hot topics

From my site:

Allie Reynolds

Hal Newhouser

The main thing for these guys is I included a year-by-year detailed account of their careers in the notes sections.

I was surprised just how poor a candidate Reynolds came off like. Please check it out anyway though.

Newhouser's peak was otherworldly.

I was shocked when Allie Reynolds actually made the VC's list of 25 candidates last time around. Actually I wasn't shocked when he made the list, I was doubly shocked when I looked up his numbers. First at how bad they were, and secondly then I was shocked that he was on the ballot.

I don't see Phil Cavaretta as top 50--maybe top 75--more likely barely top 100, but the guy had a serious career. Hats off to Phil.

Allie Reyolds was 7-2 over the course of 6 World Series, in which his team went 6-0, and in all of which he won at least one game. By my rough calculations, his ERA+ in these World Series was about 156. And to top it off, his postseason batting average was over .300!

No voters here, that I can tell, give excessive weight to postseason performance. But it is certainly a reasonable measure of "merit" (and certainly a reasonable measure of "fame"). I think if you want to give voters a wide range of choices -- but not "everyone", as we do -- then you need to include every possible definition of HoFer, including "greatest postseason."

I don't see Phil Cavaretta as top 50--maybe top 75--more likely barely top 100, but the guy had a serious career. Hats off to Phil.

He got a lot of playing time at ages 17 & 18 so he shows up on a lot of the early age-based leader lists.

His 22-year career makes him appear on quite a few of those "Kevin Bacon" lists linking old timers and modern players.

Much of his peak was between 43-45, but like Newhouser he still played quite well in 46. Not enough bat for an OF/1B candidate to get many votes here, though.

No voters here, that I can tell, give excessive weight to postseason performance. But it is certainly a reasonable measure of "merit" (and certainly a reasonable measure of "fame"). I think if you want to give voters a wide range of choices -- but not "everyone", as we do -- then you need to include every possible definition of HoFer, including "greatest postseason."

Yeah, can't be too hard on the HoF yet because I don't think he's getting a whole lot of support yet. He's nowhere near the Santo/Hodges/Oliva level of "popular" opinion as yet.

Reynolds was a halfway decent strikeout pitcher for his day but you have to check the leaderboards to see that as his was definitely not a strikeout era. Not exactly a control artist, he did his part to contribute to the post-war walk boom and no doubt took advice of great middle infielders wherever he played. I group him in with Lew Burdette as a decent (but not great) pitcher who had great run support and had a good World Series or too.

How is the vets committtee viewing Burdette?

though I'm not giving him such high credit for ww2 buu if Pesky would have had seasons like he did before and after the war I would have him high on my ballot

Prelim.

PHoM: Cool Papa Bell & Joe Medwick

1. Red Ruffing
2. Jake Beckley
3. Eppa Rixey
4. George Van Haltren
5. Mickey Welch
6. Willard Brown
7. Cool Papa Bell
8. Joe Medwick
9. Biz Mackey
10. George Sisler
11. Tommy Leach
12. Dobie Moore
13. Edd Roush
14. Sam Rice
15. Quincy Trouppe

16-20. Ryan, Averill, Powell, Childs, Elliott
21-25. Duffy, H.Smith, Streeter, White, Griffith
26-30. Jennings, Mullane, Sewell, Monroe, Strong

Newhouser is 51st.

For those that care, after my low Paige vote last year, he is up to 4th on my PHoM 'voting'.

This is in reply to Chris Cobb's comment on the Leon Day thread, re. NeL pitchers. My current ranking was posted on the Re-evaluting NeL Pitchers thread;

"Mendez--about #8-10 and in my PHoM
Redding--around #25-30
H. Smith--around #30s
Byrd--#40s
A. Cooper--#50s
Leon Day--#50s-60s but subject to change
Winters--#60s-70s

Brewer, Bell, Holland--#85-100
Trent, Madlock, Donaldson--bubbling below the hot 100 for now"

The good Doctor Chaleeko then suggested that Trent, Madlock and Donaldson probably ought to rate ahead of Brewer, Bell and Holland, especially for anybody (like me) inclined toward peak and prime rather than pure career. And he is right. So move the latter three up ahead of the next to last threesome and I can live with that list for now.

FTR here is Chris' list from the Leon Day thread:

>Mendez
Redding
Byrd
Matlock
Day
Smith

>Andy Cooper might deserve to appear among this group, as might Ted Trent.

Between the two of us and the 1959 results we have:

Mendez 1-1-2 (24)
Redding 2-2-1 (23)
Byrd 4-3-x (rec'd votes last in 1955)
Matlock 9-4-3 (64)
H. Smith 3-6-x (coulda sworn he got a few votes, can't find 'em)
Donaldson 10-x-x (rec'd votes last in 1934)
A, Cooper 5-7-x
Day 6-5-new
Winters 7-x-x
Trent 8-8-x

Bud Fowler last rec'd votes in 1909
George Stovey 1906

1960 Prelim

NAME
1. Duffy
2. Oms
3. Mendez
4. Medwick
5. Cravath
6. Matlock
7. Walters (pHOM)
8. Newhouser (pHOM)
9. Ferrell
10. Burns
11. Browning
12a (Hack) projected pHOM 1966
12b (Herman) projected pHOM 1967
12c (Lyons) projected pHOM 1968
12. Willard Brown
13. Trouppe
14. Ruffing
15. Jennings
------------------------------
16. Averill
17. Rixey
18. GVH
19. Moore
20. Roush
YOUSE NEW GUYS
-Cavaretta: Long career, short peak, well off my consideration set.

