Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Hall of Merit > Discussion
Hall of Merit
— A Look at Baseball's All-Time Best

Monday, January 13, 2020

2021 Hall of Merit Ballot Discussion

2021 (December 2020)—elect 3

Top 10 Returning Players
Kenny Lofton, Johan Santana, Sammy Sosa, Jeff Kent, Lance Berkman, Bobby Abreu, Buddy Bell, Wally Schang, Bobby Bonds, Sal Bando

Newly eligible players

Tim Hudson
Mark Buehrle
Torii Hunter
Dan Haren
Barry Zito
Aramis Ramirez
Shane Victorino
Alex Rios
Grady Sizemore
A.J. Burnett

DL from MN Posted: January 13, 2020 at 02:06 PM | 651 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

Page 5 of 7 pages ‹ First  < 3 4 5 6 7 > 
   401. bachslunch Posted: December 17, 2020 at 08:44 AM (#5994613)
Flip again.
   402. Mike Webber Posted: December 17, 2020 at 06:06 PM (#5994758)
I use Win Shares and BB Ref War as the first filters, with emphasis on career, where a player ranks among his era peers, with big seasons as a boosting factor.


1) JEFF KENT – 55.2 BBref-WAR, 339 Win Shares one MVP type seasons, 9 seasons 20+ Win Shares. We share the exact same birth date, so bonus points for that. 20th round draft choice with the misfortune of being in the same organization as Robbie Alomar, who was exactly the same age. Never drew more than 31 walks in a season until he was 29, which limited him to being a solid player rather than an all-star.

2) BOBBY ABREU 60.0 BWAR, 356 Win Shares. One MVP type seasons, 11 seasons 20+ Win Shares.

3) SAL BANDO – 61.6 BBref-WAR, 283 Win Shares, two MVP type seasons, 9 seasons 20+ Win Shares. I believe he was better than Ken Boyer, but his home parks helped disguise it.

4) SAMMY SOSA – 58.4 BBref-WAR, 322 Win Shares – three 30+ Win Share seasons, 7 seasons 20+ Win Shares. Value wise very similar to Bobby Bonds.

5) TOMMY LEACH – 46.8 BBref-WAR, 328 Win Shares, only one MVP type season, 8 seasons 20+ Win Shares. Good peak, excellent defensive player at third and in centerfield.

6) LANCE BERKMAN 52.1 BBref-WAR, 313 Win Shares. 4 MVP type seasons, 10 seasons 20+ Win Share season.

7) JASON GIAMBI 50.5 BWAR, 325 Win Shares. 4 MVP type seasons, 8 seasons 20+ Win Shares. This ranking really surprised me, his career value is so much greater than Chance, and he has similar big seasons, so he has to be ahead of him.

8) ANDY PETTITTE 60.7 WAR, 224 Win Shares, 2 20 win-share seasons. 2nd 1996 Cy Young ballot. Post season credit makes him the best available modern pitcher. He’s a jump up on this year’s ballot.
9) FRANK CHANCE45.6 WAR 237 Win Shares - I’m a career guy, but this is the peakiest of peak guys.
10) FRED MCGRIFF 52 WAR, 342 Win Shares. 10 seasons 20+ Win Shares. 1 MVP type season.

11) JOHAN SANTANA 51.4 BBref-WAR, 171 Win Shares – Cy Young Awards, 2 3rds, 1 5th, 1 7th. 45th in WAR among pitchers in the past 50 years. 32nd in the past 40. 26th in the past 30.

12) PHIL RIZZUTO – 40.6 BBref-WAR, 231 Win Shares, one MVP type seasons, 7 seasons 20+ Win Shares. With a conservative 60 or so win shares or 9 WAR during World War II, I move him to the top of the middle infielder group. Same arguments as Nellie Fox, only with a 3-year hole in his career at ages 25 to 27, plus a bad return to MLB in 1946. (No extra credit for 1946 – just noting it).

13)BOBBY BONDS57.7 BBref 302 Win Shares – Four 30+ Win Share seasons, at ages 23, 24, 25, and 27. After age 33 Bobby had 7 win shares, Barry had 286. Pete Browning without the fielding problems?

14) BUDDY BELL 66.1 BBref - 301 Win Shares, ZERO MVP type seasons, 7 seasons 20+ Win Shares.

15) KENNY LOFTON 68.2 BBref - 281 Win Shares – The reason I have him lower than others is I believe his Defensive WAR is overstated. Couple that with his lack of MVP type season’s and I have him lower than many others voters. 7 20-win share seasons.



Next group of guys off the ballot grouped by position:
Tim Hudson, Kevin Appier, Tommy John 289 WS, 62 WAR, 1 20WSS , Vic Willis, Mark Buehrle Gene Tenace, Jorge Posada, Wally Schang, Norm Cash, Orlando Cepeda, Johnny Evers, Larry Doyle, Bob Elliot, Ron Cey, Joe Tinker, Miguel Tejada, Luis Aparicio, Dave Bancroft, Fregosi, Stephens, Ken Williams, Bernie Williams, Bob Johnson, Sam Rice, Luis Gonzalez.

New Players not on ballot:
TIM HUDSON 56.5 WAR, 219 Win Shares, 3 20-win share seasons. 2nd 2020 Cy Young ballot.
Mark Buehrle 60 WAR, 220 Win Shares, 1 20 win-share season.
Hudson first pitcher off the ballot, Buehrle behind Appier and Tommy John.
After Hudson and Buehrle none of the other new players are close to ballot spots.

Other required notes:
Schang has so little peak it’s hard to move him into an elect position, but the best available catcher.
   403. kcgard2 Posted: December 17, 2020 at 08:22 PM (#5994789)
Bleed (#391): I personally think McNair is the 3rd best NgL candidate available, I have him #42 overall among available. I think he's probably better than some of the other NgL candidates already voted in, but information has changed over time on the NgL candidates compared to when they were voted in (and will still continue to). I think Newt Allen and Marvin Williams are the other two ahead for me, though I seem to be on an island about Marvin Williams :) Nobody else talks about him.

kwarren (#399): I don't think Willis has been overlooked, he's been getting votes for a long time. However, we've already elected 13 other starting pitchers who were exact or near contemporaries of Willis from a time when there were only 16 teams. The position from that era is already extremely well represented. Willis's stats are good on the surface comparing to modern pitchers who share the ballot with him, but they didn't stand out (comparatively) against what pitchers were doing in his own time. Electing the 14th best starting pitcher form his era would be like electing the 26th best starting pitcher from the 1990 to 2015 timeframe, so, maybe like Brad Radke? Maybe that's selling Willis a little short (I have Willis well above Radke, so yeah, it feels like it is), but it's the main reason he doesn't get more traction. Compare to him to his own contemporaries. I guess it's part of the fairness to eras argument. You'd have to say that on average there was 1 HOM level starting pitcher per team in his time to argue that he should be in, and that's going pretty deep into a position.
   404. Bleed the Freak Posted: December 17, 2020 at 10:02 PM (#5994805)
Bleed (#391): I personally think McNair is the 3rd best NgL candidate available, I have him #42 overall among available. I think he's probably better than some of the other NgL candidates already voted in, but information has changed over time on the NgL candidates compared to when they were voted in (and will still continue to). I think Newt Allen and Marvin Williams are the other two ahead for me, though I seem to be on an island about Marvin Williams :) Nobody else talks about him.


Cobbling together the latest I can see from Eric's MLEs, I have the following slotting:

Newt Allen - 109.6 - with the caveat that it's HEAVILY reliant on great fielding / small sample size on D
Hurley McNair - 108.6 - good not great sample size, but the best position player on the board.
Sam Bankhead - 101.9 - reliant on very good to excellent defense and baserunning, again some sample size issues
Lazaro Salazar - 99.5 - the difficulty of determining whether he was a position player or pitcher, could be a bit above 100 or more like 90.
Marvin Williams - 98 - I'm a fan too, but I don't see prioritizing above McNair at this point, unless Williams is being underrated for league quality he was in?
Bus Clarkson - 97.5 - your comfort of WWI credit / league quality can get him to the 100s.
Heavy Johnson - 97 - GREAT hitter, not a lot of data yet.
George Scales - 97 - someone to keep on our radar.
Ben Taylor - 95.1 - peak looks weak enough that he is no longer a priority electee.


For Vic Willis, if you add 5 years prior to and 5 years after career, from 1893-1914, he's in the discussion for 7th-11th best with Rube Waddell, Joe McGinnity, Clark Griffith, and Three Finger Brown. He doesn't seem low enough to be 13th / maybe I'm missing someone? But your point of caution stands, rather than being a top of ballot candidate, I've got him in the second half.

Interesting point on Brad Radke though, if I take the 22 year span like Willis (1990-2011), Radke does come in at the mid 20s.
   405. kwarren Posted: December 17, 2020 at 11:05 PM (#5994813)
In response to Vic Willis comments in #403 & 404

JAWS total for pitchers who pitched significant portions of their career in 1898 - 1910:

1) Cy Young 120.8 HOM
2) Kid Nichols 95.3 HOM
3) Cristy Mathewson 87.9 HOM
4) Eddie Plank 71.1 HOM
5) Ed Walsh 64.1 HOM
6) Vic Willis 56.7
7) Clark Griffith 56.1 HOM
8) Joe McGinnity 55.9 HOM
9) Rube Waddell 54.0 HOM
10) Eddie Cicotte 51.5
11) Mordecai Brown 51.1 HOM

Not sure how it was determined that Willis was the 14th best pitcher of his era. 6th seems to be much closer to the mark.

Looks like this may be one time the Hall of Fame got it right, and the Hall of Merit not so much. Willis is the only player in the top nine of his era to not be in the Hall of Merit, but he was inducted into the Hall of Fame. We can still put him in !!

Here are their actual WAR totals

1) Cy Young 163.8 HOM
2) Kid Nichols 116.3 HOM
3) Cristy Mathewson 106.0 HOM
4) Eddie Plank 91.0 HOM
5) Ed Walsh 65.8 HOM
6) Vic Willis 63.6
7) Clark Griffith 63.0 HOM
8) Eddie Cicotte 58.6
9) Mordecai Brown 58.4 HOM
10) Joe McGinnity 58.3 HOM
11) Rube Waddell 58.3 HOM

   406. Bleed the Freak Posted: December 18, 2020 at 11:33 AM (#5994864)
3) Mark Buehrle - debuts in an elect-me spot PHoM 2021.
15) Tim Hudson - debuts on-ballot, PHoM 2021


DL and others, what drives Buehrle to land ahead compared with Hudson?
   407. Bleed the Freak Posted: December 18, 2020 at 11:48 AM (#5994868)
15. Mickey Welch. Next best pitcher WAR after John by a decent margin. He and McCormick are the only 19th century pitchers I currently support.
16-40. Urban Shocker,


Bachslunch, any chance I can get you to flip Shocker onto your ballot?
Are you giving Urban a bump for WWI time?

As to Welch, does he standout from Tommy Bond, Tony Mullane, and Charlie Buffinton from his era, and how many short-distance pitchers should we prioritize?

Looking close at Shocker, I'm seeing The Big Train, Pete Alexander and Lefty Grove as clearly better than him, I believe Dazzy Vance a little ahead,
he's in a scrum for 5th best pitcher ~1911-1932, likely with Stan Coveleski going 5/6.

HOPE everyone takes a close look at Urban this cycle : )



   408. DL from MN Posted: December 18, 2020 at 11:49 AM (#5994869)
"DL and others, what drives Buehrle to land ahead compared with Hudson?"

1 pitching win. 3-16 are actually really close in my spreadsheet.
   409. bachslunch Posted: December 18, 2020 at 04:20 PM (#5994955)
Are you giving Urban a bump for WWI time?

"Fine with giving Negro League credit, not presently giving credit or debit for war, injury, illness, postseason play, or minor league service." I've had people give me a ton and a half of grief about this before, but I don't do so. I don't give extra credit to players like Phil Rizzuto and don't debit players like Indian Bob Johnson. And if I'm forced to do so, I'll withdraw my ballot.

I struggled a bit with Shocker (58.7 BBRefWAR) vs. Welch (62.3 BBRefWAR) on this ballot, and there's no question they'll both be on my ballot next time. The fact that Welch has almost 5 points of WAR on Shocker *and* pitched his whole career in the NL put him ahead. I'd have docked him if he had pitched in any other league during his career, but he didn't.

As to Welch, does he standout from Tommy Bond, Tony Mullane, and Charlie Buffinton from his era, and how many short-distance pitchers should we prioritize?

"While it's not a popular approach, am currently treating 19th century pitchers pretty much equally as post-1900 if they played most or all of their career in the NL; for now am assuming AA, NA, and UA stats are suspect."

Welch pitched his whole career in the NL (62.3 BBRefWAR). Tommy Bond got 9.8 points of WAR from the NA and UA, and subtracting that from his BBRefWAR of 60.9 puts him well back at 51.1 WAR. Tony Mullane pitched most of his useful career in the AA, and his total BBRefWAR for the NL years is 19.0 WAR. Charlie Buffinton got 10.2 points of WAR from a year each in the PL and AA, and subtracting that from his BBRefWAR of 60.7 leaves him at 50.5 WAR. That's a significant difference in Welch's favor. Maybe it's not nuanced and maybe it's not the way others would do it, but so be it. I also was the person who left Mariano Rivera off ballot entirely because relief pitcher WAR numbers are always low compared to starters -- it bothered me a lot that other voters went through what I considered to be a lot of twisting and turning to justify Rivera, but didn't feel it was my place to say anything (that's their choice, and I accepted it). I also would not likely have carried Sandy Koufax on my ballot, and do not carry folks like Johan Santana, Dizzy Dean, or Addie Joss on it, either. Having Santana at 31 is a concession, quite frankly, and I'm not comfortable with it -- but he's probably getting in this time, so I won't have to worry about him after this year. Yes, I value longevity over peak. Tommy John is on my ballot, for example.

I've struggled a lot with Tommy Bridges because his WAR number is not as high, but gave him a boost because of his high ERA+. I'm not so sure I can do that now with Andy Pettitte, Tim Hudson, and Mark Buehrle eligible. Will likely be rethinking that next time.

The only two 19th century pitchers I plan to support going forward are Jim McCormick (who ends up with 68.4 BBRefWAR when you subtract his 7.5 UA WAR) and Mickey Welch. FWIW, there were three other voters who had Welch on their ballot last time. I've been McCormick's sole backer from day one.
   410. Bleed the Freak Posted: December 18, 2020 at 09:32 PM (#5994980)
11) Bob Johnson - 147 wWAR, Solid hitter, bad defender, short career perhaps worthy of minor league credit, which I use to balance out era representation that goes against him. pHOM 1960.


kgard2 great explanations and or ballot, one tiny beef, Bob Johnson appears to have been above average if not a very good fielder.
TZ/DRA/Kiko all have him in the black, he's really a standout comparatively with Kiko's WAR, which should be capturing arm related value, something TZ/DRA isn't for that time era. With Johnson 45th in OF assists all-time, his defense could be understated by TZ/DRA.

Also, you mentioned Marvin Williams as your favorite Negro League candidate, though not in your top 20...
where does he land overall, same goes for Newt Allen and Hurley McNair, thank you sir!
   411. Bleed the Freak Posted: December 18, 2020 at 09:36 PM (#5994981)
"While it's not a popular approach, am currently treating 19th century pitchers pretty much equally as post-1900 if they played most or all of their career in the NL; for now am assuming AA, NA, and UA stats are suspect."

Welch pitched his whole career in the NL (62.3 BBRefWAR). Tommy Bond got 9.8 points of WAR from the NA and UA, and subtracting that from his BBRefWAR of 60.9 puts him well back at 51.1 WAR. Tony Mullane pitched most of his useful career in the AA, and his total BBRefWAR for the NL years is 19.0 WAR. Charlie Buffinton got 10.2 points of WAR from a year each in the PL and AA, and subtracting that from his BBRefWAR of 60.7 leaves him at 50.5 WAR. That's a significant difference in Welch's favor.


Fair enough on preferring guys with more career value, I would caution you though from dismissing other non-NL stats all-together from that time era. From our discussions here in the past, the PL might actually have been the best league in 1890, the AA was of pretty comparable quality for a lot of it's run. I don't trust the UA either, it had two stars and not much else.
   412. kwarren Posted: December 18, 2020 at 09:55 PM (#5994982)
I also was the person who left Mariano Rivera off ballot entirely because relief pitcher WAR numbers are always low compared to starters -- it bothered me a lot that other voters went through what I considered to be a lot of twisting and turning to justify Rivera, but didn't feel it was my place to say anything (that's their choice, and I accepted it).


Totally agree with you on this issue. I am at a loss to understand the overwhelming support he received from both Hall of Fame & Hall of Merit voters. His on field value is about equal to David Wells, but the perception of these two players couldn't be further apart. Save total & playoff stats are a big part of it. And most people look at things such as his ERA and ERA+ as if he was a starter, not what his role actually was. You would think that they would use a separate ERA+ for starters and relievers to put the stat into a little context, but they don't.

Rivera was a great reliever, clearly the best ever, but that doesn't make him nearly as valuable or meritorious as the top 15 or 20 starting pitchers in the major leagues during his career.

Relief pitchers are essentially back-up pitchers. WAR and Win shares recognize this, even with the boost given for late inning leverage. But almost everybody chooses to ignore measures of value when it comes to relief pitchers. They want to vote them in, and look for ways to justify it. As you said, going through a lot of twisting and turning. Comparing relievers to other relievers seems to be way they do it, as if relief pitcher is a position on it's own, and needs to be represented in the Hall of Merit. Apparently "saving" games that your team had a 95% chance of winning anyway with an average reliever is a big deal. Even with a 3-run lead in the 9th inning.

Just to show how much saves influences the electorate consider the following relationship between JAWS & Saves:

Bobby Shantz 29.8 48
Tom Gordon 29.1 158
Firpo Marberry 28.4 99
John Hiller 28.4 125
Greg Swindell 28.0 7
Ellis Kinder 26.7 104
Turk Farrell 26.2 83
Kerry Wood 26.2 63
Lindy McDaniel 25.8 174
Stu Miller 25.2 153
Syl Johnson 24.9 43
Lee Smith 24.8 478 HOF
Bruce Sutter 24.2 300 HOF
Trevor Hoffman 23.7 601 HOF

Pretty clear, that it is saves, and not value or merit that gets relievers elected to the Hall of Fame. And if it wasn't for the save statistic we would still be waiting for our first reliever to be elected. Really sad, that this is where we are, when you think about what a meaningless stat, saves really are. And relief pitching is the last thing general managers are concerned about when building a team.