-Pesky: Short career, foreshortened career, gets WW2 credit, but still off the edge of my consideration set. Just ahead of Gordon and Sewell, a bit behind Doerr.

-Allie Reynolds: Thanks Chris J! Pass.

-Connie Marrero: Was there a rest of his career somewhere south of the border?

-Leon Day: I didn't like Dean; I didn't like H. Smith; I don't much like Day.

-Silvio Garcia: The nice pitching helps his case, but ultimately, he's a SS with an average glove and an iffy bat. He'd need to have a lot more pitching to get onto my ballot or onto my consideration set.

Chris Cobb sometime recently brought up "sleeper" NgL candidates, and I thought I'd alert everyone to another. Marvin Williams.

Williams appears to have been a high-average, power-hitting 2B/OF type. He basically divided a fairly lengthy career among the NgL, MxL, and MiL (in that order), and seems to have hit wherever he went.

I don't yet have any idea what his translations will look like, and he's not on the docket for a few years, but I wanted to let everyone know about him in advance.

There are also three white "stealth" candidates coming up in the next two or three elections that are worth noting and whose cases might require longer than a one election cycle's worth of discussion/research.

-In 1961 we encounter Ferris Fain. If memory serves, Fain was a PCL guy whose price was high. I don't know much about him, but I wonder how much, if any, MLE credit he might be eligible for.

-In 1962 comes Al Rosen. Rosen was blocked by Keltner and retired early due to back trouble. What I'd also like to know is whether the war played any role in his delayed MLB debut.

Lastly, and most thorny
-1962 also brings us Ellis Kinder. Kinder appears to have been Eck v 1.0. He was a pretty good starter with the Sox who was converted to a fireman. At which point, he went all nutty and started posting very nice ERA+s and gathering lots of saves. He's probably the first serious relief candidate we've encountered since Marbury, and RPs are, after NgLers, probably the most difficult players we'll assess.

1960 Prelim

1. Hughie Jennings (3 lat year-1-2, PHoM 1927)
2. Dobie Moore (4-2-3, PHoM 1942)

Great peak shortstops back on top.

3. Joe Medwick (5-3-4, PHoM 1954)
4. George Sisler (6-4-5, PHoM 1938)

Great peak hitters next.

5. Tommy Bond (7-5-6, PHoM 1929)

Great peak pitcher next.

5a. Lou Boudreau (e-9-11, prelim PHoM 1960 subject to some more thought)

Another shortstop with not as consistently high a peak as the top 2.

6. Rube Waddell (8-6-7, PHoM 1932)
7. Jose Mendez (9-7-8, PHoM 1957)

More great peak pitchers.

7a. John Beckwith (na-e-10, prelim PHoM 1960 subject to some more thought)

Or was he too much of a pain in the du-pa?

8. Addie Joss (10-11-12)
9. Ed Williamson (11-12-13, PHoM 1924)
10. Willard Brown (12-10-x)
11. Hal Newhouser (new)

All (except Williamson of course) will also get PHoM consideration this year. Really gotta crunch Joss vs. Newhouser. Superficially comparable careers--i.e. a little short, high peak. Joss more consistent, Newhouser the higher peak and more of a workhorse. Also interesting to compare to Waddell.

12. Pete Browning (13-13-15)
13. Charley Jones (14-15-x, PHoM 1921)

Remaining great 19C hitters.

14. Joe Gordon (15-x-14)
14a. Stan Hack (e-14-x)

Will get their PhoM consideration some day.

15. Dick Redding (last on my ballot in 1939)

Re-eval of NeL and also of ML pitchers brings the Cannonball all the way back from #46 last year. Watchthe pitchers shuffling around from#16-50.

16-20. Doyle, Cravath, Coimbre (new), Doerr, Duffy
21-25. Trouppe, (Stovey), Rixey, Averill, Cicotte, Childs
26-30. Dean, Oms, Monroe, H. Smith, Sewell, Griffith
31-35. Roush, McCormick, Bell, Byrd, Elliott
36-40. Bresnahan, Matlock, Traynor, (Keeler), Mullane, Gomez
41-45. A. Cooper, (Faber), Lundy, Mackey, Ruffing, Bancroft
46-50. Keller, Wilson, Dunlap, (Sheckard), Day, Welch

prelim: my spots 1-4 are all pitchers

Newhouser Griffith Ferrell Walters, with Ruffing & Rixey on the ballot as well.

me rasslin agin meself:

I'm having a problem. Just when I thot the NeL players would be sorting themselves out, we have been getting info on more of the 'lesser known' ones: W Brown, A Oms, L Easter.

As I attempt to integrate their perceived value with others, I find these guys fighting two different battles.

First, their MLB equivalent stats, from our best guesses, against other MLBers.

Let's say my perception of the 3 mentioned above puts them all around #10-15 on my ballot.

Now more 'real' MLBers get pushed off. Am I OK with this? Do I really believe that 1/3rd of my ballot ought to be 'banned from MLB' players, given how many we've already elected? I start wrestling against quotas, etc.

The other problem is that some NeLers on my ballot are there partly through reputation. Biz Mackey and Cool Papa Bell would make my top 30 based on their stats, but I have elevated them due to their perceived value as given by NeL experts.

Well, if the experts had the data that we have about earlier NeLers, about Cuban players, etc., would Cool Papa still be held in as high esteem?

So, as I bring Brown/Oms/Easter up, there is a tendency for Bell/Mackey to drop.

And then I feel again I am quota-izing.

But if I don't drop any of them, more pitchers are getting shut out, and I really do feel like our Hall is short of hurlers. Especially NeL hurlers; we are doing pretty well at honoring NeL hitters.

I guess this is a just a long way of saying this whole exercise is hard.

Good thing it's fun :)

My preliminary ballot is almost 1/2 pitchers.

1. Newhouser -- Others have attributed a Jenningsesque peak to Dean. As I see it, Newhouser has that kind of peak, not Dean. A war-time discount drops him some, but he was great 1946-48 also, and his career IP aren't bad for his era. He's wouldn't be among the stronger first-ballot electees ever, but he's the best player eligible.
2. Griffith
3. Jennings
4. Rixey
5. Ferrell
6. Oms
7. Brown
8. Ruffing
9. Mackey
10. Roush
11. Sisler
12. Cravath
13. Grimes
14. Newsom
15. Gordon

By my positional count, we have 114 electees divided as follows:

8 Catchers

12 first basemen

10.5 second basemen
14.5 shortstops
7.5 third basemen

12 left fielders
12 center fielders
9.5 right fielders

28 pitchers

By positional quotas of 8 starting position players per 3 three pitchers to make a team of starters, we should have elected 31 pitchers to this point. Taken together our 2B/3B/SS total of 31.5 is about right, our 33.5 total of outfielders is a bit high, our first base total of 12 is a bit high, and our catcher total of 8 is a bit low.

I think it's time to elect some pitchers, though I disagree with Tom H. about Negro-League hurlers being especially underrepresented. 6 of the 21 NeL electees have been pitchers, which is a higher ratio (28.6%) than for major-league players, (26.5%).

I want to start by asking how many HOM voters currently live in or around New York? I just moved here to get my master's and was thinking of actually seeing someof you guys.

Second, I am having trouble seperating how much of Trouppe's value is 3B and howmuchof it is catcher. Some of this is a flaw in my system, where I lower the bar for 'average' and 'peak' seasons for catcher because they play fewer games. However, if a guy was playing 60% 3B and 40% catcher in a year it is tough to work out. Is there an MLE estimate for Trouppe only as a catcher instead of as a C/3B?

I also want to state that I really like Newhouser, I saw someone had him at #51 on their perlim. He reminds me a lot of Dazzy Vance, especially after the war discount. Right now he slots in a #2 behind Jennings for me.

We should have a fun two years as I forsee two backlog candidates making the grade. Let teh Ruffing/Jennings debate begin!

It is great to see Newhouser getting serious consideration. I was always a fan of his--I thought, his biggest fan. Now I see I have him at #11 on my prelim and Chris has him #1. Not his biggest fan anymore.

But in order for him to be #51 you've got to have a serious career fixation and a big WWII discount. I have more of a peak fixation, but also about a 10 percent discount for 1944-45. But Hal was still dominant 1946-48, as Chris said, to an extent that Dizzy Dean wasn't. Hal just didn't get the press and the post-season.

Chris, you are correct; what I should have said was, given the current ballot trends (top NeL players in the 59 election were hitters) and future NeL eligibles (Jackie R, Campy, Irvin, Doby, and no pitchers I can see), we appear to be on track to honor NeL hitters percentage-wise to a slightly greater degree than NeL pitchers. Unless a Redding or Mendez train gets really rolling.

Tom,

Quite true.

In my reckoning, though, Robinson and Campanella are mostly ML candidates, with Irvin and Doby about 1/2 NeL 1/2 ML. I see the upcoming dearth of transitional-era NeL-to-ML pitching candidates as part of a larger dearth of pitching candidates from the immediate post-war era. For all the talk about the possibility that missing war years didn't necessarily hurt pitchers as it would position players, we see very few serious pitching candidates who missed time during the war. From 1961 to 1969, we'll see only three good pitching candidates reaching the ballot: Feller, Lemon, and Wynn.

Chris, if Monte Irvin is 1/2 ML, then he ain't goin' nowhere. Say it ain't so!

I want to start by asking how many HOM voters currently live in or around New York?

At a buck fifty a gallon, I was "in or around New York." At $3.50, not so much.

Chris, if Monte Irvin is 1/2 ML, then he ain't goin' nowhere. Say it ain't so!

I wouldn't take the percentages literally. I think it just means they need contributions from both MLB/NeL to be electable.

Campanella's got a much higher percentage of NeL time than Doby does, but he doesn't really "need" it as he'll likely go in just on his Dodgers numbers. Doby crossed the color barrier at age 23, but as an OF with 6300 PA and 136 OPS+ he'd love any NeL credit he can get to separate him from the Averill's, Berger's and Wilsons of the world.