The Hall of Fame makes some sense because, as the name suggests, it is more oriented to fame as opposed to merit. Sandy Koufax, Jack Morris, Whitey Ford, Jim Rice, Kirby Puckett, Hoyt Wilhelm, Rich Gossage, Trevor Hoffman, & Lee Smith can also be viewed that way. Their fame was far more notable than their value or merit.
   413. DL from MN Posted: December 18, 2020 at 10:07 PM (#5994983)
His on field value is about equal to David Wells


That's an inaccurate exaggeration.

Pitcher PWAA
Rivera 32.5
Wells 22.0

That doesn't include any postseason credit.
   414. Howie Menckel Posted: December 18, 2020 at 10:08 PM (#5994984)
if not for Rivera's incredible postseason numbers, he would not have been on my ballot, either.

it's pretty well forgotten now, but Rivera threw 141 IP in his 96 postseason games. this was NOT a three-out robot with no "fireman" duties, by any means. Saves were not a coveted category in NYY MGR Joe Torre's playing prime, and he did not tie himself into knots avoiding having Mariano go the extra mile whenevwe necessary.

he allowed 13 R (11 ER) in the postseason. he did have a couple of crucial hiccups - but far more successes - in those 96 games.

as a setup man in 1996 WS win, he allowed 1 R in 14 IP over 8 postseason games.
in 1998 title, 13.3 scoreless in 10 G.
in 1999, 12.3 scoreless in 8 G.
in 2000, 3 R in 15.6 IP.

dinged for 1997 and dinged for 2001, sure. but however you measure leverage, he had a LOT of it.
   415. Eric J can SABER all he wants to Posted: December 18, 2020 at 10:28 PM (#5994986)
as a setup man in 1996 WS win, he allowed 1 R in 14 IP over 8 postseason games.
in 1998 title, 13.3 scoreless in 10 G.
in 1999, 12.3 scoreless in 8 G.
in 2000, 3 R in 15.6 IP.


Plus 1 R in 16 IP in 2009.
   416. Howie Menckel Posted: December 18, 2020 at 10:40 PM (#5994988)
good point!
   417. kwarren Posted: December 18, 2020 at 11:16 PM (#5994990)
That's an inaccurate exaggeration.

Pitcher PWAA
Rivera 32.5
Wells 22.0


Pitcher JAWS
Rivera 42.5
Wells 42.4

Hard to be any closer than that.

if not for Rivera's incredible postseason numbers, he would not have been on my ballot, either.


If you agree that based on his regular season performance he shouldn't be on the ballot, we basically agree.

I had him on my ballot also, but nowhere near #1. I guess it depends how much WAR or JAWS his postseason accomplishments are worth.
   418. Rob_Wood Posted: December 19, 2020 at 01:09 AM (#5994993)
There are many stats that people use to compare pitchers. JAWS is one of many. WAA and WAR are available (both BB-Ref and FIP-based). Various pennant added stats are out there which "combine/synthesize" the WAA and WAR values. I have my own CPASR (Career Pennant Added using a Sliding Replacement level).

By CPASR (regular season only), Mariano Rivera has .652. David Wells is ... (drum roll) ... also at .652. These are around 380th best all-time. For what it is worth the 10 players on either side of Rivera and Wells in the CPASR rankings are Nellie Fox, Wally Berger, Orlando Cepeda, Bert Campaneris, Chuck Klein, Vern Stephens, Vada Pinson, Sam Rice, Larry Doyle, and Matt Williams.

In full disclosure I had Rivera 4th on my 2019 HOM ballot due largely to his stellar post-season performance.
   419. kwarren Posted: December 19, 2020 at 06:52 AM (#5995003)
12. Jim Sundberg - I firmly believe that catcher defense is underrated by our current metrics, and Sundberg looks to be the strongest candidate of the eligible catchers when accounting for that deficiency. There’s a non-zero chance I’m totally off base here, but I think it’s pretty likely that his unaccounted defense pushes him at least to borderline territory.


Would love to hear more about how he is underrated by "current metrics. Why is this the case ? Why is it just him ? How should the metrics be revised ?

Sundberg's dWAR is 25.3. This puts him ahead of all catchers except Pudge Rodriguez (29.6) and Gary Carter (26.1). I find it hard to understand that his defense is so underrated that he should be ahead of Munson, Tenace, Freehan, Schang, Posada, Kendall, & Porter.

Here are the JAWS scores the these catchers with their dWAR also noted.

Therman Munson 41.5 11.9
Gene Tenace 40.9 1.8
Bill Freehan 39.2 12.0
Wally Schang 37.8 6.5
Jorge Posada 37.7 2.6
Jason Kendall 36.0 13.9
Darrell Porter 35.0 10.6
Jim Sundberg 34.6 25.3

Sundberg's WAR7 is 28.7 which is the 2nd worse of baseball's top 26 catchers, ahead of only Schang.
   420. DL from MN Posted: December 19, 2020 at 08:26 AM (#5995004)
A player's JAWS is calculated by averaging his career WAR with the total WAR from his seven-year peak -- that is, his seven most valuable seasons (consecutive or non-consecutive).


Yeah, that's not how I evaluate players for this project at all. I don't care at all about most valuable 7. I'm much more interested in how much value the player provided above the average player at their position over their career. Spitting WAR7 back at me isn't going to convince me of anything.

The best argument that we are missing something in catcher defense is the standard deviation of catcher defense is lower than the standard deviation of left fielder defense despite the fact that the catcher is involved in SO MUCH MORE of the defensive value of the ballclub. There is a strong possibility that our accounting system is wrong and we're attributing catcher defensive value to other positions, mostly pitchers. People argued against this for years; then came the framing research. If we artificially double the stdev for catcher defense the best ones (Sundberg, Rodriguez, Carter) become a lot more valuable and the worst (Posada, Tenace) become much less so.

Also dWAR measures defensive contribution (RField) and the replacement value of the position itself (RPos). RField is a better way to compare value among defenders at a particular position.

Since you do use WAR7, if you double Sundberg's defensive contribution he adds 8 wins to that WAR7 number (36.7). Posada would lose at least 3 wins (34.7).
   421. progrockfan Posted: December 19, 2020 at 09:04 AM (#5995005)
The best argument that we are missing something in catcher defense is the standard deviation of catcher defense is lower than the standard deviation of left fielder defense despite the fact that the catcher is involved in SO MUCH MORE of the defensive value of the ballclub. There is a strong possibility that our accounting system is wrong and we're attributing catcher defensive value to other positions, mostly pitchers.
I agree with this 100%.

As for relievers:

Can't speak for anyone else here, but personally I don't give a damn about saves. Leverage, though, is something different. Rivera's 0.70 ERA and 0.759 WHIP in 98 post-season innings grabs my attention irespective of "starter" or "reliver" status.

Same with Gossage. He had an ERA at least one full point under leaage averge in 11 different seasons, pitched a lot of innings vis-a-vis modern relievers, and was deadly in the '78 and '81 Series. Let's pretend for a moment that the 'save' stat doesn't exist (which would be OK with me). All pitching staffs carry multiple pitchers other than their starters, and were I a manager, I assure you that I would find regular employment in high-leverage situations for a pitcher like Gossage.

But I'm not wholly in the relief camp either. There are exactly three (3) relievers in the history of baseball who I consider worthy of HoM status - Rivera, Gossage and Wilhelm, that's it - as against dozens & dozens of starters. That's not exactly reliever-worship. Guys like Hoffman, Smith, Quisenberry, Sutter, Fingers &c. strike me as very over-rated (that's just me, ymmv). Occasionally there's a crossover guy like Marberry, Smoltz or Eckersley, and I guess I can see a case being made for them, but I'll take a dominant long-career starter over any of them.

Koufax, for me, is a difficult one. In terms of raw dominance at any given moment in time, he stands with Grove and Paige at the very summit. On the other hand, a goodly chunk of that dominance is ascribable directly to park effects. But then, the Dodgers did win & win & win when he was around... In the final analysis, I'd take Paige or Grove over Koufax in a heartbeat, and Koufax over Quiz & Co. in a heartbeat. As for Rivera vs. Koufax - a long career of short appearances vs. a short career of long appearances - I'd probably take Rivera, but only because a HoM-worthy reliever is a much rarer commodity than a HoM-worthy starter.

It's clear, to my mind, that a great starter is worth far more than a great reliever. That doesn't mean, though, that there are no relievers worthy of Merit - any more than the fact that first basemen generally hit far better than catchers means that there are no catchers worthy of Merit.

I have yet to see an uber-stat which takes every meaningful factor into account. As Bill James notes, good sabermetric analysis reaches outside statistics and considers all available evidence. Baseball is a complex and nuanced game, and I think our discussions, analyses, and voting should reflect this fact.
   422. Jaack Posted: December 19, 2020 at 12:03 PM (#5995031)
Would love to hear more about how he is underrated by "current metrics. Why is this the case ? Why is it just him ? How should the metrics be revised ?

Sundberg's dWAR is 25.3. This puts him ahead of all catchers except Pudge Rodriguez (29.6) and Gary Carter (26.1). I find it hard to understand that his defense is so underrated that he should be ahead of Munson, Tenace, Freehan, Schang, Posada, Kendall, & Porter.

Here are the JAWS scores the these catchers with their dWAR also noted.

Therman Munson 41.5 11.9
Gene Tenace 40.9 1.8
Bill Freehan 39.2 12.0
Wally Schang 37.8 6.5
Jorge Posada 37.7 2.6
Jason Kendall 36.0 13.9
Darrell Porter 35.0 10.6
Jim Sundberg 34.6 25.3

Sundberg's WAR7 is 28.7 which is the 2nd worse of baseball's top 26 catchers, ahead of only Schang.


bWAR does not incorporate pitch framing or game calling into it's metrics at all. Fangraphs does incorporate them for seasons since 2008, and it has a huge impact - 123.5 runs for Buster Posey, 147.4 for Yadier Molina, 165.6 for Brian McCann, and 165.8 for Russell Martin. And all those except Posey are missing a few seasons too. On the other hand, it has a pretty big negative impact on poor fielders like Jorge Posada and AJ Pierzynski. Now some people are hesitent to trust the magnitude of these numbers, but throughout a number of different attempts at developing a method for evaulating pitch framing, the results have consistently put the same guys at top with similar levels. I am a believer, but I do regress the raw framing data we have to some degree.

In order to avoid short-changing catchers who play before the PITCHf/x data, I use an admitedly haphazard method based on their existing catcher defense stats and their playing time at catcher to try and guesstimate the territory that their framing would put them in (generally, catcher who are good at framing are good at other things).

In practical terms, guys with good defense and longevity at the position move up, and guys with shorter careers, poor defense, or limited playing time at catcher move down relative to the rest. Sundberg is the biggest winner because he's on the fringes of my consideration set already and has very strong fielding numbers. Most other guys with as strong of fielding numbers are already elected or too far out to make it into ballot territory.
   423. DL from MN Posted: December 19, 2020 at 01:29 PM (#5995044)
Jason Kendall is interesting to me since he was frequently praised for his ability game calling (to the point where he was calling games from LF) and getting strikes called by the umpire. He kept catching even after he tore his thumb ligament, dislocated his ankle, etc.
   424. kcgard2 Posted: December 19, 2020 at 08:45 PM (#5995096)
Bleed (#410): Yes, in my head I remembered Johnson as having a reputation of poor defender, and I seem to have been confused. I should have checked. Seems Johnson was roughly an average or above average corner outfielder defensively, which overall makes him defensively not a valuable player, but I had a misconception of his rep and phrased it poorly. I didn't miscalculate his WAR stats if you were worried about that :)

Bleed (#404): Also working from chaleeko's MLE's, I have the following slotting:

Newt Allen 4.6 WAR/650, 9800 PAs - you mentioned everything of importance here, huge caveats about defense currently, but if they prove out I would move Allen on-ballot. I have Allen #35 overall, BTW
Marvin Williams 4.7 WAR/650, 8400 PAs - I like him better than McNair because 2B vs corner OF basically
Hurley McNair 4.5 WAR/650, 10200 PAs - McNair also has a lower completeness score, and that's meaningful to me because of large corrections that have come via more complete records, there's plenty of examples (Taylor is prominent)
Sam Bankhead 4.1 WAR/650, 10700 PAs - you mentioned the issues here
Conrado Marrero 4.4 WAR/250, 3400 IP - completeness score lower than McNair, I am also inherently more skeptical of NgL pitchers
George Scales 4.2 WAR/650, 9600 PAs
Heavy Johnson 5.4 WAR/650, 6700 PAs - low completeness score and low projected PA total, defense questions
Ben Taylor 3.9 WAR/650, 10100 PAs - highly complete records, low impact player. Best comp to me is Mark Grace. If MLEs had started this way, he would not have made it onto people's ballots, but inertia is causing him to stay there anyway.
Lazaro Salazar 4.6 WAR/650, 8500 PAs - I like Salazar, but it's an odd case with the split between position player and pitcher. Neither of which dominant but both highly effective is just not a profile you see in MLB except for Babe Ruth, and Salazar wasn't Babe Ruth (I feel moderately comfortable with that assertion)
Roy Welmaker 5.0 WAR/250, 2900 IP - pitcher skepticism, WAA lags other players from the MLEs
Bill Holland 4.2 WAR/250, 3500 IP
Webster McDonald 4.0 WAR/250, 3900 IP
Dave Barnhill 4.5 WAR/250, 3300 IP - lower completeness, lower WAA

Regarding Vic Willis, I used 1890-1920 as his "cohort," so not quite 10 years each way. Since Willis already had a short career, it seems fair. 30 years (31 I guess technically) is kind of my standard window for "contemporaries." Quibble if you like. To kwarren, that gives us Young, Johnson, Mathewson, Nichols, Plank, Alexander, Waddell, Brown, Walsh, Griffith, Rusie, McGinnity (with an outside consideration of Rixey, so really it should have been 12 perhaps, my mistake) as the contemporaries who put up a huge chunk of their HOM value in this timeframe, the 13 that I referenced before. We will not all agree on the ordering of these 13 vis-a-vis Willis, but I used it as a shorthand since those 13 are in the HOM. By my ballot, you can tell I also prefer Cicotte to Willis, but neither here nor there. Taking the HOM voters as the yardstick, Willis would be 14th best from his time, as I've defined it here. Now you can trace the genesis of my comment. Given that we already have 13 (or 12) from his time, the argument remains how many is an appropriate amount for one position from one timeframe, to explain why Willis hasn't gotten more love. The voters preferred those other candidates.

Bleed (#406):
DL and others, what drives Buehrle to land ahead compared with Hudson?

Higher bWAR and fWAR, same WAA (in negligibly more innings), higher wWAR, less disparity between bWAR and fWAR estimates of value (I like this as a bit of a tiebreaker sometimes), small narrative credit (perfect games, extreme prevention of baserunning), and Buehrle was still an effective pitcher when he retired while Hudson was toast. At the end of the day, there's not much daylight between spot #30 and spot #60 on the ballot, honestly, which is roughly my rankings of those two.
   425. kcgard2 Posted: December 19, 2020 at 08:56 PM (#5995102)
Regarding Mariano, I split the difference on him (I believe I had him 11th on my ballot). His dominance was literally unparalleled and the postseason performance is crazy. On top of that, his raw WAR total was not out of line for HOM consideration even as a reliever, which I wonder whether will ever happen again. wWAR and other things that weight peak WAR value aren't going to paint him that favorably, thus my splitting the difference. And I think I'm OK with it, though like others, I would have 3 relievers in my HOM at most, unless you count Eckersley as a reliever.
   426. kcgard2 Posted: December 19, 2020 at 09:03 PM (#5995105)
Sammy Sosa. The corked bat. The multi-year clubhouse disruption. The abandonment of his team on the final day of the season. His persona non grata status with the team for which he won his MVP

progrockfan...I'm not sure these are allowed reasons to dock Sosa on your ballot? I doubt cancelling them out will make you move him high enough anyway, but the second sentence of the requirements on the ballot page is "You do need to stick to what happened on the field (or your best estimate of what would have happened if wars and strikes and such hadn’t gotten in the way)." Later it explains that things like this can be used for a first-year boycott.
   427. Bleed the Freak Posted: December 20, 2020 at 08:16 AM (#5995190)
kcgard424, thanks on your depth of explanations.

I figured Bob Johnson was just verbiage to clean-up : )

For Welmaker and in general, do you award some level of war credit to pitchers, or are you on the side of and arm only has so many bullets to spare?

I hope Kiko can drop in on Hudson vs Buehrle, as Hudson's an easy HOF pick with his W-L records while Buehrle is off the map.
   428. kcgard2 Posted: December 20, 2020 at 08:45 AM (#5995191)
For most of these MLEs, the guys are already projected for 3000 IP or well over. I'm taking those as fair numbers. As a small effort towards answering a bit for Kiko's system, (assume Hudson and Buerhle are equal) pWORL will like Hudson better than Buehrle just because Hudson pitched on better teams than Buerhle did, I think.
   429. progrockfan Posted: December 20, 2020 at 08:52 AM (#5995192)
progrockfan...I'm not sure these are allowed reasons to dock Sosa on your ballot? I doubt cancelling them out will make you move him high enough anyway, but the second sentence of the requirements on the ballot page is "You do need to stick to what happened on the field (or your best estimate of what would have happened if wars and strikes and such hadn’t gotten in the way)." Later it explains that things like this can be used for a first-year boycott.
I'm willing to have a debate about this; it's a potentially valid point. I'm not saying you're wrong, not at all. But surely, year after year of clubhouse disruption, walking out on his team before the season is over, etc., DOES affect the quality of play on the field...? What does the electorate think?
   430. Bleed the Freak Posted: December 20, 2020 at 08:53 AM (#5995193)
...it's not a team thing though.

Hudson - 64.9 team, 65.2 neutral
Buehrle - 49.6 team, 46.7 neutral
   431. progrockfan Posted: December 20, 2020 at 08:56 AM (#5995194)
Regarding Mariano...his raw WAR total was not out of line for HOM consideration even as a reliever, which I wonder whether will ever happen again.
Yeah, I don't see the next potential Mariano on the horizon either. He was, I suspect, a 100% unique outlier.
   432. kcgard2 Posted: December 20, 2020 at 11:23 AM (#5995213)
I don't recall years of clubhouse disruption, but I think the difficulty of docking for it would be quantifying how much effect you think it had on his teammates' performances, if you want to dock Sosa for it. Otherwise clubhouse disruptions are not something that happened on the field, and it seems to amount to a revival of the character clause. Did you check whether being the center of that magical home run chase inspired his teammates as much as the disruptions harmed it? As for walking out on his team the final day, totally fair to dock since it amounted to one game missed on the field. It sounds like I'm being facetious, because if you did quantify these things somehow, I think they'd amount to a drop in the bucket, but heck, maybe they'd amount to more. The quality of Sosa's play on the field was what it was even with these distractions, at the end of the day.

I guess I'd like to hear other viewpoints also. Corked bat is the interesting one to me because that was on-field and rule-breaking, and presumably someone could argue it impacted his performance, I guess (best of luck proving anything, but the argument is there to be made somehow, within the parameters of things voters can consider).
   433. bachslunch Posted: December 20, 2020 at 04:26 PM (#5995242)
If what I've read is correct, corking a bat gives the hitter no advantage. To me, it's a clear "cheating" issue as it's considered illegal, but if folks like Shoeless Joe and Rose are in for far worse things, bat corking is very, very small beer. Surely smaller than illegal pitches, which HoM-ers Ford, Drysdale, Sutton, Perry, and Maddux all indulged in (reportedly, anyway), and do impart some advantage.
   434. kcgard2 Posted: December 20, 2020 at 04:32 PM (#5995244)
I agree. After posting that I actually began to wonder whether even cheating is allowed to be held against players, considering illegal pitches and steroids not being allowed to keep guys out of HOM.
   435. Kiko Sakata Posted: December 20, 2020 at 05:10 PM (#5995248)
I hope Kiko can drop in on Hudson vs Buehrle, as Hudson's an easy HOF pick with his W-L records while Buehrle is off the map.


To be honest, I'm not 100% sure why Hudson looks so much better than Buerhle. Here's what I wrote on their respective discussion pages back in January.

Buerhle:

Using my preferred weights from the link in the previous paragraph (one-year, treat SP and RP as different positions w/ positional averages based on pitchers who did both in the same season), I get Buerhle at 25.8 eWOPA and 28.6 pWOPA.

Baseball-Reference has him at 28.4 WAA (29.4 pitching, -1.0 batting). Using BB-Ref's WAR-WAA gap and applying it to Fangraphs' WAR, Buehrle has 20.8 FIP-WAA and 29.9 RA9-WAA.

So, we all seem to be in the same general ballpark here. Buerhle had a great pickoff move, threw very few wild pitches, and fielded his position very well - all of which Player won-lost records acknowledge, and all of which would lead Buerhle to out-perform his FIP. And, indeed, all of his BB-Ref WAA, his eWOPA, and his pWOPA end up better than his FIP-WAA but not quite as good as his RA9-WAA.

Beyond the numbers, my view of Buerhle is basically that he's a career candidate who didn't have enough of a career - Tommy John if he'd retired a decade earlier. And while I'm one of Tommy John's biggest fans here (maybe his biggest), he definitely needs that last decade of his career to be a serious HOM candidate.

Hudson:

Player won-lost records are a big fan of Tim Hudson. Using my default weights, he's at 40.4 eWOPA (eWins over positional average) and 40.8 pWOPA (pWins ...).

Baseball-Reference only gives Hudson 31.3 WAA (30.0 pitching, 1.3 batting). Fangraphs doesn't show WAA, but since they use the same replacement level as BB-Ref, you should be able to just subtract the difference between BB-Ref's WAR and WAA from Fangraphs' WAR to get fWAA.

Fangraphs calculates their WAR two ways: using FIP and using RA9. Converting from WAR to WAA, they give Hudson 23.6 FIP-WAA and 37.7 RA9-WAA.

So, Tim Hudson's runs allowed (he had a career ERA+, for example, of 120) is way better than you'd expect from his peripherals (career ERA of 3.49, career FIP of 3.78). BB-Ref's WAA number comes in between the two Fangraphs numbers - as you'd expect. BB-Ref thinks that maybe 50-60% of Hudson's RA-FIP gap is to his credit, which certainly seems plausible on its face. Whereas, if we accept Fangraphs' numbers, Player won-lost records say Hudson actually deserves MORE credit than even his excellent RA9 would suggest. Which certainly raises an eyebrow, even from me.

Here's Tim Hudson's pitching page on my website. Hudson rates as above average at everything (he was also an above-average fielder and an above-average hitter for a pitcher). Within pitching, Hudson was very good against the running game (Component 1), good at preventing hits on balls in play (Component 5), good at limiting extra-base hits (Component 6), good at generating double plays in double-play situations (Component 7).

Component 3 is plate appearances where the batter doesn't put the ball in play - strikeouts, walks; Component 4 is plate appearances where the batter does hit the ball - including home runs. These are jointly estimated, so they're jointly .500 but not both .500. For Hudson's career, pitchers' Component 3 win percentage was over .500 (in aggregate, the wins from strikeouts were more than the losses from walks/HBP); Component 4 was below .500 (in aggregate, the wins from pop-ups, fly balls, and ground balls were less than the losses from line drives and home runs). Tim Hudson was above average - and above .500 - in both. Basically, beyond FIP, Tim Hudson was really good at avoiding line drives and generating relatively weak contact.

If I take these numbers at face value, Hudson's easily on my ballot, almost certainly in an elect-me spot, and might even be #1 overall. I have to think about whether I might want to temper them a little bit, though - and, if so, by exactly how much.

[end of my year-old quotes]

I don't know that I have anything more insightful to add to that. And I completely understand if you want to temper my system's love of Hudson in constructing your ballot. I will re-iterate what Bleed said in 430 - it's not a pWins vs. eWins thing. To the extent Hudson pitched for better teams than Buerhle, the strength of those teams tended to be their pitching - i.e., Hudson.

I guess the one thing I'd note from the above is that Hudson shows up as above average at every breakdown I make - good control, good at avoiding home runs and line drives, good at getting outs on balls-in-play, good at avoiding extra-base hits, good at controlling the running game, good at fielding his position, good hitter (for a pitcher), etc. And one thing that I found from my system is that being a little bit above average at everything is a little better than being really good at some things but having glaring holes. I've mostly noted this is true of teams - and ironically, sort of my go-to example of this is the 2005 White Sox, who basically had no holes, just a bunch of guys who were somewhere between average and pretty good - including, of course, Mark Buehrle.
   436. Kiko Sakata Posted: December 20, 2020 at 05:26 PM (#5995250)
Actually, I know we all disdain pitcher wins as a meaningful statistic, but it turns out, my numbers track quite well with Buerhle's and Hudson's win-lost records.

Buerhle had a career record of 214-160, which is 27 wins over .500. And, per my last comment, Fangraphs, BB-Ref, and I all have Buerhle somewhere between 21 and 30 wins above average. So, we all pretty much agree that Buehrle's career record is an accurate reflection of his ability. Which, having lived in Chicago for his entire career, seems reasonable: the White Sox were probably around .500 for his career, so they weren't costing him a lot of wins but he was pretty much earning the ones he got.

Hudson had a career record of 222-133, which is 44.5 wins over .500. BB-Ref and Fangraphs have him at more like 24 - 38 wins over average (note: there's a pretty huge gap between Fangraphs' number based on RA-9 vs. based on FIP). So, they both think that Hudson's record is a little too good relative to his performance. And, at first glance, this doesn't seem unreasonable (see comment #428). I give Hudson a pWOPA of 40.8 (pWins over positional average) - which is still a tick below his traditional record but close to it. But then I get the really interesting result that he ALSO looks just as good when I control for context - 40.4 eWOPA. Which I find kind of fascinating - my Player won-lost records seem to be saying that there's some signal in Hudson's won-lost record that's not entirely reflected in his peripheral stats.
   437. kcgard2 Posted: December 20, 2020 at 07:03 PM (#5995263)
Within pitching, Hudson was very good against the running game (Component 1), good at preventing hits on balls in play (Component 5), good at limiting extra-base hits (Component 6), good at generating double plays in double-play situations (Component 7)...Beyond FIP, Tim Hudson was really good at avoiding line drives and generating relatively weak contact.

Do you control for quality of defense when measuring these things? They all sound like things that very well could be the product of playing in front of good defenses, which would explain why bWAR/WAA is not giving Hudson credit for them. I'm trying to get at how we can explain the disparity between your player win-loss records and WAR.

It's hard for me to follow an argument that says Hudson might be the #1 overall player on a ballot, and Buehrle is a compiler with not enough career length, when they have essentially the same WAR, same WAA, and same career length. Hudson's best seasons were slightly better than Buerhle's best seasons.
   438. Kiko Sakata Posted: December 20, 2020 at 08:20 PM (#5995275)
Do you control for quality of defense when measuring these things? They all sound like things that very well could be the product of playing in front of good defenses, which would explain why bWAR/WAA is not giving Hudson credit for them. I'm trying to get at how we can explain the disparity between your player win-loss records and WAR.


eWins try to control for teammates: pWins just split credit between pitchers and fielders but give both sides the same credit/debit depending on the play; eWins go back at the season level and try to control for the quality of teammates. I can't swear that it does it correctly - the adjustments tend to be a little more modest, I think, than BB-Ref's adjustment for team fielding, for example. Hudson's gaining about 0.14 wins for "teammate adjustments" in pWins relative to eWins.

Without digging through all 489 games of Tim Hudson's (regular-season) career, my best guess is that the game-level breakdown of Tim Hudson's performance was more favorable than you'd expect given his season- and career-level statistics, which translates into a few more eWins/pWins/traditional pitcher wins than you'd expect. To be perfectly honest, I'm not entirely sure what that would look like in a game log, but I think it would show up if, for example, Hudson had a tendency to have one or two truly atrocious starts every year that tended to blow up his ERA but only result in one or two losses.

Okay, take 2007, for example. I have that as Hudson's best season by eWOPA/eWORL (5.5/7.1) whereas BB-Ref sees that season as a more modest 2.9 WAA / 4.7 WAR. The Braves went 22-12 in Hudson's starts in 2007. He had three games that season where he failed to pitch 5 innings and the Braves won 2 of them. He also had two games where he pitched exactly 5 innings and gave up 5 runs; the Braves split those two games. That seems a bit lucky for Hudson but I don't know. Honestly, I don't know how to judge Hudson's 2007 game-level performance because I really don't have a great sense of what one would expect - 5 crappy games out of 34 doesn't jump out at me as unusual and none of the 5 were especially terrible - he allowed 27 runs in 19.2 IP in those 5 games. Is that unusual for a good pitcher?
   439. progrockfan Posted: December 21, 2020 at 09:55 AM (#5995302)
OK, accepting for the sake of argument the idea that I can't penalize Sosa for clubhouse antics / the corked bat / etc., where does that leave me?

My ballot is already OF-heavy, as 4 of 8 players I regard as electable are OFs.

There's no way Sosa passes Duffy on my ballot. I regard Duffy as a legitimate all-time great. (I wonder if his .440 average is being subconsciously held against him in some quarters? It's kind of a 'freak stat,' and may make him seem a bit of a freak candidate. But I ask you: What other candidate was 1) clearly the best defensive player at his position - three positions, actually - for a full decade; 2) was a very good-to-excellent hitter throughout his career; and 3) was the postseason masher of his time, but hasn't been elected? There may well be a name out there, but I personally can't think of anyone.)

Bonds had power - not as much as Sosa, obviously, but very solid- and also had superior defense, speed, and on-base ability. That's a big hurdle for Sosa to covercome, and I don't think he clears the bar. I always opt for the all-around player over the one-dimensional talent, and Bonds was, in my view, much the superior multi-tool player.

Johnson was far more consistent than Sosa, was an above-average hitter literally every year he played, and deserves significant MLEs (in my view) both before and after his MLB career. The case for Sosa vs. Johnson is I think more plausible than for either Duffy or Bonds Sr., but I still come out in favor of Johnson.

And then there are further OFs - Kenny Lofton in particular - whose cases I also regard as superior to Sosa's.

Which means, in sum, that Sosa still doesn't make my top 15. I can re-word my reasons why, and will gladly do so if the electorate wishes, but it doesn't change my ballot at all.

I'll act if the electorate asks me to do so; otherwise I'll stand with my ballot as is.
   440. cookiedabookie Posted: December 21, 2020 at 04:20 PM (#5995371)
I helped count ballots last year, and I'm more than willing to do so again. Let me know if help is needed.
   441. cookiedabookie Posted: December 21, 2020 at 04:29 PM (#5995372)
Here's my preliminary ballot through 25, which I've shared as I've tinkered throughout the year. Will probably post on the ballot thread tonight, with a more detailed write up.

1 Andy Pettitte
2 Thurman Munson
3 Joe Tinker
4 Wally Schang
5 Tim Hudson
6 Kenny Lofton
7 Buddy Bell
8 Tommy John
9 Bobby Bonds
10 Jorge Posada
11 Urban Shocker
12 Jeff Kent
13 Lance Berkman
14 Dwight Gooden
15 Bob Johnson
16 Eddie Cicotte
17 Sammy Sosa
18 Vic Willis
19 Ron Cey
20 Sal Bando
21 David Wells
22 Willie Davis
23 Mark Buehrle
24 Art Fletcher
25 Norm Cash
   442. kwarren Posted: December 21, 2020 at 04:56 PM (#5995381)
1. Luke Easter. I’ve put in the sweat and done the research and math (which I’m working to get published as part of a much larger project, so please pardon my not posting it here). An unrecognized McCovey or Stargell, an absolute no-brainer HoMer, the top unelected player by a huge margin.


Here's a link to the Luke Easter story:

https://sabr.org/bioproj/person/luke-easter/
   443. Chris Cobb Posted: December 21, 2020 at 10:18 PM (#5995434)
Still working on fine-tuning my ballot, but here's a close-to-final preliminary, in case there's interest in discussion.

I rank players within decade-by-decade cohorts, using a combination of career value, value above average and five-year peak rate. For position players, I average BWAR and FWAR results, then add a fielding modifier from DRA, weighted equally with the BWAR and FWAR fielding assessments. For pitchers, I use BWAR. I integrate my decade rankings into a ballot based on rank position, scaled to a 30-team league. In other words, in a 16-team league, a player with an in-decade ranking of 16 is adjusted, by multiplying 16 by 30/16, to a scaled rank of 30. I do this to aim for fairness across eras. I consider players’ integrated WAR values, their scaled ranking, and positional distribution in assembling the ballot.

This year’s ballot is challenging to assemble because there are not large differences between the candidates. All of HoM-worthy players would be among the bottom third of the HoM if elected, and most of the eligible are crowded into the bottom tenth. I’m pretty comfortable with the elect-me spots, but after that, the choices are more uncertain.

1. Buddy Bell. Total = 136.98. 1980s. Rank in Decade: 20.6. Stellar defense, above-average offense, and a fairly long career put Bell at the top of this year’s ballot, narrowly ahead of Lofton.
2. Kenny Lofton. Total = 136.07. 1990s. Rank in Decade: 22.9. A similar set of strengths to Bell, but his defense was not as strong.
3. Bobby Abreu. Total = 121.41. 2000s. Rank in Decade: 22.5. Nearly the top offensive player on the ballot. Having considerably more defensive value than Giambi and Berkman, who have comparable offensive value, and a longer career at high productivity than Bobby Bonds or Brian Giles, who have comparable all around production, puts him in an elect-me spot.
4. Johan Santana. Total = 120.6. 2000s. Rank in Decade: 23.5. Great peak and excellent quality throughout his career put him ahead of the pitching cohort.
5. Kevin Appier. Total = 125.6. 1990s. Rank in Decade: 25.6. Peak is not quite the match of Santana’s, but is well ahead of other pitchers. Career after his injury is not as high quality but has more quantity than Santana’s, making the two very close in value and in-decade rank.
6. Jeff Kent. Total = 116.07. 2000. Rank in Decade: 27.5. Near in raw value to his outfield contemporaries at a position that tends to shorter careers. These considerations have caused me to move Kent up considerably. He stands out relative to most other infield options.
7. Robin Ventura. Total = 114.94. 1990. Scaled Rank in Decade: 29.1. Generally overlooked by the electorate, but very similar in value to Kent, although his value lies more on the fielding side.
8. Sammy Sosa. Total = 133.07. 1990s. Rank in Decade: 24. Contextual value issues cause me to slide Sosa down the ballot a bit, but he’s still so far ahead of the next best outfielders of his time that he’s still a solid lower-tier HoMer. He’s a clear step behind Abreu and Lofton, however, so he lands mid-ballot rather than moving to an elect-me spot: I’ve more or less swapped him and Abreu.
9. Brian Giles. Total = 117.72. 2000s. Rank in Decade: 25.5. So far seems undervalued by the electorate. Very similar but a little better than Lance Berkman, who is a top-10 returning candidate. Bat not as big as Berkman’s but more defensive value. This placement does not include any minor-league credit, although I do zero out his last season, which was dreadful. His drop from a 4.8 WAR season at age 37 to -1.9 WAR at age 38 is one of the more epic declines I’ve seen.
10. Sal Bando. Total = 124.71. 1970s. Scaled Rank in Decade: 28.2. Solid career; very good offense at third base; varied assessments of his defense hold him back.
11. Tim Hudson. Total = 115.3. 2000. Rank in Decade: 28.5. The top new arrival. Although he doesn’t have the dramatic peak of Santana, his peak is still strong, and his career is quite substantial for a post-2000 pitcher. Would probably rank higher if he had played his entire career in the National League, where his bat would have played throughout his career.
12. Ben Taylor. Total = 118.73. 1910s. Rank in Decade: 25.3. Bringing him up on to my ballot after looking more closely at first basemen 1900-1920; I would like a firmer sense of his quality, though.
13. Bobby Bonds. Total = 123.9. 1970s. Rank in Decade: 29.4. A complete ballplayer; his case is hurt by his short career, but he has bat, baserunning, and defense. He should win election eventually.
14. Art Fletcher. Total = 116.56.1910. Rank in Decade: 26.8. The best remaining infielder from the underrepresented 1910s. A little better than Joe Tinker and Dave Bancroft; considerably better than Joe Sewell.
15. Thurman Munson. Total = 114.71. 1970. Rank in Decade: 30.6. Another look at him brings him onto my ballot. He’s in the last “decade quota” spot for the 1970s. Because he’s a catcher, he gets the last ballot spot over Hershiser.