As to the #51, first off it IS prelim.

Secondly, I DO have a career fixation. As far as I am concerned when you are giving out career accolades, as we are, you should be looking at career paerformance. The analogies of showing Newhouser around names like Jennings, Dean and Vance work for me. I didn't vote for any of them either.

Re Irvin: I've been working on his MLEs this past week, and his career spans roughly 1937-1957. Of this he played about a decade in the white majors and minors. He played one year in Mexico, and he missed two or three seasons due to the war.

All that said, if what I'm seeing from his MLEs is accurate, the top of the 1962 ballot could be one of the most fractious in recent memory, because Irvin was a big-time player with a big-time peak, and a long career too.

Check it:
-Feller: around 300 WS, and may rec'v WW2 credit.

-Jackie: around 300 WS, plus one year of NgL credit, with WW2 credit an interesting and open question because he didn't play before the war, as well as the open question of how much merit his integrating the league should account for, if any.

-Irvin: A VERY strong candidate who gets WW2 credit.

-Bus Clarkson: A power-hitting NgL/MiL infielder who hit at every stop; he may be among the best sleeper/stealth candidates of them all, though I'm still working on translations so I don't want to oversell him.

-Phil Rizzuto: Better than Gordon/Doerr, gets WW2 credit, and he's a SS, so he could draw serious support among the peaksters and the infield-friendly crowd.

-Ellis Kinder: Depending on how we evaluate his relief work, he could or could not develop into a serious candidate

-Al Rosen: What's his case look like? His peak is massive.

-Mel Parnell: Fans of Dean take note, his ERA+ and peak are great, but could be S.J. Wood v 2.0.

-Hank Thompson: Gets WW2 credit; I've just about finished translations on him, and he's a very intersting candidate. Knowing why he gave up the game after his age 31 season would help and that information will make or break him as a candidate.

Tack onto that the returning backlog, perhaps enhanced by Kiner (does he get WW2 credit?), V. Stephens, and Sid Gordon (he does qualify for WW2 credit), and we could have a serious brouhaha on our hands.

Should be a blast!!!!

open question of how much merit his integrating the league should account for, if any.

JRobinson certainly should get credit for his year in Montreal. As well as perhaps some 2b-fielding credit for being forced to play 1B his first year with the Dodgers.

As much as he is revered as an important historical figure, he is perhaps underrated as a player. His 1949-54 peak is legitimately outstanding with 1949-52 being inner circle type of stuff. I'm not exactly sure how much extra credit he needs. He's easily better than any of the other eligible 2B without any credit.

Early prelim:

(1) Jennings
(2) Newhouser--These two are too close to call.
(3) Van Haltren
(4) Dobie Moore
(5) Ferrell
(6) Averill
(7) Rixey
(8) Childs
(9) Duffy
(10) Oms
(11) Ruffing
(12) Medwick
(13) Ryan
(14) Sisler
(15) Roush

David, thanks, I forgot about his MiL season, that's important. His peak is excellent. Pretty good for a guy who was a football and track star in College.

I want to start by asking how many HOM voters currently live in or around New York?
I live in New York and it will show when Mattingly appears on the ballot

Anyone know why Allie Reynolds didn't get a real shake in the bigs until he was 26? I ask in part because I don't know how light skinned SuperChief was.

Is it OK if I rejoin? This was a busy summer for me and I'll be glad to be invited back.

Anyways, here's how I see the current candidates:

1) Mendez
2) Averill
3) Newhouser
4) Brown
5) Sisler
6) Mackey
7) Gordon
8) Medwick
9) Sewell
10) Griffith
11) Schang
12) Wes Ferrell
13) Oms
14) Rixey
15) Beckley

Jennings has too little outside of his peak. Ruffing's RA+ is underwhelming.

Anyone know why Allie Reynolds didn't get a real shake in the bigs until he was 26?

The Indians signed him in 1939 at age 22. He went to Oklahoma State (then Oklahoma A&M) for college.

His "deadballera" obit says "Reynolds received a track scholarship to Oklahoma A&M, but carried only 140 pounds on a 5-foot-11.5-inch frame at the time and coudl not immediately make the college's baseball team. So to build up his body, Reynolds competed in the javelin and discus throws, and he evenutally added 60 pounds".

Unless there is evidence that he was excelling in the Cleveland minor league system and was unreasonably held back it just appears that he was a late bloomer (at least late by HOM-candidate standards).

welcome, Ardo!

attempting to derail the Medwick express for perhaps one last time.....

Most of us agree that Indian Bob Johnson is not a HoMer, based on his showing up on only 10% of our ballots.

Similarities between Johnson and Medwick:
same era
same position (Bob played 1/2 yr more CF)
similar defense (FRAA has Bob with a teensy lead)
similar hitters
similar career value (WS has Ducky ahead, WARP has Bob)

Differences:
Medwick better peak. Johnson better off-peak.

Okay, now each of us can determine how much to put on 'peak'. But first answer this question, and then look at the data: If player A was X better than player B in their best year, how much better does player B have to be in other years to make this up? 2X? 3X?