16. Orel Hershiser. Total = 118.5. 1980s. Rank in Decade: 30.
17. Chuck Finley. Total = 117.0. 1990s. Rank in Decade: 26.7.
18. Jason Giambi. Total = 116.35. 2000. Rank in Decade: 26.5. He and Olerud are a placed a bit below their rank in decade numbers because of there’s such a glut of post-1990 first basemen.
19. Ned Williamson. Total = 128.3. 1880s. Rank in Decade: 26.9.
20. John Olerud. Total = 116.69. 1990. Scaled Rank in Decade: 27.8
21. Lance Berkman. Total = 114.65. 2000. Rank in Decade: 29.5.
22. Bucky Walters. Total = 107.56. 1940. Rank in Decade: 29.2.
23. Phil Rizzuto. Total = 111.45. 1940. Rank in Decade: 30.
24. Mark Buerhle. Total = 111.7. 2000. Rank in Decade: 30.5.
25. Don Newcombe. Total = 99.19. 1950. Rank in Decade: 30.
26. Urban Shocker. Total = 120.94. 1920s. Rank in Decade: 26.3.
27. Vic Willis. Total = 122.14. 1900s. Rank in Decade: 27.4
28. Jim McCormick. Total = 123.33. 1880. Rank in Decade: 31.
29. Bert Campaneris. Total = 100.78. 1970. Rank in Decade: 38.
30. Frank Chance. Total = 116.75. 1900. Rank in Decade: 34.4.

Required disclosures

Lance Berkman – Has some very good years mixed among his best, so his peak isn’t quite as good as Giambi’s or Giles, so he’s one and two steps behind them, respectively. I don’t oppose his election, but I think he’s farther back in the line than many do. As usual, bat-first candidates do a little better with the electorate than they ideally should.

Wally Schang – ranks somewhere between 30 and 45. His lack of peak, as mentioned by others, weakens his case. Among catchers, my current ranking of eligible is Munson, Posada, Schang. I hope to get caught up on recent advances in analyzing catcher defense by the 2022 election.
   444. kwarren Posted: December 22, 2020 at 07:42 AM (#5995451)
4. Johan Santana. Total = 120.6. 2000s. Rank in Decade: 23.5. Great peak and excellent quality throughout his career put him ahead of the pitching cohort.

27. Vic Willis. Total = 122.14. 1900s. Rank in Decade: 27.4


Nice to see Santana so high on your ballot, but why is Vic Willis with a higher value way down in 27th place.
   445. DL from MN Posted: December 22, 2020 at 10:00 AM (#5995470)
I'm going to answer for Chris using his words:

I rank players within decade-by-decade cohorts, using a combination of career value, value above average and five-year peak rate.
Johan Santana. Rank in Decade: 23.5
Vic Willis. Rank in Decade: 27.4


   446. Chris Cobb Posted: December 22, 2020 at 10:04 AM (#5995473)
why is Vic Willis with a higher value way down in 27th place?

Good question, thanks!

Several reasons, the most important of which for the consideration of the electorate is that WAR is not a period-neutral indicator of value: straight-up WAR comparisons will consistently favor earlier players from periods when competition is weaker. BWAR makes competition adjustments between different leagues at the same time, but it doesn't make all-time competition adjustments, and there is no currently available comprehensive metric that does so.

It is because of this limitation of WAR that I rank players first only within a cohort of contemporaries, organized by decade. I then create the ballot by comparing players first by decade ranking, with those rankings scaled to a consistent measure based on the size of the player talent pool, which I take (conservatively) to be roughly comparable to the number of major-league teams, with adjustments for segregation (once the Negro Leagues begin to organize, I add teams to the major-league pool, so that there are more hypothetical slots for the teens, twenties, and thirties than, for example, the 1880s). For the comparison of Santana and Willis, then, the more important number for explaining their relative rankings is that Santana ranks higher in his cohort (Rank in Decade: 23.5) than does Willis (Rank in Decade: 27.4). Willis's career was mainly in the 1900 decade, with 16 teams and nascent organization in Black Baseball, so it has a soft quota of 17 HoM slots in my system. Santana pitched in the 2000 decade, which has a soft quota of 30. That is the largest slot number, so I scale all other decades to 30. Willis's raw in-decade ranking is 15.5, which scales to 27.4 when there are 30 slots between the top and bottom of the HoM-worthy cohort instead of 17. The difference between Santana's in-decade rank and Willis's is not small, so that, in context, I am confident that Santana was a more meritorious pitcher than Willis.

On my ballot, I also rank Willis fairly conservatively because his decade is well represented in the Hall of Merit (18 players, 17 slots). Its pitching cohort is particularly well represented (contemporaries Young [1/2 career assigned to 1890s in my decade system], Mathewson, Plank, R. Foster, Waddell, McGinnity, and Brown accrue 6.5 HoM spots for pitchers in a decade whose size would suggest 4.5 spots) so adding Willis on top of the already elected players of that era would hinder the election of equally or more deserving players from the last 50 years of baseball history and from other positions.

To put it another way, Willis is a borderline candidate from a well-represented era in which the abundance of elected pitchers suggests that it was easier for pitchers to dominate. That's a context that leads me to rank Willis conservatively because all the factors point to his type of player being a little bit overrated by the total system. By contrast, Jeff Kent is a borderline candidate with the same in-decade rank as Willis (27.5) who played in an era that is not yet well represented at a position that is slightly underrepresented historically and in a position group (infielders) that is definitely underrepresented both historically and in the 2000 decade. That's a context that leads me to rank Kent aggressively, because all the factors point to his type of player being a little bit underrated by the total system.

In a context in which the top eight candidates by in-decade rank fall into in the bottom 70th and 80th percentiles and the next 20 candidates are all in the bottom percentile of the HoM, that leads to a ballot spread of 20 places between Willis and Kent because of the era and positional factors that create the context in which I assess their merit.
   447. DL from MN Posted: December 22, 2020 at 10:07 AM (#5995474)
5 6 Kenny Lofton 323 20 1 5 2 1 4 2 1 1 1 1 1
6 10 Johan Santana 246 19 2 1 1 2 1 4 2 1 2 2 1
7 12 Sammy Sosa 231 18 1 1 2 1 3 1 2 1 2 1 1 1 1
8 9 Jeff Kent 231 14 3 1 2 1 2 1 1 1 1 1
9 13 Lance Berkman 216 13 1 4 1 2 3 1 1
10 n/e Bobby Abreu 200 15 1 1 1 2 1 3 2 1 1 2


Looking over the finishers from last year Kenny Lofton has the pole position. Sosa and Kent were tied with Kent having more upside due to not appearing on as many ballots. If Kent gets just a couple ballot adds he's going to leapfrog over Sosa.
   448. DL from MN Posted: December 22, 2020 at 12:01 PM (#5995514)
test
   449. Bleed the Freak Posted: December 22, 2020 at 01:40 PM (#5995535)
DL 447 / Berkman and Abreu are close enough they could leapfrog without adding much support / if they vault to elect-me spots.

Or we could have a vaulting candidate, like Hurley McNair or Gavvy Cravath, that have relatively impressive MLEs.
   450. Bleed the Freak Posted: December 22, 2020 at 01:43 PM (#5995536)
Returnees
Sammy Sosa (RF #15): We have plenty of RFs, so I don't feel especially moved to include Sosa
Lance Berkman (1B #26): Just off the end of the position for me, but we already have about five or six more 1Bs than we need
Bobby Abreu (RF #25): He'll be a HOM borderliner for me until we reach a place where the 26th best hitter at each position
Bobby Bonds (RF #21): We have plenty of RFs, so I don't feel especially moved to include Bonds either


Doc, for my benefit and others, you likely have a deep list already constructed, could you share a full top ~30 if it's readily available?
   451. Al Peterson Posted: December 22, 2020 at 02:33 PM (#5995542)
Prelim ballot, we’ll see if there is other adjustments prior to final ballot.
2021 ballot – we go through the backlog as a couple fringe candidates arrive, mostly continue looking into the past of some forgotten electees.

Methodology in brief: The system used for my ranking entails a little bit of everything including WS, WAR, OPS+/ERA+. Ratings include positional adjustments, additions to one’s playing record for minor league service, war, and NeL credit and for our real oldtimers some contemporary opinion thrown in. The results of this work tend to favor prime/peak players over career types but that is not 100% tried and true. Last year’s placement is in parenthesis.

1. Kenny Lofton (3). Lofton has served his group-vetting time, I’m fine seeing him as top slot at this point. Baserunning fiend, table setter for some excellent offenses. Played for 11 teams total – 1276 games for Indians, 827 games for the other 10 combined. The what if for Lofton: had he focused on baseball in college, dropping the basketball play with Arizona, would he had gotten to the majors earlier?

2. Phil Rizzuto (2). Holy Yankee shortstops Batman!! I’ve done my minor league & WWII absence calibration so Scooter scoots to ballot position. Glove first but the offense during prime years was nothing to sneeze at either. Lofton and him are close to me, I’ll go Phil 2nd since there is more projected work done here.

3. Bobby Bonds (4). Even with the constant trades, drinking problem and whatnot his combination of speed/power made him a very valuable player. He wasn’t the next Mays, or as good as his son, but we’re talking about a RF who could steal bases and field his position. All five tools on display.

4. Tommy Leach (6). Combination hot corner/centerfielder could field a little, hit a little. Second all-time in inside-the-park home runs to Wahoo Sam Crawford. Someone else stated he was uniquely valuable in his particular era and I agree he meant more in the particular era he performed in – you don’t get to play 900+ games at 3b & CF without having a good defensive skill set. Useless trivia: Still holds World Series record with 4 triples in a single series.

5. Bobby Abreu (9). Doubles power, good batting eye, sneaky speed in a corner outfielder. Kind of guy you overlook – supported by the 2 All-Star appearances – who did things that help your team win ballgames.

6. Tony Mullane (7). Old time pitcher who threw plenty well, a good hitter to boot. Had some playing time issues since he missed seasons due to being blacklisted. He’s amongst the best of his era when accounting for the time outside of baseball due to conflicts with different leagues. Goes on the all-Nickname team as well.

7. Lance Berkman (10). He’s the type of extended prime player my system tends to favor. Bob Johnson of his era, likely to have the same fate where 30 years from now his name will be unfamiliar to many casual baseball fans.

8. Mickey Welch (11). 300 game winner in the house. Was it due to luck, run support, bad opponents? Still a feat to accomplish, sometimes I need to remind myself that and not totally overlook Smilin’ Mickey. Seemed to pitch well against the other front line starters of his day.

9. Bob Johnson (13). Always a bit underrated in Win Shares due to quality of teams he played on. His career has war years that need discount. But also a couple years at the beginning of his career were in the PCL where he was more than major league quality. The tail of his career is nonexistent since the 1946 avalanche of returning War players pushed him back to the minors.

10. Sammy Sosa (14). Peak power that was enough to make people start walking him. This increased his value as it upped his OBP skills, doubling the value added. Early in his career he had base stealing and defense as assets.

11. Jeff Kent (16). Highest 2nd basemen I’ve got, the glove holds him back but not a deal breaker. Still might be underrating position slightly…

12. Ben Taylor (17). Holding steady here. He lingers on ballot fringe, we have 1B to spare. Some NeL updates have downgraded Taylor but still feel ballot spot warranted.

13. Tommy John (18). He remains in a Tiant/Willis/Pettitte cluster. Offer value in different ways but all fine pitchers. As Tiant now elected he’s on that in/out line.

14. Buddy Bell (15). The gap between top-tier 3Bmen is not large for the position when he played in the 70s and 80s. Body type didn’t really look the part of a great glovemen but few would deny he was outstanding.

15. Vic Willis (19). Another pitcher this one from the turn of the century..the 20th, not the 21st. I’d like to squeeze one more hurler on the ballot and feel his era is owed another slot. HOM worthy? Eh, no strong feelings.

Next up, but off ballot:

16. Andy Pettitte. Not far off as I thought he would be. Above average production for awhile does get you somewhere in life.
17. Tim Hudson. Neck-and-neck with Pettitte. Extended prime, thought of decently by media at the time with 4 All-Star appearances and 4 Top 5 Cy Young finishes.
Next five– Sal Bando, Willie Davis, Jason Giambi, John Olerud, Mark Buerhle

Disclosures: .
Top 10

Johan Santana: Get in the pitcher line. Probably somewhere in the 50s.
Wally Schang: We’re doing this guy? Alrighty, not my cup of tea. If ordering catchers have him below Posada (Jorge is around #30), real close to Tenace/Munson/Lombardi but that’s aways back.
   452. kwarren Posted: December 22, 2020 at 03:26 PM (#5995556)
Thanks Chris for your explanation in Post #446. You obviously put a lot of time, thought, and analysis into your ballot.

On my ballot, I also rank Willis fairly conservatively because his decade is well represented in the Hall of Merit (18 players, 17 slots). Its pitching cohort is particularly well represented (contemporaries Young [1/2 career assigned to 1890s in my decade system], Mathewson, Plank, R. Foster, Waddell, McGinnity, and Brown accrue 6.5 HoM spots for pitchers in a decade whose size would suggest 4.5 spots) so adding Willis on top of the already elected players of that era would hinder the election of equally or more deserving players from the last 50 years of baseball history and from other positions.


One problem with this example is that Willis is better four of the pitchers inducted from this era - Griffith, McGinnity, Waddell, & Brown. This is like the Hall of Fame voters saying "we can't put Whitaker and Grich in the Hall of Fame because there are only so many spots available and Mazeroski & Huggins have taken their spots".

JAWS total for pitchers who pitched significant portions of their career in 1898 - 1910:

Vic Willis 56.7
Clark Griffith 56.1 HOM
Joe McGinnity 55.9 HOM
Rube Waddell 54.0 HOM
Mordecai Brown 51.1 HOM

It would be interesting to know how your system values Griffith, McGinnity, Waddell, & Brown and how they compare with Willis.

You do need to treat all eras of baseball history fairly. It’s because we want informed voters making informed decisions on the entire electorate, not just the players they remember.”

This is the Hall of Merit's guide on the issue of different eras. It does not say that there needs to be equal representation for all positions for all eras. It just says that all eras must be considered fairly. From a statistical point of view the probability of pitching talent being equally distributed over the different eras is essentially zero. We do know that the talent will be distributed randomly, but randomly is clearly not equivalent to equally. So if the talent is not distributed equally over eras why are so many voters going through such hoops and machinations to attempt to make the representation of each era equal. When they refer to the authority of the instructions, as requiring that, I think it is just a misunderstanding of the requirements.

We are currently in an era when starting pitchers in general have less value than any other time in history. When current pitchers start to become eligible to the Hall of Merit, I hope that the electorate will not feel bound to elect starting pitchers in the same numbers that they have done for other eras. The role of pitching is just important as ever, but that value is now being spread over 20 to 25 pitchers per team (as opposed to 12 to 15 in earlier eras), so the average value of each individual pitcher has never been lower. If the best pitchers continue to pitch 200+ innings per season, this may not be a huge issue, but seeing the way that Blake Snell was used so successfully by the forward thinking Rays, I doubt that will be the case. There seems to be a universal belief in the game that a rested reliever will perform better than an ace starter when it comes to facing a line-up for the 3rd time, and there is lots of evidence to back this thinking up.

   453. DL from MN Posted: December 22, 2020 at 03:55 PM (#5995567)
From a statistical point of view the probability of pitching talent being equally distributed over the different eras is essentially zero.


Yes, it's much less likely to be distributed mostly at the turn of the previous century.
   454. kcgard2 Posted: December 22, 2020 at 03:56 PM (#5995569)
kwarren (#452): There are more ways to judge pitchers than just JAWS. Even granting that premise, the ones you've listed are basically washes except for Brown. So why would/might other voters disagree with you on whether Willis was better than the others? Waddell was much more effective on a per-inning basis, and relied FAR less on the defenses behind him for his effectiveness. I'd easily prefer Waddell. McGinnity has black ink all over the place, though personally I do think Willis may have been better, at least I would not push back against that assessment. Call them a wash. Griffith, they're close but I think I agree with you I'd personally give Willis the edge. Brown also has a bunch of black ink and is basically an equal of Willis, I prefer Brown, even though his JAWS is worse than McGinnity or Griffith. Looking at the four cases you presented, I prefer Willis only to one of them, but you are presenting the argument as if it's cut and dried that Willis is clearly better than all four. Clearly I'm not alone in those assessments.

I sympathize with the points of your final paragraph, but different voters may interpret that directive differently. Saying that Willis is lower down because compared to contemporaries he didn't stand out as much as other candidates, even though his raw value may be higher than those other candidates, is a fair way to handle era representation/comparisons for HOM.

Dr. Chaleeko (ballot thread): you'll need to revise your ballot, since Andruw was inducted last year. Also, I find it interesting that you list Berkman as a 1B. He spent more time in corner OF (but not more time at either of LF or RF specifically). Even still, he seems to have generated the most career value in LF as best I can tell, but it's close. I've always preferred that LF and RF be treated as a combined position (which would make Berkman much easier to classify), but maybe that's just me.

Chris Cobb (#443): I like your ballot :) You also came along at just the right time to explain in more depth what I was saying about Vic Willis and why voters haven't necessarily overlooked him a bit further upthread.
   455. Rob_Wood Posted: December 22, 2020 at 04:11 PM (#5995572)
I generally agree with kwarren's points above.

I have looked back at my early HOM ballots and see that I did have each of Griffith, McGinnity, Waddell, and Brown on several ballots (typically in the middle or lower half of my ballot). I had Willis in my "consideration" set but I don't think Willis ever appeared on any of my ballots.

In retrospect, I think I undervalued Willis at the time and will definitely consider him going forward to be a legitimate HOM candidate.
   456. kwarren Posted: December 22, 2020 at 06:03 PM (#5995588)



   457. kwarren Posted: December 22, 2020 at 06:09 PM (#5995590)
Rob Wood (#418)

By CPASR (regular season only), Mariano Rivera has .652. David Wells is ... (drum roll) ... also at .652. These are around 380th best all-time.


Looks like your CPASR is a "hell of a system". :)

   458. Carl Goetz Posted: December 22, 2020 at 07:44 PM (#5995605)
I don't think I've posted my prelim ballot yet so I thought I'd let it marinate here for a couple weeks before posting to the real thread.
I use a system based on WAR, WAA and WAG (WAA and WAG are Zeroed for negative years). I use a blend of WARs on Baseball Gauge that's roughly Baseball Reference WAR for Offense, 70% DRA-30%DRS/TZ for defense and 75% BBRef-25%FIP for Pitching. I make several adjustments to these for Schedule length, defense regression, postseason bonus, catcher framing etc. I blend Career WAR with WAA to derive my Career Score and I blend WAA and WAG to derive my Peak Score. I do have an overall rating score which is simply the geometric mean of my career score and peak score, but I honestly look at the 2 separate scores when making my rankings. I do look at both relatively equally, but I weigh peak a bit more heavily. I also look at subjective criteria like where the player ranked with his contemporaries and also where he ranks using other methodologies such as WS and Kiko's PWORL. General beliefs/biases; I think we are short on Catchers, long on pre-1894 pitchers, but a little short on pitching overall, and that there was more value at 3B than SS in the 70s.