By RCAA, in descending peak order, we have
Bob Johnson 61 60 45 37 35 32 28 26 22 19 18 30 413
Johnson*.92 56 55 41 34 32 29 26 24 20 17 16 28 380
Joe Medwick. 92 63 55 31 28 25 25 20 13 11 09 -4 368

The last column is career total; the 2nd to right is 'years other than top 11'.
I did 'Johnson*.92' to account for more offense in Bob's league.

By this measure, Medwick's best year beat's Bob's best by 36 runs. Years 2 thru 9 are even. From year 10 on, Bob wins by 48.

By WARP3,
Bob Johnson 95 89 85 84 76 74 66 62 60 59 56 55 57 918
Joe Medwick 110 97 85 81 70 66 64 62 57 56 52 30 49 879

Joe's best year is 1.5 wins better. Years 2 thru 4 are even. Years 5 thru end, Bob is ahead by 5.4 wins.

By Win Shares, Medwick is ahead. The question is, does Win Shares know something hidden from other tools? In this case, as many have pointed out, Medwick's high totals are from his team's outperforming their RS/RA totals, and maybe from an advantage of Medwick having great teammates compared to Bob.

By Any Other Method Than Win Shares, these guys look even. In order to have Joe up high on my ballot, I must
a. trust Win Shares over all other metrics, and/or
b. believe that a peak year is worth multiple times non-peak years, or
c. I must also be a fan of Johnson.

I do not fathom how much of our electorate hold to a. or b., but that is how the voting has gone.

Thanks, David. Added it to my Reynolds page.

Tom,

I think a few years ago, someone posted that Medwick's Cardinals teams only overperformed by a few wins over his tenure there. I think 1934 was a big overperform, but every other year was around 0. This includes 1937, Medwick's big triple crown year. The only real difference I see is that Johnson played on consistently bad teams. So maybe Johnson is being slighted but Medwick isnt' recieving much of a boost during hsi Cardinals peak.

I would answer yes to both A and B in your post, Medwick may make my PHOM this year, ranking something like #5 while Johhnson ranks around #33 or so because I dont' think he was ever one of the best players in the game. However, 5-33 is much smaller now than it was in 1940 or 1920 or 1898. I see a gap between them, but it isn't a gulf. I wish I could put about 28 players or so on my ballot, if that helps you to see how close they really are.

Thanks, David. Added it to my Reynolds page.

Sure, though the obituary implies that he was 140 pounds early in college and did not make the team right away. I would guess that he'd have bulked up some and made the team eventually or the Indians would not have signed him when he graduated from Okla A&M.

The obit also mentions the cause of his retirement. He hurt his back when the Yankees team bus crashed into an overpass in Philadelphia. Ouch.

allie reynolds deadball era article

Here is what I found and wrote in the 1959 Ballot discussion thread about Medwick's teams and Pythag.


Data from BB-Ref, of course

To the Pythag question I will respond this way:

a) really just the 1936 team was way over their pythag.

b) When someone exceeds their pythag now they are lucky or they have a good bullpen. Dizzy Dean may be the reason, he led the league with 11 saves that year in 17 relief appearances.

c) What difference does it make? Do you think 27 win shares over a period of 7 years somehow have all ended up with Medwick? In 1937 Medwick's really big year they did not meet their pythag, but just narrowly missed it.


All the guys still on the ballot have flaws, I guess I just see fewer flaws in Medwick's arguement than most of the others.

I do think Medwick and Johnson are really close, but that is the problem with these guys right now!

How far apart are these guys really,


Bobby Doerr (18th place) vs Johnny Evers (no votes)

Joe Gordon (28th place) vs Tony Lazzeri (no votes)

Bob Elliot (on 8 ballots) vs Pie Traynor (1 ballot)

George Sisler (11th place) vs George Burns (57th place)

jschmeagol,
I am in and around NY.
Born in NJ, moved to NY, back to NJ, back to NY, and now back to NJ. About 30 miles from Manhattan.

c) What difference does it make? Do you think 27 win shares over a period of 7 years somehow have all ended up with Medwick?

As to what difference it makes, there are two ways to answer the question.

(1) The difference it makes will depend on how you are using win shares. If you place a lot of weight on peaks of three or five years, Medwick's peak is substantially helped by the fact that, in one of his peak seasons, his team had an extra 30 win shares, which translates to an extra 3 or 4 for Medwick.

That's no so many in a career context, but in a 3-5 yr. context, it can be significant.

Couple that with the fact that Bob Johnson, in _his_ peak, has a great season when his team underperformed its pythag by 9 wins (iirc). That costs him about 3 win shares. So, in a head-to-head comparison of peaks, there's a 6-win-share swing between the two. That could well lead to a significant difference in evaluation: Medwick gets a 35-ws season and Johnson gets a 29-ws season for statistically similar performances.

(2) Obviously these differences matter. If they weren't large enough to matter, Bob Johnson would probably be as close with Medwick as Gordon is with Doerr, because every other metric I know of has Johnson superior for career and Medwick only slightly superior on peak.

I'm also in NYC. Born in Queens, raised on the Upper West Side of Manhattan, living now in the far West Village.

Not that he's especially close to making my ballot, but I'm curious: Why not more love for Johnny Pesky? Are those who vote for Sewell (but not him) mainly career and timelining types? Or voters who don't give war credit?