1) Thurman Munson: With Catcher ADj, has the top peak Score and overall score for this ballot. The next few are close, but Thurman is easy #1 for me.
2) Wally Schang: Ditto to Thurman, only he's #2. He's easily the best MLB catcher between Bresnahan and the 30s stars. I rate him better than Santop and I agree with Santop's election.
3) Buddy Bell: He'd be top in career and peak scores if it weren't for a catcher adjustment. He ranked here even after regressing his fielding 15%.
4) Johan Santana: Peak is even more important to me for pitchers than position players. Santana is an easy #1 pitchers for me.
5) Kenny Lofton: Best OF on the board. Great blend of career and peak scores.
6) Art Fletcher: Phenomenal defense even with a lot of regression. I did 50-50 for Tinker and Fletcher between DRA & TZ since I think DRA overrates deadball SSs a bit. I also did my standard 15% regression.
7) Joe tinker: Really nice defense and a strong peak. See Fletcher above.
8) Andy Pettitte: He's more career, but there's a lot of postseason value here. In some ways, I don't think its fair since he had way more postseason opportunities than most, but he still played well for a long time with a lot of inning on his arm.
9) Bernie Williams: Moved up my list last year once I started regressing defense by 15%. Also, he gets a lot of postseason credit.
10) Tommy Leach: Good blend of Career and peak. I go back and forth on where he is position-wise. He's definitely behind Bell if 3B and Lofton if OF. Only slightly behind Bernie if OF.
11) Orel Hershiser: I was surprised to see his career score was higher than peak score. I usually think of him as a peak guy from the mid to late 80s Dodgers years.
12) Vic Willis: Example of a guy that doesn't score great in my system. I believe my system is missing something and that he's a really nice peak candidate. He rates far above any other pitcher in WSAB.
13) Roy White: Scored above Leach and Bernie in raw numbers. I just don't quite trust the defensive numbers, so I scaled him back a little more than most players.
14) Bobby Veach: Same as Roy White.
15) Sammy Sosa: Was 19 last year; 4 got elected, no newbies on my ballot so simple math.

Top 10 Returnees
Kent Ranks 35th on my ballot.
Berkman My system really doesn't like him (at least compared to the rest of you. He's roughly 45th on my ballot.
Abreu Probably about 20th. Bonds is definitely better by my system. I did expect him to rank higher than this.
Bonds Ranks 17 just off ballot.
Bando Ranks 33 right now. I have Ron Cey at 32.

Let me know your thoughts.
   459. Chris Cobb Posted: December 22, 2020 at 08:31 PM (#5995608)
Kcgard covers basically the ground that I would cover, based on a perspective on the pitchers in question that is similar to mine: not one of them stands out greatly in relation to the others. I should emphasize that I certainly do consider Vic Willis, as Rob Wood puts it, "a legitimate HoM candidate." The fact that he is in my top 30 indicates, essentially, that I think he should be elected. The issue is one of relative priority in relation to the other 30 or so current candidates who should be elected at some point. If this discussion causes voters who had lost track of Willis to bring him back into consideration, that is all to the good.

With that preamble, let me flesh out the account of my assessment of Willis relative to some of his pitching contemporaries.

JAWS is a somewhat blunt instrument for comparing candidates, and it is tailored to model the career value orientation of the Hall of Fame. The Hall of Merit electorate has historically (and rightly, in my view) tended to give more weight to peak value in its assessment of merit than the HoF has. Willis is a career-type candidate, so JAWS will naturally favor him over more peak-oriented candidates: none of the pitchers kwarren has compared to Willis using JAWS have as much bulk in their careers. However, on a per-inning basis and, to a lesser extent, on a seasonal basis, they compare favorably to Willis.

My "system" sees Griffith as quite notably superior to Willis (I was a strong backer of Griffith back in the day), Waddell as a little bit better than Willis, McGinnity as just about the same, and Brown as a little bit worse. To show how they line up in my system quantitatively, I’ve included their raw totals, their rank in decade, and also one important measure that is part of my assessment, which is the WAR per 250 IP that they earned across their best five (or more, if a longer stretch advantages them) consecutive seasons. This measure shows most clearly where these other pitchers are making up ground on Willis that they lose on career value.

Clark Griffith. Total = 133.7. Rank in Decade (1890): 17.6. Peak Rate 1894-98 = 5.76 WAR/250 IP
Rube Waddell. Total = 123.62. Rank in Decade (1900): 23.8. Peak Rate 1902-06 = 6.70 WAR/250 IP
Joe McGinnity. Total = 121.93. Rank in Decade (1900): 25.6. Peak Rate 1899-1904 = 5.53 WAR/250 IP
Vic Willis. Total = 121.84. Rank in Decade (1900): 27.4. Peak Rate 1899-1903 = 4.71 WAR/250 IP
Mordecai Brown. Total = 115.21. Rank in Decade (1900): 32.6. Peak Rate 1906-10 = 6.06 WAR/250 IP

If I were setting up the HoM purely according to my own assessments at this point in the history of my engagement with the HoM project, Willis would be in and Brown would be out, but I can easily see how more peak-inclined voters could prefer Brown to Willis: Brown was an outstanding pitcher at his peak.

The fact that all of these pitchers, except Griffith, are in the lower third of the HoM together makes the "can't put in Grich because Mazeroski has his spot" comparison kwarren has made irrelevant. Grich is a clear, no doubt, top half of the HoM player, and Mazeroski is nowhere near the HoM. Just their career WAR is enough to show that Grich was twice the player Maz was (71.1 vs. 36.5 BWAR, to be exact). To exclude Grich because Mazeroski has filled the 2B quota already would be an egregious error. Leaving out Vic Willis while inducting Mordecai Brown is an error, if it is indeed an error, about on the scale of leaving out Chuck Finley while inducting Roy Oswalt. We haven’t made that mistake with the modern players, in part because we haven’t come close to inducting either pitcher but I think that Finley and Oswalt stand in similar positions, relative to their era, as Willis and Brown, and for similar reasons.

And now for an aside that gets back into the historical weeds--prompted by Rob Wood's reflection on his having set Willis aside a while back:

When I dig back into the old records having to do with my own degree of support for Willis, I am reminded that when I first studied Willis's case seriously in 2004, I was working with an interesting tool that has long since fallen into neglect, the Run Support Index. Invented and (for a time) popularized by Chris Jaffe (no relation to Jay Jaffe) who was involved in the Hom project for a time, RSI used retrosheet to analyze pitchers' actual run support, relative to league average, as a way of assessing their won-lost records: how much was it due to their run support? One could also consider, relatedly, whether pitchers might have been "pitching to the score" and so on. At this time, win shares, and an early version of Clay Davenport's WARP (published at Baseball Prospectus) were the only comprehensive metrics available, and it was quite clear that there were significant limitations to both. So I set out to work out pitchers' wins above average by using RSI to calculate their run support in each season, their fielding support based on team defensive efficiency relative to league average, as adjusted by Davenport's calculation of their hits saved or excess hits allowed, which I took from WARP. Willis fared very poorly by that measure: for example, his 1898 rookie season, which BWAR evaluates as 2.4 wins above average, I found to be just 0.9 wins above average, because with the run support and fielding support Willis received, he only won about one more game than an average pitcher would have done, given that level of support. Overall, in his career, I found Willis to have a W+ of only 10.8, which placed him far behind the pitchers to which we are now comparing him. McGinnity had a W+ of 38.3, Griffith was at 36.5, Waddell at 30.1, and Brown at 21.2. These totals more than offset Willis's advantages in innings pitched. These findings led me not to support Willis, and I didn't revise my assessment of him until I revised my system to work off of BWAR, with no reference to Win Shares, WARP, or my home-grown W+ analyses, and ran all of the early pitchers through the system.

The consistency and calculative rigor that we now rely on from metrics like BWAR and FWAR (not that they are perfect or beyond dispute and adjustment!) was not available back in those days, and we were figuring out analytical tools as we went along in the project. As the metrics changed our assessments tended to change with them, but sometimes assessments and attitudes towards certain players became set in ways that might properly have shifted more. On the other hand, it could be worth revisiting some of that prior work. I abandoned my RSI analyses because Chris Jaffe's published numbers eventually disappeared from the web and because as we reached the modern era in which complete games were no longer the norm, RSI became pretty much irrelevant to analysis of pitching quality. But it might be interesting to look back at the early pitchers from that perspective, especially now that more rigorously quantified accounts of team defensive quality are available.

One last historical note. Some indication of how the electorate's views have changed over the years can be seen in the history of when the pitchers in question here were elected:

Mordecai Brown 1925
Joe McGinnity 1928
Clark Griffith 1971
Rube Waddell 1986

   460. kwarren Posted: December 22, 2020 at 10:39 PM (#5995633)
To show how they line up in my system quantitatively, I’ve included their raw totals, their rank in decade, and also one important measure that is part of my assessment, which is the WAR per 250 IP that they earned across their best five (or more, if a longer stretch advantages them) consecutive seasons. This measure shows most clearly where these other pitchers are making up ground on Willis that they lose on career value.

Clark Griffith. Total = 133.7. Rank in Decade (1890): 17.6. Peak Rate 1894-98 = 5.76 WAR/250 IP
Rube Waddell. Total = 123.62. Rank in Decade (1900): 23.8. Peak Rate 1902-06 = 6.70 WAR/250 IP
Joe McGinnity. Total = 121.93. Rank in Decade (1900): 25.6. Peak Rate 1899-1904 = 5.53 WAR/250 IP
Vic Willis. Total = 121.84. Rank in Decade (1900): 27.4. Peak Rate 1899-1903 = 4.71 WAR/250 IP
Mordecai Brown. Total = 115.21. Rank in Decade (1900): 32.6. Peak Rate 1906-10 = 6.06 WAR/250 IP


I agree that this is a valid metric to measure a players peak level of performance and has value. It may also indicate why Willis is still be on the outside looking in.
   461. Bleed the Freak Posted: December 23, 2020 at 09:52 AM (#5995662)
When I dig back into the old records having to do with my own degree of support for Willis, I am reminded that when I first studied Willis's case seriously in 2004, I was working with an interesting tool that has long since fallen into neglect, the Run Support Index. Invented and (for a time) popularized by Chris Jaffe (no relation to Jay Jaffe) who was involved in the Hom project for a time, RSI used retrosheet to analyze pitchers' actual run support, relative to league average, as a way of assessing their won-lost records: how much was it due to their run support? One could also consider, relatedly, whether pitchers might have been "pitching to the score" and so on. At this time, win shares, and an early version of Clay Davenport's WARP (published at Baseball Prospectus) were the only comprehensive metrics available, and it was quite clear that there were significant limitations to both. So I set out to work out pitchers' wins above average by using RSI to calculate their run support in each season, their fielding support based on team defensive efficiency relative to league average, as adjusted by Davenport's calculation of their hits saved or excess hits allowed, which I took from WARP. Willis fared very poorly by that measure: for example, his 1898 rookie season, which BWAR evaluates as 2.4 wins above average, I found to be just 0.9 wins above average, because with the run support and fielding support Willis received, he only won about one more game than an average pitcher would have done, given that level of support. Overall, in his career, I found Willis to have a W+ of only 10.8, which placed him far behind the pitchers to which we are now comparing him. McGinnity had a W+ of 38.3, Griffith was at 36.5, Waddell at 30.1, and Brown at 21.2. These totals more than offset Willis's advantages in innings pitched. These findings led me not to support Willis, and I didn't revise my assessment of him until I revised my system to work off of BWAR, with no reference to Win Shares, WARP, or my home-grown W+ analyses, and ran all of the early pitchers through the system.


"Back in the day." Joe Dimino's pennants added metric agreed with this, incorporated Baseball Prospectus WARP/I think defensive evaluations, Willis was completely off the radar.
   462. Chris Cobb Posted: December 23, 2020 at 01:00 PM (#5995701)
"Back in the day." Joe Dimino's pennants added metric agreed with this, incorporated Baseball Prospectus WARP/I think defensive evaluations, Willis was completely off the radar.

Yes; we've probably spilled more e-ink and done more revision of thinking about pre-1920 pitchers than any other group of players. Although I am not a supporter of Mickey Welch, I became a lot more understanding of some recent support showing up for him when I reviewed the 1880s last year and looked closely at what WAR systems are saying about him now: his WAR is coming in much closer to McCormick and Radbourn than it used to. I remember a time when Welch's career WAR/P was in the 40s while the acknowledged top 1880s pitchers were up in the 60s to 80s. Welch is still at the back end of the group, but he's a lot closer than he was at one time by the then-standard metrics.

If I had (a lot) more time, I'd be interested in digging in more deeply to the changes that have led to the shifts in comprehensive metrics for the early pitchers and in kicking the tires on how RSI meshes with WAR-based assessments.
   463. Bleed the Freak Posted: December 23, 2020 at 02:18 PM (#5995715)
Yes; we've probably spilled more e-ink and done more revision of thinking about pre-1920 pitchers than any other group of players. Although I am not a supporter of Mickey Welch, I became a lot more understanding of some recent support showing up for him when I reviewed the 1880s last year and looked closely at what WAR systems are saying about him now: his WAR is coming in much closer to McCormick and Radbourn than it used to. I remember a time when Welch's career WAR/P was in the 40s while the acknowledged top 1880s pitchers were up in the 60s to 80s. Welch is still at the back end of the group, but he's a lot closer than he was at one time by the then-standard metrics.

If I had (a lot) more time, I'd be interested in digging in more deeply to the changes that have led to the shifts in comprehensive metrics for the early pitchers and in kicking the tires on how RSI meshes with WAR-based assessments.


And if you incorporate Baseball Gauge WAR 50/50 with Baseball Reference for the early ball pitchers, Charlie Buffinton and Tommy Bond look REALLY good too, I would argue they appear to be a cut above Jim McCormick even. Bond has the added challenge of earlier leagues / smaller teams to face, how much value comes from his defense and not having to face great Boston teams.
   464. Michael J. Binkley's anxiety closet Posted: December 23, 2020 at 04:41 PM (#5995744)
And if you incorporate Baseball Gauge WAR 50/50 with Baseball Reference for the early ball pitchers, Charlie Buffinton and Tommy Bond look REALLY good too, I would argue they appear to be a cut above Jim McCormick even. Bond has the added challenge of earlier leagues / smaller teams to face, how much value comes from his defense and not having to face great Boston teams.


I use this exact ratio for my pitching evaluations. And as a peak voter, I care only about above average and above greatness (same level above average that replacement level is below average).

Spoiler alert: I have both Bond and Buffington on my upcoming ballot. And while I know that Dr. Chaleeko and others have expressed concern that we already have enough pre-1893 pitchers in the HoM, I look at it this way: Along the lines of "It's better to let a guilty man go free than imprison an innocent one," I feel it's better to make sure that all qualified players are in, regardless of existing members and era/position quotas. Not to mention this presupposes that everyone the HoM in the past would still be considered worthy today. Back when most of the pitcher's box era pitchers were elected, most voters relied on WS and the old BP WARP. And while I know WS still has some of it's adherents here today, a significant portion of voters have moved on to different WAR systems which may yield different valuations.

That said, in my system, I have Bond as the 3rd best pre-1893 pitcher (behind Spalding and Clarkson, respectively) and Galvin as clearly out and also Ward (1/2 pitcher, 1/2 shortstop) as lacking. Here is my evaluation of pre-1893 pitchers (PEACE+ score, with 100 being the approximate in/out line, mWAA and mWAG, with negative seasons zeroed out):

1. Al Spalding - 161.13/45.3/18.7
2. John Clarkson - 153.02/44.6/14.2
3. Tommy Bond - 143.12/39.3/17.1
4. Charlie Buffinton - 125.94/35.7/15.6
5. Tim Keefe - 119.57/42.2/9.3
6. Bob Caruthers - 112.94/33.9/11.8
7. Old Hoss Radbourn - 107.99/34.8/9.4
8. Silver King - 89.06/25.8/9.1
9. Monte Ward - 78.07/29.5/4.3
10. Jim Whitney - 74.68/25.9/6.5
11. Pud Galvin - 71.90/28.3/4.7
12. Mickey Welch - 58.09/24.3/2.6
13. Tony Mullane - 57.27/25.2/2.1
14. Jim McCormick - 54.66/23.8/2.2
   465. Kiko Sakata Posted: December 23, 2020 at 06:11 PM (#5995753)
Maybe this is more for the discussion thread, Kiko, but I have a tremendous amount of respect for your system. I saw your ballot is very friendly to "bat-first" candidates who were suspect on defense: Kent, Stephens, Berkman, Giambi, Strawberry, Harrah. I'm curious why this is so.


(from Ardo in the ballot thread)

First, thank you.

Fielding is a smaller share of value in my system than in WAR systems, basically because fielding credits are shared between pitchers and fielders in my system. So, yeah, bad fielders are less hurt by my system - and good fielders get less credit (this is a big part of my system's attitude toward Buddy Bell and a smaller part of my system's view of Kenny Lofton).

I and others have discussed this in past ballot threads (I want to say the last two years, not so much in the current thread). I wrote an article that was in the Baseball Research Journal a few years ago that touched on this. This article on my website is basically that article although I think I may have updated some of the math.

As an aside, my system actually thinks that Vern Stephens was a pretty good fielder in his prime (and then a fairly bad fielder as he aged, so he ends up around average for his career). When the Red Sox acquired him, they moved Pesky to third base and installed Stephens at shortstop. My system also views Harrah and Strawberry as essentially average fielders over the course of their career (and technically the relative spread of fielding won't matter for average fielders - multiply 0 by 2 or 5 or 10 or 100 and you still get 0). (My system agrees that Berkman, Kent, and Giambi were below-average fielders - pretty bad in the case of Kent; very bad in the case of Giambi: for these three, it's what I have previously called a case of "valuation rather than evaluation".)
   466. The Honorable Ardo Posted: December 23, 2020 at 06:20 PM (#5995755)
For me, it's more of a thematic question: what's the difference between pitching 55.5 feet off flat ground versus pitching 60.5 feet off (gradually) a mound?

So I decided to look at the NL's ten best pitchers in 1892, the last year of the pitcher's box. How did they adapt?

1. Cy Young. Almost unaffected. Kept on being Cy Young.

2. Gus Weyhing. Collapsed. 1892 was his fifth straight good season in a row. He wasn't even league average in any season at 60.5 feet.

3. Kid Nichols. Led the 1893 NL in WHIP. Had an off year in 1894, then kept on being Kid Nichols.

4. Bill Hutchison. Collapsed, though more understandably than Weyhing; he worked an unsustainable workload in 1890-92, turned 33, and didn't need to support himself through the game (he was a Yale graduate and already a successful businessman)

5. Amos Rusie. Just as dominant at the longer distance, though his IP totals slowly tapered off. Missed 1896 in a salary dispute well-known to aficionados of 19th century baseball. Had his career curtailed by injury in 1898.