To this peak, nontimelining, war-credit-giving voter, Pesky almost looks better...

I would have to give Pesky 3 very prime years of WWII credit to even him up with Sewell. Which he might have had in him (or even more!), and unfortunately, we will never know. His candidacy for me is somewhat like Luke Easter's; those who belong in the Hall of what Might Have Been.

The late 40s Red Sox had three near-misses (Doerr, DiMaggio, Pesky), all of whom were from reputation nice guys and good citizens and all of whom I feel bad about not putting high on my ballot (Doerr's on it, but only just). On the other hand I know nothing whatever about Jake Beckley's reputation or personality, and have no psychic relation with any of the teams he played for. Must mean I'm doing something right :-)

For the ballot counters:

There is a new version of the ballot counter available on the Yahoo groups site. See the 1959 results thread for (limited) discussion.

1960 prelim.

Another year to clean out one or two from the backlog.

Newhouser is the only newbie to make my prelim, and boring old Sam Rice re-enters. Newhouser is a Ed Walsh type pitcher. Recall that Walsh only won 195 games, and was a regular pitcher for only 7 seasons. With appropriate war discount, 1945 is still one of the great IP/ERA+ combinations we've seen.

Leon Day was very good, but falls short of my ballot. I forgot to mention Dandridge on my last ballot- he had the biggest discrepancy between reputation and statistics that I've seen since Judy Johnson. I can't help but feel that I've been unfair in the aggregrate to NeLers who played in 4-5-6 defensive positions, but none of the unelected ones individually had both the peak and the career that would make me confident that they were a HoMer. Perhaps Lundy comes the closest.

I looked carefully at Easter. While I had hoped that he would be ballot-worthy, his late-career ML performance was, at best, characteristic of the Thompson-Charley Jones kind of slugger that I am not too fond of. Still, his ML track record and other accomplishments might make him worthy of more buzz.

Medwick and Bob Johnson are fairly close in my system, but the slight different puts Medwick near the bottom of the ballot, and Johnson just below. I concede that the difference is probably within the noise of the Win Shares system that I use as the basis of my rating system.

1. Wally Schang
2. Joe Sewell
3. Hal Newhouser
4. Red Ruffing
5. Joe Gordon
6. Bobby Doerr
7. Earl Averill
8. Charlie Keller
9. Tommy Bridges
10. Cool Papa Bell
11. Lefty Gomez
12. Jose Mendez
13. Joe Medwick
14. Biz Mackey
15. Sam Rice

EricC - there has been a discussion on the Q Trouppe thread about how the top candidates at catcher compare. Since you have Schang at the top of your ballot and Mackey at 14, could you explain why you have Schang at the top of your ballot, and how you compare Schang, Mackey, Troppe, Bresnahan and Lombardi.

Thanks,
Mike

If I went purely by my spreadsheet system, Schang would be #1 on my ballot as well.

The basis of my method is career value over a high "replacement" threshhold, where the threshhold is basically guys who had significant careers as regulars or near regulars, but weren't very good. Guys like Tim Leary, Ed Kranepool, and Omar Moreno, who never made any real progress toward the HoM.

Plus I give a catcher bonus of about 25%. On my actual ballot, I manually bump Schang down because his impact was so diluted in any one season. But we would be #1 by the career numbers. The OPS+ and career length at catcher are impressive.

Looking at the similarities between my ballot and EricC's I wonder if we have similar systems.

Welcome back, Rube. Newhouser #11, Day off ballot around #45. Elliott and Medwick for my PHoM this year.

1)Averill--Looks like a HoMer to me even without PCL credit, but I do give him some PCL credit as he was obviously major league quality before arriving in the majors. Compare him to Goslin: Averill has a higher OPS+ (133/128), and is an A+ CF vs. a C+ LF. Goslin has career length, mostly because Averill plays in the PCL for a while. PHoM in 1948.
2)Elliott--I like him better than Hack. PHoM in 1960.
3)Gordon--Fixed my war credit, he and Doerr moved up. PHoM in 1958.
4)Doerr-- PHoM in 1958.
5)Sewell--109 OPS+, reasonably long career, good shortstop (A- Win Shares). Yes, I am allowing for his switch to 3B at the end of his career. PHoM in 1939.
6)Schang--His rate stats would put him in the HoM, but a look at each individual year isn’t impressive. Still, a hitting catcher with his career length isn’t common...Bill James rates him a C+ fielder in Win Shares, but says he was a good catcher in the NHBA. PHoM in 1938.
7)Medwick-- PHoM in 1960.
8)Mackey--#2 on my 1949 prelim, but more data on his hitting has dropped him to here.
9)Cool Papa Bell--If Max Carey is in, Cool Papa should be too.
10)Bob Johnson--A very underrated player. Usually I'm a WS guy but this time I think Warp has it right.
11)Newhouser--
12)Doyle— His hitting is legitimately outstanding, he played 2nd base, and a C+ defender by Win Shares. 126 career OPS+, compare to contemporary George Cutshaw, who was a regular 2B for 11 years with an OPS+ of 86. #19 all time in innings at 2B. Regularly in the 2B defensive Win Shares leaders, WS Gold Glove in 1917. Top 10 in Win Shares 1909-12, 1915. PHoM in 1926.
13)Beckley— Behind the big 3, much better than other dead-ball 1B. Win Shares best fielder at 1B in 1893, 1895, 1899, and 1900. Add in 2930 hits, with power and walks. No peak but a lot of consistent production, we’re not talking about Ed Kranepool here. PHoM in 1913.
14)Rixey—Early Wynn will be the next pitcher with more IP, his W/L percentage isn’t high because he didn’t get a lot of support. ERA+ is very good at 115 for such a long career. PHoM in 1939.
15)Waddell—Waddell has a run of 7 years (1902-1908) in which he was blowing people away, in three of those years with an ERA+ over 165. A seven year peak for a pitcher is much more rare than a seven year peak for a hitter, I give the short peak pitchers a lot more credit than the short peak hitters. PHoM in 1916.