6. Ed Stein. Had one good season in 1892 (his first full season at age 22), then never repeated it afterwards.

7. Nig Cuppy. A rookie in 1892, he seamlessly adapted to the longer distance and was on a Hall of Merit trajectory thru 1897, when he suffered an arm injury. He hung on as a part-time pitcher through 1901.

8. Frank Dwyer. Consistently good-not-great from both the shorter and longer distance. Suddenly lost effectiveness in 1899, ending his career at age 31.

9. Tim Keefe. In his decline phase anyway. After sustained dominance thru 1890, he fell apart in 1891, then had a strong bounce-back season in 1892. Made 22 okay starts in 1893 and voluntarily retired in August.

10. Tony Mullane. Pitched a lot of innings for two teams in 1893 at a league average rate. Utterly collapsed in 1894 (granted, he was 35 and a hard liver) and that was it.

Only Young (one of the top pitchers ever) and Nichols (a Whitey Ford figure who had great support from dynastic teams, overrated by history, but still clearly above our threshold) avoided taking a hit.
   467. kcgard2 Posted: December 23, 2020 at 08:18 PM (#5995765)
Binkley, what adjustment if any do you make for Caruthers pitching almost entirely in the AA? It's hard to believe he'd rank that high if you're making any adjustment. I guess same question for Keefe since 1883 was an utterly ridiculous piece of work. Actually most of these guys have at least one of their super high WAR seasons in the AA, UA, NA, or PL. But Caruthers has 80% of his career value there. Silver King has no career to speak of outside the AA and PL, so I'd wager you're making no adjustments here.
   468. Chris Cobb Posted: December 23, 2020 at 08:39 PM (#5995768)
Bleed the Freak wrote:

And if you incorporate Baseball Gauge WAR 50/50 with Baseball Reference for the early ball pitchers, Charlie Buffinton and Tommy Bond look REALLY good too, I would argue they appear to be a cut above Jim McCormick even. Bond has the added challenge of earlier leagues / smaller teams to face, how much value comes from his defense and not having to face great Boston teams.

What is the rationale for incorporating Baseball Gauge WAR into evaluations for the early pitchers? I am not familiar with how it handles them.
   469. Michael J. Binkley's anxiety closet Posted: December 23, 2020 at 09:06 PM (#5995769)
kcgard2-

The WAA numbers I use are not the ones straight from The Baseball Gauge using a straight average of bWAA and gWAA. Because the WAA figures are league-specific, but the WAR figures are league-neutral, I reverse-engineer league neutral WAA for each player season from the player's WAR for that particular year. I then multiply that WAA number by my standard deviation adjustment to get my final mWAA number.

For example, 1885 Bob Caruthers in the AA. According to BG, Caruther's 1885 50/50 bWAA/gWAA WAA is 7.3. His WAR under this is 10.1. Since he pitched 482.1 inning that year, I convert that 10.1 WAR to 5.6 league-neutral WAA. The 1885 American Association then has a standard deviation adjustment of .898 for 1885. 5.6*.898=5.0. Thus, I credit Caruthers with 5.0 mWAA (and consequently .5 mWAG) for 1885 instead of the 7.3 league-specific WAA he is allocated in the base custom metric.

Edited to add: Although I do make these adjustments, my final PEACE+ numbers do also include some bonuses for being the best position player/pitcher in a particular league/all of baseball. So for example, Silver King gets bonus points for being the best player in baseball in 1888, even accounting for the AA, and for being the best pitcher in baseball in 1890 in the PL. My system is very peak-heavy.
   470. Chris Cobb Posted: December 23, 2020 at 11:25 PM (#5995788)
Michael J. Binkley's anxiety closet, I have to ask some follow-up questions about the data behind your numbers, specifically with respect to the numbers provided by GWAR, since it appears that they constitute 50% of numbers you use. I'm not sure what I'm seeing at the Baseball Gauge site when I pull up GWAR numbers for the early pitchers, but I have a hard time seeing how the numbers are credible or envisioning a mode of analysis that would lead to the published results.

Just spot-checking, I looked at the numbers for Al Spalding, whom your system identifies as the greatest pre-1893 pitcher. For 1872, which BWAR agrees is one of his best seasons, GWAR presents Spalding as having earned 19.1 pitching WAR and 14.7 pitching WAA (versus BWAR's findngs of 11.3 pWAR and 8.6 pWAA). Spalding threw all but 25 of his team's innings (his relief pitcher, Harry Wright, was pretty much just as effective), and he went 38-8. This suggests that an average pitcher for Boston would have been barely a .500 pitcher, with a record of 23.3-22.7. Given that the team scored runs at 1.38 R/G above league average, it seems impossible that an average pitcher would have had an average record when backed by this offense.

In addition, when I took a look at Spalding's pitching numbers at Fangraphs, it calculates his RA9-WAR for 1872 at 16.0, with 11.8 of those wins being assessed as fielding dependent! It's entirely unclear to me, therefore, by what reckoning GWAR assigns 19.1 pitching WAR to Spalding. I don't see how that number can be derived from the underlying data.

The kind of adjustments you describe as applying to Caruthers' numbers in your post at #469 sound reasonable (although I am not sure they are entirely sufficient to adjust for the relative weakness of the AA), but my first look at the GWAR numbers does not give me confidence that they are reasonable at all, and if they are 50% of your system, that would seem to be leading to preposterous results for these early pitchers.

I'll add that it is notable to me that your list of results puts toward the low end all of the pitchers who peaked in the strongest league context of the pre-1893 period: 1879-83. Radbourn, Whitney, Galvin, and McCormick all did most of their best pitching during this period. It seems implausible to me that most of the best pitching was being done in the National Association Years and in the weaker American Association, with the best NL seasons also coming after the AA had begun attracting a fair amount of legit talent. I suspect quality of competition factors are skewing your numbers in ways that you haven't compensated for, especially if such issues are baked in to the GWAR numbers.

Finally, I am also wondering how you are handling "two-way" players like Ward, Caruthers, and Whitney, or more generally how your system handles players with significant numbers of both batting and pitching WAA in the pre-1893 period.

I apologize if you've been over all of this multiple times in recent years -- I was away from the Hall of Merit discussions for several years before returning around the 2020 election. I am partly trying to get up to speed on recent sabermetric developments, but I am also a bit boggled at the variance between your results and my own analyses of these players, which are radically different. I can readily understand that a peak-based system will prioritze different players from a system that balances peak and career, but I do look at players' peaks, and my findings look so different from yours that I want to try to account for the difference.

   471. kcgard2 Posted: December 24, 2020 at 09:13 AM (#5995813)
Another weird thing about the Player's League of 1890: The Buffalo Bisons had a putridly, unbearably awful pitching staff, which single-handedly made every other team's pitching staff in the entire league roughly average to well above average, except the Pittsburgh Burghers, who were still roughly 25% better than Buffalo by adjusted ERA. Buffalo as a team was below replacement level, even though their offense was competitive within that league; that's how bad their pitching was. It really is hard to overstate how bad their pitching was.

If that league simply hadn't had Buffalo in it, Silver King's bWAA goes from 10.6 to 7.1 after you run through all the calculations. Presumably, standard deviation should pick up some of this? I don't know. I agree with Chris Cobb that your system doesn't seem to be accounting for league quality strongly enough. Perhaps we simply disagree about that. Silver King may have still been the best pitcher in baseball in 1890, WAR and ERA+ even after doing any league adjustments would still be at least somewhat competitive, although IMO there's no way he's ahead of Kid Nichols or Kid Gleason, and there's 5 or 6 others who are in the conversation there. In 1888 he does have a solid argument as best pitcher even accounting for AA.
   472. kcgard2 Posted: December 24, 2020 at 09:24 AM (#5995814)
Actually, wait, why do you say that WAR figures are league-neutral? They come from exactly the same baseline as WAA figures do, at least for BBRef, and all of it is league-specific. Perhaps this is where some disconnect is happening?
   473. kcgard2 Posted: December 24, 2020 at 10:26 AM (#5995825)
Kiko (#465): It's weird to me that Campaneris or Concepcion would even be close to your ballot if you system actually downplays the value of defense. Campaneris and Concepcion accumulated only 10-20 WAR from their bats, with downplayed defensive value of 15% they don't even reach 40 WAR. Even if you dock Bell more defensively than you dock either of the shortstops, they have to make up 130 runs of offense in the case of Campaneris, and 180 runs of offense in the case of Concepcion (deltas already include baserunning value). Dock Bell 45% of his defensive value (3x) compared to their 15% and he still has 51 WAR to their 41 and 36 WAR. Shoot, Dock Bell HALF of his defensive value and don't adjust the shortstops at all, he's still ahead.

I know your system doesn't match up to WAR, but these are wildly off. You don't have Bell in your top 200 because you downplay defense. Obviously I don't agree with that but even if I grant that it's legit within your system, I just don't see how that leaves any avenue to these shortstops where 70% of their value or more came from defense.
   474. Bleed the Freak Posted: December 24, 2020 at 10:48 AM (#5995831)
Even if you dock Bell more defensively than you dock either of the shortstops, they have to make up 130 runs of offense in the case of Campaneris, and 180 runs of offense in the case of Concepcion (deltas already include baserunning value). Dock Bell 45% of his defensive value (3x) compared to their 15% and he still has 51 WAR to their 41 and 36 WAR. Shoot, Dock Bell HALF of his defensive value and don't adjust the shortstops at all, he's still ahead.

I know your system doesn't match up to WAR, but these are wildly off. You don't have Bell in your top 200 because you downplay defense. Obviously I don't agree with that but even if I grant that it's legit within your system, I just don't see how that leaves any avenue to these shortstops where 70% of their value or more came from defense.


As an aside, Dan R's work previously had Campaneris and Concepcion ahead of Bell, so having Bell lowering in and of itself isn't necessarily unusual. Campaneris actually gets rated average defensively at Kikos, though his baserunning metrics are excellent, positional value should be coming in to play versus a poor crop of fellow SS.

As to Kiko, I'll try to dig up the thread, but Bell doing poorly isn't just a defensive review, his studies indicate that power type categories are underrated versus contact weights that Baseball-Reference uses.
   475. Michael J. Binkley's anxiety closet Posted: December 24, 2020 at 11:00 AM (#5995835)
First, let me quickly address the question of WAR being league-neutral vs. WAA being league-specific. bWAR I know is league-neutral - and by that, I mean across leagues in a certain year, not across all leagues/years (e.g. it only adjusts across the NL, AA, and UA in 1884 and not across the 1882 AA compared to the 1954 NL).

Towards the bottom of this page, https://www.baseball-reference.com/about/war_explained_position.shtml , under the table-header, "Replacement level for MLB and the WAR assigned to a league by year and league," B-R lists the replacement level run values it uses for each league each year. The most extreme example of this is, of course, if you look at Fred Dunlap's 1884 in the UA. Currently B-R has his WAA for that year as 7.8 and his WAR at 7.9 (which is intriguing to me because in the past they were exactly equal - which makes sense because the average player in the UA was replacement-level, or possibly even lower) and this lines up with the zero that the UA has for replacement-level runs if the table referenced above.

(and checking now, gWAR does the same, since Dunlap's gWAA and gWAR for 1884 are both 9.5).

For more recent examples, as with the conventional wisdom, B-R has the NL as stronger than the AL in the post-integration years, presumably due to its integration at a faster rate than the junior circuit and it has the AL as the slightly stronger league in recent years.

This is why I "reverse engineer" my WAA figures from WAR rather than using the given WAA numbers on the site, since they are only relative to the league the player played in that year. Fred Dunlap certainly wouldn't be 7.8 WAA had he played in the AA in 1884, let alone the NL (I have him down at 4.5 extrpolated to 162 games WAA for 1884).



   476. DL from MN Posted: December 24, 2020 at 11:05 AM (#5995836)
shortstops where 70% of their value or more came from defense


I think you are conflating defensive value and replacement value. For Campaneris he has 6.3 bWAR from defense (63 RField) and an RPos of 128 (12.8 WAR). Bell has RField of 174 (17.4 WAR) and RPos of 44 (4.4 WAR). If you regress RField but change the replacement value of each position (RPos) that explains it.

In contrast in Dan R's standard deviation adjusted WAR with different positional replacement (zeroing out negative seasons) I see the following

Player BWAA2 BRWAA2 FWAA2 Rep2 WARP2 WAPA2
Campy -1 76 48 484 60.5 29.2
Bell 130 -22 180 275 57.1 17.7

That actually gives Bell a BIGGER advantage on defense by 2 WAR. However he gives it all back and then some in positional value (20.9 WAR difference instead of 8.4 WAR difference). That is the difference between the systems in calculations of positional replacement value. You'll also see that Campaneris provides 29 more standard adjusted wins than the average shortstop over the course of his career while Bell only provided 17.7 above his position.
   477. Michael J. Binkley's anxiety closet Posted: December 24, 2020 at 11:38 AM (#5995840)
Which leads me into some of Chris Cobb's questions.

To be quite honest, I have ended up tweaking my system every year since I started voting (and even before I started voting, believing I needed a "perfect system" before I started doing so - I was letting "perfect" be the enemy of "good."). And I'm sure I will again before I cast my ballot next year as well.

And I fully appreciate the questions. Peer-review is how we all get better at this.

Prior to this year, for pitching, I actually incorporated bWAR, gWAR, and F-G FIP-based fWAR as my basis. I finally decided to abandon the FIP fWAR since I was finally convinced by others here that it was a more predictive stat that evaluatative. I also previously used a salary-estimator based system based on the peak-friendly version that Dan Rosenheck used in his initial version of his WARP. Using the FIP fWAR served an important counter balance to the pre-1893 pitchers because otherwise, due to their huge seasonal WAR numbers compared to later-era pitchers, they all came up as over the line without fWAR being included.

Under this old system, I was more in line with consensus in having Clarkson at the top of this cohort, followed by Spalding and Keefe closer together, then Bond and Buffinton, and then the Caruthers/Radbourn/Galvin. They were all over the line. Next was Whitney (FIP didn't punish him as much as every other pitcher of the era, if at all), and then all the others. Under all previous of my systems, all peak-heavy, Bond and Buffinton have always been over the line comfortably, while Mullane/Welch/McCormick have always been outside my PHoM. A majority of the reason for this, especially this last trio, is that while they were fairly consistent relative to era, their relative peaks weren't that high.

For example, McCormick accumulated 63.7 mWAR in my system, a total, since it's over 60, many would consider Hall-worthy. However, a ton of that value is just showing up and pitching the high number of innings of the day. Now I'm not saying that that didn't have value to his teams - it most definitely did. It just didn't make him great. I have him at only 23.8 MWAA and 2.2 mWAG. There a huge difference in greatness in my opinion, between putting up 10 WAR in 600 innings and 16 WAR in 600 innings, which is what all of this era's pitchers in my PHoM did at least once or twice.

Going back to why I include gWAR in my pitching calculation. It goes back to a concept I learned about in business school regarding "wisdom of the crowds," (and I know I am a little shaky here with a sample size of n=2), but while a wsidom of crowds answer may not be the most accurate answer, it tends to be more accurate than any individual composite answer (again huge caveat with my current sample size of just bWAR and gWAR in my amalgamation). And I don't pretend to be smart enough to say without a shadow of a doubt that bWAR, or gWAR (or fWAR) is right, so I just used a combination of them all, until I eliminated the FIP fWAR this year - perhaps next year I will add back in the RA-fWAR.

As to relative league strengths across years, like Dan R's WARP, I don't apply a "league strength" adjustment, but I do apply a standard deviation adjustment to hitters/pitchers each year based upon the standard deviations of WAA/pa or WAA/ip of regular players in those leagues.
   478. Chris Cobb Posted: December 24, 2020 at 12:00 PM (#5995844)
However he gives it all back and then some in positional value (20.9 WAR instead of 8.4 WAR difference). That is the difference between the systems in calculations of positional replacement value. You'll also see that Campaneris provides 29 more standard adjusted wins than the average shortstop over the course of his career while Bell only provided 17.7 above his position.

Thank you for this clear account of how Dan R's replacement value system works. Is there an up-to-date source for his WAR, so that it could be used for assessing post-2005 seasons? I still refer to my now ancient spreadsheet with seasons through 2005, but it becomes less useful as we move farther beyond that year.

Although I consult Dan Rosenheck's work, and I used to include it in my system until it became inapplicable to too many post-2000 players, I have difficulty with this way of measuring value in this particular case to the extent that it appears that shortstop replacement value was pushed downward at this time by teams' perverse unwillingness to put players at shortstop who could actually hit. As has been discussed elsewhere, it certainly appears that Buddy Bell, for example, could have played at least an average major-league shortstop, but teams made a value-perverse decision not to put players like Bell at shortstop, which had the result of causing a severe divergence between replacement values for shortstops and third basemen. If Bell could have played shortstop but was excluded from that position by bad management decisions, then that isn't a reflection on Bell's merit as a player; it is a reflection in management's failure to realize his potential value.

It seems to me that the "third base glut" of the 1970s and 1980s came about in part because management's unreadiness to maximize that talent by spreading it between shortstop and third base resulted in a concentration of talent at third base. Concentrations of talent at positions may to some extent be a product of random variation, but they are also demonstrably produced by conditions that move certain types of players toward specific positions and enable their success at those positions. First base and catcher are obvious examples of how concentrations of talent are produced by conditions. Eras whose offensive conditions have heavily favored power hitters see large clusters of HoM first basemen; when catching became a physically brutal position to play, players with high offensive value were steered away from it, resulting in a dearth of high-value catchers. What we see in the 1960s through the Yount/Ripken era is a playing and cultural condition that concentrated talent at third base by pulling it away from shortstop. The fact that direct analysis of fielding numbers cannot find fielding value that justifies the amount of hitting that management sacrificed at shortstop to populate that position with light/quick players suggests that the replacement value of shortstops, as accurately documented by Dan R's mode of analysis, was artificially low during this period.
   479. Chris Cobb Posted: December 24, 2020 at 12:31 PM (#5995847)
Michael J. Brinkley's Anxiety Closet wrote:

Going back to why I include gWAR in my pitching calculation. It goes back to a concept I learned about in business school regarding "wisdom of the crowds," (and I know I am a little shaky here with a sample size of n=2), but while a wsidom of crowds answer may not be the most accurate answer, it tends to be more accurate than any individual composite answer (again huge caveat with my current sample size of just bWAR and gWAR in my amalgamation). And I don't pretend to be smart enough to say without a shadow of a doubt that bWAR, or gWAR (or fWAR) is right, so I just used a combination of them all, until I eliminated the FIP fWAR this year - perhaps next year I will add back in the RA-fWAR.