jschmeagol and other NYers, I will be in the NY area (SE CT) Sept. 24-25, I could maybe swing something if you all are getting together, but the 25th I have a fantasy hockey draft.

Originally from LI, high school in CT, slowly making my way down the coast, currently in the WV panhandle . . .

*******

I think the Medwick being overrated by playing for teams that overachieved in terms of pythag is extremely overblown, it's a very minor thing at most, if you even consider that a negative. Teams don't necessarily overachieve their pythag (somewhat) and their expected runs created (much moreso) because of luck. I think it's also quite likely (especially with regards to RC) they overachieve because of things that aren't explained by the formula. RC is an estimate, not a fact.

Sorry for the rant :-)

could you explain why you have Schang at the top of your ballot, and how you compare Schang, Mackey, Troppe, Bresnahan and Lombardi.

Mike- the short answer is that I use a rating system based on strength plus length. Strength is based on Win Share rates minus a factor that is inversely proportional to the square root of career/prime length. Peformance of the whole career/prime is rated, without regard to how it is distributed over individual seasons. The top WS/162 rates among eligible ML C are (1) Bresnahan 25.9 (2) Schang 21.6, (3) Lombardi 19.1. Bresnahan loses his strength advantage to Schang in my system through his shorter prime and having less catcher bonus due to playing proportionally more games at other positions. Since Schang has a strong length advantage, he comes out on top of the eligible ML catchers. Lombardi's WS rates are too low to overcome.

The NeL catchers are more difficult to judge, due to limited information but I try to follow the same philosophy. I have Mackay close to the consensus rating. I need to reevaluate Trouppe, as I have not given proper credit for his WWII gap nor his post-NeL career.

So I THINK, based in part on Ron Wargo's numbers, that we have:
29 in the Hall of Merit, but NOT in the Hall of Fame.
56 in the Hall of Fame, but NOT in the Hall of Merit.

I think we've had 114 elected.
That would mean 85 of our electees already were in the Hall of Fame.
And it would mean that we've had 141 Hall of Famers become eligible so far, and elected 85 of them to a possible 114 slots.

Is that right?

Hey, Dimino, if my wife is out in Vegas on business that weekend, that could be perfect! I get bored when she's not around. I'll check with her in the next day or two.

Yeah, the 24th would be cool if we could get something together. I live downtown on William and Maiden, a few blocks from wall street and the former WTC. I don't really know the city well, I just moved here last sunday, but I am up for hitting a bar or something and talking about how much Joe is wrong by not voting for Hughie Jennings.

The ballot's so deep and complex I started to fart around with something I've been avoiding. I tried to come up with some basic mathmetical model to loosely rank the hitters (just by hitting ability).

I took OPS+ (actually, AOPS because I was using the 'cycoledia, not b-ref), and typed in it and games played for each year of a player's career. Then I took AOPS, and doubled the difference between it an 100 (in other words 140 became 180, 97 became 94, etc), and then multiplied it by games played in a season and divided it by the games played by the team that year. So it's nice souped up version of AOPS combined with ability to play every day. That's for single season.

For career I just figured out how many season's worth of games a player played in his career and multiply that by a souped up version of career AOPS.

Then, I try to combine career & prime. I took the 7 best years and declared them the prime, and added them together - not equally though. I weighted them differently with the best season getting weighted the heaviest and the 7th best season weighted the lightest.

To combine this with career I figured the prime & career numbers for 16 guys, I figured out how much I'd have to adjust the career numbers by to make it worth 40% of the whole with prime being 60%.

Lots of crappy crappy math here, but it helps give me a basic sense of what the hell I'm doing instead of randomly groping in the dark with my own general sense as a guide. Why double the difference from 100 with AOPS? I want the best seasons to really stand out from the average or OK ones. Why 60/40 prime/career? Because it sounds right to me.

At any rate, here's how it all came out:
1. George Sisler
2. Joe Medwick
3. Jake Beckley
4. Bob Johnson
5. Jimmy Ryan
6. Gavy Cravath
7. Earl Averill
8. George Van Haltren
9. Edd Rousch
10. Larry Doyle
11. Cupid Childs
12. Ernie Lombardi
13. Tommie Leach
14. Lave Cross
15. Joe Sewell
16. Wally Schang

Now this doesn't take into account defense, baserunning, minor league credit, position, quality of competition, and a host of other factors. Still interesting to me for several reasons.