In general, I agree with this principle, and I've tried to follow it when it makes sense. I use both BWAR and FWAR in my evaluation of position players, but I exclude FWAR from my evaluation of pitchers, for the same reason that you have chosen to do. I don't incorporate systems into my evaluations until I understand what they do, and for me personally right now GWAR is a black box that I am starting to examine.

A challenge to applying any system to 19th-century baseball is the highly variable conditions as competition levels and game conditions changed suddenly and drastically. These make many year-to-year adjustments necessary, which the comprehensive metric or our own systems may not always have made. It looks to me like the results you are getting from GWAR without FWAR to ballast it are skewing your results in some cases, perhaps because GWAR needs adjustments for certain game conditions. My initial look at just a couple of GWAR cases suggests that in many circumstances GWAR probably provides a different but reasonable perspective on 19th-century pitchers: Bob Caruthers' 1885 in GWAR is not so different from BWAR's. If they diverge somewhat, each may be capturing some factor that the other misses. When GWAR diverges dramatically from BWAR, as in the case of Spalding's 1872 season, the magnitude of the divergence suggests that one perspective is reasonable and the other is not. In this case, FWAR's different way of looking (its results are accurate for what the metric measures, but its measures are significantly incomplete) suggests that GWAR's results must be counterfactual: there is some element in how GWAR calculates value that is thrown off by the conditions of early baseball, resulting in Spalding's GWAR being inflated beyond any possible real value. Now, I know almost nothing about how GWAR is calculated, so I can't hazard even a guess at what is happening or why or if a tweak could be inserted that would correct its extreme results. But it does look to me like it is skewing your results significantly in the case of Spalding. Depending on what the problem is, it could also be skewing the results for other pitchers.

I'd greatly appreciate any insights you or others can give on GWAR, and I will continue to look more deeply at specific pitcher seasons; I am intrigued by kcgard's observations about the PL and the impact of the Bison's pitching on that league, and I'd like to get a better sense of what is happening across the 1870s with Spalding and Bond.
   480. kcgard2 Posted: December 24, 2020 at 05:35 PM (#5995883)
The 1890 Bisons are the team version of managers refusing to play decent hitters at SS in the 70s :)

Dl thank you for the response in 476. It makes sense to me if you work backwards from a WAPA perspective, because SS was so depressed at that time and 3B was so stacked, as we have discussed many times. I always come at things from a runs produced/runs saved framework and work up from there. BTW, I was using Fangraphs numbers for offense, position, baserunning, and replacement to get the numbers I cited. For me it's conceptually easier to start from a place that says Player A produced X number of runs from batting, Player B produced Y. To close that gap, you have to find X-Y runs for Player B from somewhere else, is it even reasonable to reach that?
   481. DL from MN Posted: December 24, 2020 at 05:46 PM (#5995886)
Is there an up-to-date source for his WAR, so that it could be used for assessing post-2005 seasons?


Unfortunately no, I've been using baseball reference since then.
   482. DL from MN Posted: December 24, 2020 at 06:05 PM (#5995887)
management's unreadiness to maximize that talent by spreading it between shortstop and third base


I think that may have been a reaction to artificial turf. The difference goes away when grass infields make their comeback in the 1990s.
   483. Chris Cobb Posted: December 25, 2020 at 02:47 PM (#5995949)
I think that may have been a reaction to artificial turf.

The problem I have with this theory is that Dan R's numbers show that the drop in shortstop replacement value begins in 1960 and the decline in replacement value continues across the full decade of the 1960s to reach bottom in 1969. Then it says at about that level until the mid-1980s, which it goes on an upward trajectory, returning to the range of values that had been typical prior to 1960 by about 1995. So the decline precedes the advent of most artificial turf, and the rebound begins prior to the move away from turf. I've wondered if the change in the strike zone, which reduced the ability of non-power hitters to take walks contributed to the change.

I think it's not unlikely that the advent of artificial turf reinforced the trend, especially for teams playing their home games on turf, but it can't explain the shift in the first place.
   484. Dr. Chaleeko Posted: December 25, 2020 at 08:09 PM (#5995994)
Re 483

Forgive me if I’ve said this before and recently. I remember posting it somewhere but can’t recall where. Anyway, Chris, i think you are 100% correct. My guess is that
A) People who managed in the 1960s probably recognized that in an extremely low-run environment maximizing defense at one or more skill positions might pay off (e.g. the Cards with Maxvill and Javier or the Bucs with Maz and Alley)
B) People who started coaching or playing in the 1960s were weaned on this thinking
C) The run environment improved in 1969 but it didn’t really get back to 1950s levels until the silly ball 1990s. That’s true straight up in the NL. In the AL it got back to 1950s levels in the early 1980s, but that’s WITH a DH that the 1950s AL didn’t have.
D) As all the people in B became managers or GMs or coaches, they kept on with what they knew. That generation started to age out in the mid-late 1980s-mid 1990s and the big shortstops became commonplace in the 1990s, particularly since the next generation had watched Ripken, Yount, and Trammell (and Dickie Thon) have big seasons as shortstops. Though, Yount and Tram and Thon were not necessarily big guys, they didn’t hit like Ray Oyler and played good defense.
   485. Howie Menckel Posted: December 26, 2020 at 03:58 PM (#5996049)
(since I had an error on my ballot last year, as my penance I'll post this one here before moving it to the main board)

2021 ballot - our (and my) 124th since we began this version of the journey in 2003 (real time) with an "1898" ballot. Honored to be "The First Voter" (which is right up there with Delaware being "The First State," I suppose).

props to any other remaining "voting Ripkens" as well (I think there are a couple of others left).

I had 2020 electees on my ballot - Jeter, Tiant, AJones, Helton - at 1, 12, 3, and 11. That's more "unconventional" than usual for me. :)

Have looked through the fine 2021 Ballot Discussion, and some players moved a bit in response to the analysis.

The annual fine print:

Overall, I think there is a bit too much slavish devotion in some quarters re WAR, WAA and an ever-increasing number of acronyms, which are intriguing tools but which still may not yet be sufficiently mature (though they continue to improve).

So my fondness (but not blind allegiance by any means, especially where durability is an issue) for ERA+ and OPS+ helps, I think, as a reality check even as fielding issues are quite significant for pitchers/team defenses - and especially for hitters, of course.

I tend to be mostly prime-oriented with hitters, and prime and career with pitchers. But a huge peak sometimes catches my eye, and a remarkably long, effective hitting career also works for me. I voted for Joe Jackson on his first try, and Pete Rose, and Mark McGwire - and that pattern will continue re any new steroid/PED/other history's greatest monsters accusees.

1. JEFF KENT - Quiet start in his first six seasons - OPS+s between 101 and 111 each time, so he reaches age 30 with nary an All-Star Game selection. And then - 142-125-162 (MVP)-131-147-119-123-133-119-123 - with pretty good durability to boot in a "who saw that coming" decade of INF mashing. His defense didn't help his team win games - directly. But his offensive output crushed most of the opposition thru the prime (shades of Jeter, sort of), allowing his team either to have a turbocharged offense or to be able to carry a great glove elsewhere and still have a competitive offense. That gets overlooked.

2. LANCE BERKMAN - Fascinating battle with Crime Dog (and Helton, until he was elected last year without an 'elect-me' vote from me), and Berkman's peak is a bit peakier.

3. FRED MCGRIFF – Liked him by a nose five years ago over Palmeiro, who has a weaker peak but a longer prime. McGriff 134 OPS+ in 10174 PA to Palmeiro’s 132 in 12046 PA to Sheffield's 140 in 10947 PA - and Berkman's 144 in 7814. I really like the 157-166-153-147-166-143-157 peak from 1998-94, all in 600+ PA or equivalent. Underrated.

4. JORGE POSADA - Moves up two more slots and now up nine in two years. My team will outhit your team (ballot!) I know about the defense. I also know about OPS+s of 153-144-139-131-125 and career 121. Either the Yankees would have won 115 games a year (at least) if Posada and Jeter could field, or we may be overrating the cost of Posada's defensive shortcomings. This guy had eight straight seasons of 540+ PA, which is amazing. Like Kent, in some respects.

5. BOB JOHNSON - I like this sort of consistency over a long span, though I'd hardly say he's a 'must-elect.' Interesting discussion in 2020 thread if he got a slight career delay from native American status. Sort of the Joe Gordon of OFs in career shape, or a slightly longer and flatter version of Ralph Kiner. Or McGriff without the tail, offensively. I am concerned by 1944 being his highest OPS+; seems like he took advantage of the weak competition; but talk of PCL credit reassures me. Has more than a decade's worth of excellent hitting, for a prime that I like better than, say, many key holdovers have.

6. BOB ELLIOTT - Good to see him at least occasionally mentioned in discussions starting about 10 'years' back, at least. A lost cause - but he's my lost cause - so I have to vote as long as I believe. Six seasons of at least 134 OPS+, ALL of them as a 3B (compare: Ventura never had any that high, Bell had only one higher). Wish he'd played all 3B and not much OF, but c'est le vie - Joe Sewell seemed to get treated as a full SS by some back in the day. Beats out HOMer Boyer for me and compares remarkably well with HOMer Santo as a hitter. Better hitter than HOMer Hack as well, and better than HOMer DaEvans (see those guys' threads for details).

7. KENNY LOFTON - Back on ballot last 3 years upon reconsideration.145 OPS+ in star-crossed 1994 made him an incredibly great player, and anytime he put up 120+ it also would be pretty true. But he never did - outside of a 121. He's a very good player in all 10 of these 100 to 119 OPS+ seasons due to defense. It's difficult to say how many points he can give up there and still be a better player than a slugger, though.

8. JOHAN SANTANA - But enough about hitters. THIS is why you have to reevaluate every top 10 returnee! Geesh, 182 166 162 155 148* 131 130 129 - and two 1sts and 2 2nds in IP to boot. Shot up my ballot a bit last year, but settles for here on my third look.

9. TOMMY BRIDGES - Go 8 to 10 seasons deep, and he catches up to peakier rivals.

ERA+, full seasons
Pettitte - 177 156 132* 129 112 112 111 111 111 (111) 110 106 104 100 097 (132 of 264 IP in two half-seasons)
Hudson -165 145 138 137* 131 129 129 121 119 113 110 097 092 (137 of 278 IP in two half-seasons combined) ViWillis - 165 154 153 131 128 115 111 109 104 098 096 096 089
Walters - 170 154 146 146 141 127 127 123 120* 107 103 094 092 090 (120 of 302 IP in two half-seasons)
Buehrle - 146 144 140 130 122 121 121 121 112 112 109 108 100 099 095
Bridges - 146 144 141 140 139 139 137 133 119 118 115 111* 091

10. TIM HUDSON - Did not see that coming! Holds his own in peak and prime against predecessors and doesn't lose much on the back end, ultimately.

11. TORII HUNTER - Note the five 120+ OPS seasons, compared to Lofton.

12. ANDY PETTITTE - I didn't realize that he will have to be on my annual consideration list, but turns out he earned it. Better peak than I had realized gets him a ballot slot.

13. BOB ABREU - RF, looong career, 128 OPS+. top 20 but only one Gold Glove. Baserunning really helps, hence he'll stay on my radar.

14. PHIL RIZZUTO - I'll grant a lot of war credit and stipulate to the great, great fielding. But even 3 war credit years gets him only to 13 main years, and the fielding made him above-average overall but not excellent in most seasons. Will dance in and out of my ballot, I suspect.


15. SAMMY SOSA - Here the big, short prime is 5 yrs, and it's more obvious - because there wasn't a ton before or after. This is his case: Very durable with OPS+s in that prime of 160-151-161-203-160.


MANDATORY MENTIONS

WALLY SCHANG - He keeps bouncing on and off my ballot as well. May need others to carry his water well enough to get me back on board for good.
BUDDY BELL - One of a number of 3B guys from this era, and I prefer Sal Bando (heck, I once preferred Ron Cey). Solid all-around player and 1980-84 peak is a very strong offense-defense case. I just don't see enough beyond that, though I appreciate why he gets some love.

SAL BANDO - See Bell, Buddy. SS vs 3B discussion this year intrigued me; might get either or both on my ballot next year.

BOBBY BONDS - 8090 PA, and best OPS+ is 151. A 130 to 143 six other times, which is nice - but the strong prime is just not quite long enough for an OF.

OTHERS OF INTEREST

BUCKY WALTERS - Seemed to get Palmer-like defensive support, without enough super-stats to make that irrelevant. Proved his mettle outside of 'war years.' Bob Lemon-esque, though I wasn't a big fan there.


BEN TAYLOR - Long career, excellent fielder, consistent player. I'm not 100 pct sold on the hitting MLEs, but very good reputation and the reevaluation has made me comfortable enough to place in a ballot slot at times. It is true we're not short on players from his era.

TREVOR HOFFMAN - He dropped off my ballot in 2018 - under 1100 IP, and table constantly set for 1-inning success. A weird player no matter what your system, as are almost all of the modern closers. Consistency is a real plus, but what was he doing? Mainly allowing the Padres to avoid the part-season hiccups that rivals had when a closer spit the bit - costing those teams a couple of extra games sometimes. Rivera has the insane postseason stats; no such luck here.

BILLY WAGNER - Claim to fame is utter dominance - but of what? Instead of saving the vast majority of attempts with fewer Ks and the greater likelihood of a runner, he blew you away. But guess what, if you're down 1-2-3 runs against a well-rested good pitcher, you're usually dead either way. So longevity and consistency of Hoffman easily trumps Wagner's fewer key opportunities. Also had those postseason hiccups.

DAVE CONCEPCION - Peak is as good or better than Nellie Fox's; not quite as consistent, but a slick fielder and a very useful offensive weapon many times. Not fully buying the "other teams were stupid enough to play ciphers at the position, so give Davey bonus points" argument; that helped the Reds win pennants, but Concepcion can't get full credit for that stupidity.

DON NEWCOMBE – A passionate, detailed Newcombe backer might also get me there someday - there were some efforts on the 2020 chatter in particular. I think he had the skills, but he didn’t quite actually produce quite enough. I think.
   486. kwarren Posted: December 26, 2020 at 05:55 PM (#5996056)
From Hall of Merit Ballot:
I don’t see Santana as the Koufax-level guy that some voters do
Comparative Stats with Koufax listed 1st

IP 2,324 2,026
ERA+ 131 136
bWAR 48.9 51.7
JAWS 47.4 48.3
WAR7 46.0 45.0
WAR/250 5.26 6.38

ERA 2.76 3.20
W-L 165-87 139-78


bWAR (high to low)

Koufax 10.7, 10.3, 8.1, 7.3, 5.7, 4.4, 2.2, 1.5, 1.3, .....48.9

Santana 8.7, 7.6, 7.2, 7.1, 5.0, 4.7, 4.1, 3.3, 2.7, .....51.7

CYA

Koufax 1, 1, 1, 3
Santana 1, 1, 3, 3, 5, 7

In summary it looks like Santana had slightly the better career, but Koufax at his best was better than Santana at his best by a significant margin. Had Santana pitched in the same environment as Koufax he would likely have had an ERA between 2.60 & 2.65 and likely an estimated win loss record of 173 - 79.

However, he would not have been able to replicate Koufax's 1963 & 1966 seasons:

19-5 1.74 & 27-9 1.73.

An astounding achievement (even allowing for the favourable influence of Dodger Stadium) for Koufax was leading the NL in ERA for five consecutive seasons.

Santana had 3 ERA titles, and also had ERA's of 2.99, 3.07, 2.87, 2.98 when he didn't win. This is also amazing given that he was pitching in the Homerdome. !!

These two guys were both truly dominant and are very deserving to be in the Hall of Merit. The way Santana was treated by the BBWAA is rather sad, but typical.

Koufax was a national legend at the time that he pitched because of high K's, the unbelievable ERA stats, the win-loss record, the Dodger name on his uniform, the post season success and the Yankee-Dodger rivalry.

And Santana with almost the identical regular season career is an after thought for most. Career too short, pitched for the Twins in a hitters paradise and the excellence is easily lost. Too bad !!
   487. kcgard2 Posted: December 26, 2020 at 07:14 PM (#5996072)
Hey Howie, my main observation about your ballot is that it seems very lenient on guys having terrible defense. A few guys make the back of you ballot for some defense, but nobody seems dinged at all for being negative defenders (either by value or within position). Also there are 3 other voting Ripkens with you.
   488. Jaack Posted: December 26, 2020 at 11:52 PM (#5996094)
Comparative Stats with Koufax listed 1st

IP 2,324 2,026
ERA+ 131 136
bWAR 48.9 51.7
JAWS 47.4 48.3
WAR7 46.0 45.0
WAR/250 5.26 6.38

ERA 2.76 3.20
W-L 165-87 139-78


bWAR (high to low)

Koufax 10.7, 10.3, 8.1, 7.3, 5.7, 4.4, 2.2, 1.5, 1.3, .....48.9

Santana 8.7, 7.6, 7.2, 7.1, 5.0, 4.7, 4.1, 3.3, 2.7, .....51.7

CYA

Koufax 1, 1, 1, 3
Santana 1, 1, 3, 3, 5, 7

In summary it looks like Santana had slightly the better career, but Koufax at his best was better than Santana at his best by a significant margin. Had Santana pitched in the same environment as Koufax he would likely have had an ERA between 2.60 & 2.65 and likely an estimated win loss record of 173 - 79.

However, he would not have been able to replicate Koufax's 1963 & 1966 seasons:

19-5 1.74 & 27-9 1.73.

An astounding achievement (even allowing for the favourable influence of Dodger Stadium) for Koufax was leading the NL in ERA for five consecutive seasons.

Santana had 3 ERA titles, and also had ERA's of 2.99, 3.07, 2.87, 2.98 when he didn't win. This is also amazing given that he was pitching in the Homerdome. !!


By my system, Koufax's peak outshines Santana by a pretty substantial margin. For their respective top three seasons, Santana is comparable on a rate-basis, but Koufax has substantially more volume in his peak seasons, which makes a pretty big difference. I have Koufax's top three seasons nearly at the top of the list for pitchers - up with Pedro and Walter Johnson. Santana has an impressive peak too, but it's really not astounding compared to other HoM pitchers for me - his top three seasons are around the same level as Roy Halladay's top three. That sort of peak is impressive, but it really isn't comparable to Koufax.

Now after those top three seasons, both guys have one more really nice ace type season. Then Koufax has two All-Star type seasons and a few average/filler ones, while Santana has three All-Star level seasons and two above-averagish ones. I'd say that does lean to Santana, but not by a whole lot. Adding in postseason puts a sizeable gap between them for me at least. Koufax is a solid, middle of the road HoMer for me, while Santana is around the borderline. He doesn't do well enough by my system to reach my ballot this year but I have no problem with his election.

These two guys were both truly dominant and are very deserving to be in the Hall of Merit. The way Santana was treated by the BBWAA is rather sad, but typical.

Koufax was a national legend at the time that he pitched because of high K's, the unbelievable ERA stats, the win-loss record, the Dodger name on his uniform, the post season success and the Yankee-Dodger rivalry.

And Santana with almost the identical regular season career is an after thought for most. Career too short, pitched for the Twins in a hitters paradise and the excellence is easily lost. Too bad !!


I'm not sure who this is intended for - I don't think any voter here thinks of Santana as an after thought. He's got a pretty good shot at going in this year, and is surely going to go in by 2023.
   489. Dr. Chaleeko Posted: December 27, 2020 at 01:40 PM (#5996127)
Just for giggles, and because I particularly like the topic of Santana v. Koufax...

KOUFAX
Ebbets field: 98 IP, 4.04 ERA, 91 H, 97 K, 44 BB, 14 HR
LA Colesium: 344.2 IP, 4.33 ERA, 299 H, 367 K, 188 BB, 56 HR
Dodger Stdm: 715.1 IP, 1.37 ERA, 446 H, 754 K, 142 BB, 34 HR

Koufax Home 1962-1966: 715.1 IP, 1.37 ERA, 446 H, 754 K, 142 BB, 34 HR
Koufax Road 1962-1966: 661.2 IP, 2.57 ERA, 513 H, 690 K, 174 BB, 54 HR

So what happens when you take a guy who:
-walks too many hitters
-plays in tough HR parks
-has good strikeout numbers
-has strong offensive support behind him,

and you
-put him into a pitcher's paradise
-redefine the strike zone so that's it bigger.

You get Sandy Koufax's record at Dodger Stadium. This is no knock on Koufax. He threw 661 innings of 2.57 ERA baseball away from Chavez Ravine in a league that averaged around a 3.50ish ERA in those five years, so, of course, he was a fine pitcher. But the combination of DS's high mound, big foul ground, and general offensive-suppression qualities combined with the new strike zone allowed him to throw more strikes, come at hitters from a more downhill plane that made his breaking pitches that much tougher, not have to worry about hard contact because of his park, and throw more innings at home game because he could be more efficient at home (up 1-2 IP/G overall over pre-DS Koufax). His overall BABIP went down, his overall rate of extra-base hits per hit went down.

Koufax's talent level was very high. We know that. But the manifestation of that talent at the statistical level is extremely context-sensitive. More than most players, IMO. It's not that we wouldn't see him as a great pitcher in another park, it's that we wouldn't hear people calling him the greatest lefty ever or the greatest pitcher ever. If Johan Santana had pitched with the exact same advantages Koufax had, he would likely have led the league in innings or strikeouts more often. Or thrown more no-hitters? Or had microscopic ERAs. That to me seems like a reasonable imaginary outcome.

But as Jaack said, no one is not taking Santana seriously, so this is just one of my pet topics. :)
   490. Howie Menckel Posted: December 27, 2020 at 02:42 PM (#5996136)
Hey Howie, my main observation about your ballot is that it seems very lenient on guys having terrible defense. A few guys make the back of you ballot for some defense, but nobody seems dinged at all for being negative defenders (either by value or within position).


this year, anyway. the 3B/SS talk, the fact that we elected AJones last year but we are still trying to reconcile his D more precisely, and other issues had me in an "offensive" mood. my point stands re Posada and Jeter - "Either the Yankees would have won 115 games a year (at least) if Posada and Jeter could field, or we may be overrating the cost of Posada's [or Jeter's?] defensive shortcomings."

well, mild hyperbole there. but if they each only move the needle by 2 G each annually.....

"Also there are 3 other voting Ripkens with you" - love it!
   491. rwargo Posted: December 27, 2020 at 06:19 PM (#5996157)
Beacuse it seems to track our elections, I'm just using bbref WAA or Seamheads in the case of Negro Leaguers. Early 19th century pitchers get a discount while hitters get a premium. I set negative seasonal numbers to zero. One weird thing I do is take the greater of the pure batting or pitching numbers vs. total numbers. (i.e. Dick Stuart would use his batting WAA not total.). All time rating is MLB only or Negro League only.

The numbers in parentheses are the Best WAA (total, batting, or pitching) for each player.

1. CF: Kenny Lofton (39.5) - #12 all time CF. Higher WAA than Edmonds, Dawson, Wynn, Ashburn, Doby, Browning, Averill, Cary, Roush, Hines, Gore, and Pike.

2. 3B: Buddy Bell (36.1) - #15 all time 3B. Higher WAA than Boyer, Darrell Evans, Groh, and Hack, plus 19th century guys.

3. RF: Sammy Sosa (36.0) - #16 all time RF. Higher WAA than Winfield, Dwight Evans, Guerrero, Flick, Kelly, Keeler, Slaughter, and Thompson.

4. 3B: Sal Bando (35.5) - #16 all time 3B. Higher WAA than Boyer, Darrell Evans, Groh, and Hack, plus 19th century guys.

5. SP: Vic Willis (34.9) - #54 all time SP. Higher WAA than 20 elected pitchers. 12 contemporaries (debut between 1896-1920) have been elected, Willis is higher than M. Brown, Waddell, McGinnity, Faber, and Rixey.

6. SP: Johan Santana (34.1) - #59 all time SP. Higher WAA than 17 elected pitchers. behind only Halladay, Kershaw, Greinke, Verlander, Scherzer and Hamels among those debuting beetween 1996-2020 (for now).

7. C: Ernie Lombardi (26.6) - #16 all time C. Higher WAA than Torre (adjusted), Freehan and Campanella, plus 19th century guys not named Ewing.

8. C: Wally Schang (26.0) - #18 all time C. Higher WAA than Torre (adjusted), Freehan and Campanella, plus 19th century guys.

9. RF: Bobby Bonds (33.9) - #19 all time RF. Higher WAA than Guerrero, Flick, Kelly, Keller, Slaughter, and Thompson.

10. SP: Tommy Bond (39.2) - #36 all time SP. Higher WAA than 19th century pitchers Caruthers, Galvin, Radbourn, Griffith, Rusie, and Spalding.

11. SP: Babe Adams (33.8) #60 all time SP. Higher than contemporaries Faber and Rixey.

12. SP: Jim McCormick (38.8) - #38 all time SP. Higher than 19th century pitchers Caruthers, Galvin, Radbourn, Griffith, Rusie, and Spalding.

13. SP: Roy Oswalt (32.9) #66 all time SP. Higher than contemporaries Hudson and Buehrle, below Halladay, Santana, and the current crop of Kershaw, Greinke, Verlander, Scherzer and Hamels.

14. SP: Kevin Appier (32.8) #69 all time SP. All 14 contemporaries (debut between 1971-1995) above him have been elected and none below him. He is the next best available pitcher from 1971-1995.

15. C: Gene Tenace (29.3) #12 all time C, but not 100% C. Higher than Simmons, Bresnahan, Bennett, White, Freehan, Campanella. Higher than Mauer, who might sail in.

Required Comments

2B: Jeff Kent (29.8) - #20 all time 2B. Higher WAA than Doerr, Herman, McPhee, Fox, Childs, Richardson, Barnes.
LF: Lance Berkman (32.6) - #13 all time LF. Closer to the ballot than I thought. Surprisingly Higher WAA than 14 players elected to the HOM, specifically Magee, Goslin, B. Williams, Burkett, Minoso, Medwick, Wheat, Keller, Kiner, Storvey, Kelley, O'Rourke, Sheckard, and C. Jones. Would be a fine selection.
RF: Bobby Abreu (31.9) - #22 all time RF. Higher WAA than Flick, Kelly, Slaughter, and Thompson. I actually think Giles might be a better selection, definitely Sosa and Bonds over Abreu.

Newbies

SP: Tim Hudson (31.4) - #72 all time SP. Higher WAA than McGinnity, Ford, Koufax, Ruffing, Faber, Pierce, Lemon, Sutton, Rixey. He will be in my consideration set on future ballots.
SP: Mark Buehrle (27.9) - #99 all time SP. Case is similar to Andy Pettitte (28.0), both fall just outside my consideration set unless we elect a lot above him.

Other MLB players in my consideration set (top 20H/80P):

SP: Charlie Buffinton (36.1) - #47 all time SP. Higher than 19th century pitchers Radbourn, Griffith, Rusie and Spalding.
SP: Urban Shocker (31.2) - #74 all time SP. Tied with McGinnity, higher than Ruffing, Faber, Pierce, Lemon, Sutton, Rixey.
SP: Dwight Gooden (30.8) - #78 all time SP. Higher than Ruffing, Faber, Pierce, Lemon, Sutton, Rixey.
SP: Orel Hershiser (30.6) - #79 all time SP. Higher than Ruffing, Faber, Pierce, Lemon, Sutton, Rixey.
C: Thurman Munson (25.5) - #19 all time C. Higher WAA than Bennett, White, Freehan and Campanella.
3B: Toby Harrah (32.6) #18 all time 3B. Higher than Darrell Evans, Groh, and Hack.
3B: Ned Williamson (20.2) - #42 all time SP. Higher than Ezra Sutton.
SS: Joe Tinker (30.5) - #19 all time SS. Higher than Jennings, Sewell, Ward, Wright, Pearce.
LF: Bob Johnson (30.9) #18 all time LF. Higher than Minoso, Medwick, Wheat, Keller, Kiner, Stovey, Kelley, O'Rourke, Sheckard, and Jones.
CF: Chet Lemon (33.4) #17 all time CF. Higher than Doby, Browning, Averill, Carey, Roush, Hines, Gore, and Pike.
CF: Bernie Williams (33.0) #18 all time CF. Higher than Doby, Browning, Averill, Carey, Roush, Hines, Gore, and Pike.
CF: Cesar Cedeno (31.1) #20 all time CF. Higher than Browning, Averill, Carey, Roush, Hines, Gore, and Pike.

Negro Leaguers over 16 WAA (through addition of 1926 data to NLDB)

3B: Carlos Moran (23.2) - #2 all time 3B, ahead of Beckwith, behind only Jud Wilson
SP: William Bell (21.9) - #7 all time SP, ahead of Dick Redding and Rube Foster
SP: Juan Padron (21.0) - #8 all time SP, ahead of Redding and R. Foster
SP: Bill Byrd (20.9) - #9 all time SP, ahead of Redding and R. Foster Foster
1B: Ben Taylor (19.2) - #3 all time first base in the Negro Leagues.
SP: Nip Winters (18.8) - #10 all time SP, ahead of Redding and R. Foster
RF: Hurley McNair (17.7) - #2 all time RF, behind only Dihigo. Just ahead of Heavy Johnson, who has much less PA
RF: Heavy Johnson (16.7) - #3 all time RF, and a high rate comparable to Mule Suttles and John Beckwith
SP: Hilton Smith (16.3) - #12 all time SP. 400 IP would bring him in line with the Bell-Padron-Byrd-Winters group, but Smith's rate would put him at the top of that cluster.
SP: Carlos Royer (16.2) - #13 all time SP. Lower rate, more data may not help this deadball pitcher.


   492. Eric J can SABER all he wants to Posted: December 27, 2020 at 08:07 PM (#5996170)
14. SP: Kevin Appier (32.8) #69 all time SP. All 14 contemporaries (debut between 1971-1995) above him have been elected and none below him. He is the next best available pitcher from 1971-1995.

This is kind of an arbitrary way to assign contemporaries? For instance, you would have Appier as a contemporary of Steve Rogers, even though Rogers retired four years before Appier's debut and they pitched in completely different contexts (different leagues, different scoring environments). Rogers would be more of a contemporary of Vida Blue, but Blue debuted in 1969 and would be considered part of the next group back.

Willis and Rixey/Faber is similar - Willis retired in 1910; Rixey debuted in 1912, Faber in 1914. Rixey and Faber pitched large portions of their careers in the early live-ball era; Willis debuted a few years before foul balls were strikes.
   493. DL from MN Posted: December 28, 2020 at 08:41 AM (#5996197)
I'm just using bbref WAA


How do you account for below average seasons? Do they not count at all or do they subtract from the WAA of the player?

   494. kwarren Posted: December 29, 2020 at 12:56 AM (#5996333)
By my system, Koufax's peak outshines Santana by a pretty substantial margin. For their respective top three seasons, Santana is comparable on a rate-basis, but Koufax has substantially more volume in his peak seasons, which makes a pretty big difference. I have Koufax's top three seasons nearly at the top of the list for pitchers - up with Pedro and Walter Johnson. Santana has an impressive peak too, but it's really not astounding compared to other HoM pitchers for me - his top three seasons are around the same level as Roy Halladay's top three. That sort of peak is impressive, but it really isn't comparable to Koufax.

Best three seasons ERA+ & (Career ERA+)

Pedro - 291, 243, 219, (154)
Johnson - 259, 243, 215, (147)

Koufax - 190, 186, 160, (131)

Santana - 182, 166, 162, (136)
Halladay - 185, 167, 163, (131)

I know that "best three seasons" was cherry picked because that put Koufax in the most favourable light, but he is nowhere near Johnson & Pedro even by that measure. He is much more aligned with Santana & Halladay.
   495. Howie Menckel Posted: December 29, 2020 at 01:05 AM (#5996336)
Pedro was 4th, 8th, and 7th in IP in his 3 monster seasons.

Koufax led his league twice in those seasons.

I'm not a big Koufax backer, but a little birdie tells me that if Pedro had more than 12 total CG in his top two seasons, his ERA might have slipped a tad.

Koufax had 59 in his three monster seasons.
   496. progrockfan Posted: December 29, 2020 at 08:00 AM (#5996346)
There's another major difference between Santana and Koufax aside from CG, and that's post-season performance. Some of y'all may place no weight on this, which is a legitimate approach; I do so when the data is available, and in the cases of Johan and Sandy we have rather a lot of evidence.

Koufax was brilliant in all four of his World Series appearances, helping his team win three rings in four tries and taking home two Series MVPs:
57 IP, 0.95 ERA, 0.825 WHIP, 5.55 K/W

Santana was brilliant in one postseason series, good in another, and pretty awful in three others, with his team losing four of five:
34 IP, 3.97 ERA, 1.324 WHIP, 3.20 K/W

It's also worthwhile noting that while a big chunk of Koufax's regular-season dominanace from 1961-66 was courtesy of his park, Sandy was lethal in Series play both on and off the road.

This, to me, is a big separator between the two. That's not to say Santana isn't worthy - he is after all pretty high up on my ballot - but I read Sandy as the greater pitcher. Irrespective of popular punditry, neither (imo) rank with the true Inner Circle greats.
   497. Bleed the Freak Posted: December 29, 2020 at 10:27 AM (#5996366)
What's everyone's thoughts on Gavvy Cravath?

DL has him at #5, not seeing anyone else taking note of him.

Docs MLEs were very impressive, just making sure myself or others don't overlook him, happy holidays!
   498. Jaack Posted: December 29, 2020 at 10:54 AM (#5996373)
Best three seasons ERA+ & (Career ERA+)

Pedro - 291, 243, 219, (154)
Johnson - 259, 243, 215, (147)

Koufax - 190, 186, 160, (131)

Santana - 182, 166, 162, (136)
Halladay - 185, 167, 163, (131)

I know that "best three seasons" was cherry picked because that put Koufax in the most favourable light, but he is nowhere near Johnson & Pedro even by that measure. He is much more aligned with Santana & Halladay.


As I said before, Koufax's advantage comes from his volume in those peak seasons, not his rate. Santana and Pedro were pitching 70-75% of the innings that Koufax was per season.

And I dont know what your getting at with the cherry-picking thing. I'm not trying to make the case for Koufax - he's already in the Hall of Merit. I'm just trying to establish why I don't see Santana as highly as most other voters. He's a peak candidate, and I don't see his peak as an earth shattering type that merits immediate election. It wouldn't be out of place among Hall of Merit pitchers - I think Halladay and Jim Bunning had similar peaks. At the same time, so did Sam McDowell. Overall, Santana is right on the borderline for me and behind enough guys that he doesn't make my ballot.
   499. bachslunch Posted: December 29, 2020 at 11:55 AM (#5996388)
Short career guys like Cravath (and Santana) are tough for me to figure out, given how strongly I value WAR. I have the latter 31st and the former 32nd, which I may revisit. I especially like Cravath's 151 OPS+, which may be the highest for someone with enough of a career for HoM consideration who isn't in, and that's primarily why he's as high up as he is for me.

He gets minor league credit from those who support him. I don't give it, myself, but it could help his case.

The fact that he took advantage of a very hitter-friendly park in the Baker Bowl is likely a demerit for some voters. I'm unsure on this point as well. There's much to be said for taking advantage of what your home field gives you (Babe Ruth, Roger Maris, and other left handed hitters re Yankee stadium, Wade Boggs going the other way to take advantage of the Green Monster). There's also the thinking that the advantage might not be fair to credit, that adjustments need to be made, and that such alterations diminish his case.
   500. kwarren Posted: December 29, 2020 at 01:23 PM (#5996401)
What's everyone's thoughts on Gavvy Cravath?

DL has him at #5, not seeing anyone else taking note of him.


Using JAWS he is not even close, down at the Harold Baines & Ross Youngs level, but offensively he is clearly well qualified.

Here are some other relevant right fielders for comparative purposes with OPS+, WAR/600 PA, PA, & dWar shown:

Gavvy Cravath - 151, 4.25, 4,663, (9.5)
Dwight Evans - 127, 3.81, 10,569, (3.8)
Reggie Smith - 137, 4.81, 8,051, 3.0
Sammy Sosa - 128, 3.55, 9,896, (0.3)
Dave Winfield - 130, 3.12, 12,358, (22.7)
Bobby Abreu - 128, 3.58, 10,081, (10.9)
Vlad Guerrero - 140, 3.94, 9,059, (10.0)
Bobby Bonds - 129, 4.30, 8,090, (1.8)
Gary Sheffield - 140, 3.32, 10,947, (27.7)
Harold Baines - 121, 2.09, 11,092, (19.5)

This issue with Cravath is clearly the length of his career. For the time he played he was excellent.

Interesting to note how great Reggie Smith, and how Winfield's long career hurts him in WAR/PA and his dWAR. He is probably not as well qualified and we tend to assume.

Page 5 of 7 pages ‹ First  < 3 4 5 6 7 > 

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
cardsfanboy
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 1.3677 seconds
41 querie(s) executed