I'm amazed Beckley did that well, but though he had no strong peak, he did have a very nice and very long prime. Against him is that a lot of his best years were in weak leagues.

Sisler was another surprise, but he did have a terrific peak, and a very good top 7 years with a long enough career to score well there. He wasn't on my ballot last time but he will be this time. Especially helping him is that the next few guys behind him all had some of their best years in weak leagues (WWII, expansion).

Helps Lombardi move up, and Schang move down. Leach will likely drop off my ballot. Lave Cross wasn't on my ballot but he wasn't too far off. Now he'll drop down as well.

Averill, Van Haltren, & Roush are all real close. It'll be tricky to try to work that one out.

Childs and Doyle came out almost identical, so that ends whatever chance Doyle had of leaping Childs on my ballot.

Sisler is no surprise at all! ;-)

I've been saying for years that if he had had a normal career curve, he would have been elected by now. Take 3 of his last 7 years and put them at the beginning of the career, leaving a 4 year decline. Voila! HoM!

And among the guys who also carried a glove--Doyle and Childs!

Excellent system system, Chris!

Finally getting to look at the new candidates:

Newhouser will be ranked high. Has some postwar success to go with wartime dominance. Allie Reynolds? To show you how a team performance can distort images of players, I doubt Ken Raffensberger is any worse than Reynolds. Difference is the original Raffy spent time in the PCL, little wartime service, and also on poor teams. Allie got to ride the Yankee train to success. Leon Day is not going to make the ballot - other NeLers above him in the queue.

I'll have to double check Pesky. Not likely. He's honored with a pole in Fenway anyways.

Oh, and vote early and often for Indian Bob.

Quick little note on a guy who is either eligible this year or the previous year...I think.

Claro Duany was a Cuban outfielder who played a couple years in the NNL for the Cubans. He apparently was a late bloomer because he didn't hit a lick until around his age 27 year when he suddenly started winning Latin-American batting titles left and right. He appears to have also been a pretty decent power hitter and to have drawn around an average number of walks relative to his leagues.

A quick OPS+ and WS summation of his career follows. I'm not sure my translations indicate his power very well because the CWL, which I used to help balance out sample sizes, was not a very power-oriented league. It was very frequently an extreme pitcher's league in the 1930s and 1940s, and though offense picked up a bit in the late 1940s, it was not a hitter's paradise by any stretch of the imagination. Anyway, I think the extreme nature of this power-dampening environment is having the effect of ratcheting down players' converted SLGs.

Regardless, here's how my translations see Duany (assuming he was an average outfielder).

YEAR AGE POS  G  OPS+ SFWS
---------------------------
1942  25 OF  30   -4  -1.0
1943  26 OF  89   96   7.2
1944  27 OF 143  129  20.9
1945  28 OF 153  113  19.0
1946  29 OF 126  146  22.2
1947  30 OF 143  118  21.1
1948  31 OF 109  110  13.7
1949  32 OF 154   89  14.4
1950  33 OF  91  108  12.2
1951  34 OF 139   93  13.8
1952  35 OF 105   98   9.9
============================
total      1282 108 153.3

Not much of a candidate, but since I'd done the translation, I figured someone else might find it interesting too.

Doc,

You are right not much of a candidate, but it is always interesting to learn about different guys so thanks. The more of Duany's we do the better udnerstanding we may get of a Brown or a Cepeda.

I have a question regarding Red Ruffing as he may go in this year and will probably be the top candidate heading into 1961:

Why is he better than Eppa Rixey?

I am not a Ruffing supporter (#35 last ballot) but if I could be convinced that he was the best long career, moderate peak pitcher on the board (we haven't elected oen since Ted Lyons) I could live with his induction. But I don't see how he is better than, let alone as good as, Rixey or for that matter Redding.

I would rank them (I know I am missing some):

Redding
Rixey
Ruffing
Grimes

Could any Ruffing supporters shed some light here?

I really think that the translation method is underselling Duany's power a bit, and I think we'll see it again with Campy and Minoso. I can't tell whether it's something I can fix or not, but I'll definitely be looking for feedback when we get to candidates whose profiles are more familiar to us.

Ruffing hit better than Rixey.

Career length, if you adjust for usage patterns in their days, is essentially the same.

League strength a big factor if you believe BP cards or Dallas Adams' old study.

Better peak for Red.

Aside from all that, Rixey would be a bit ahead, absed on ERA+ and Ruffing's lack of great W-L record with N.Y.

In the absence of a 1961 thread, this goes here. Somebody asked about Kiner being better than Berger, Klein and Hack.

WS
Kiner #18 LF--242/37-35-30/155/26.6 no WWII credit
Berger #13 CF--241/36-33-31/152/28.9
Klein #40 RF--238/31-30-28/140/22.0
Hack Wilson #19 CF--224/35-32-31/152/26.9

Very close on totals (except Hack) and peak (except Klein and I am skeptical of these numbers). Advantage Berger on rate, possible advantage Kiner if you think he deserves WWII credit.

WS Peak
Kiner 37-35-30-30-23-23-19-19-15-11=242
Berger 36-33-31-26-26-23-21-17-16-11-1=241
Klein 31-30-28-26-25-19-17-17-13-11-9-7-5=238
Wilson 35-32-31-28-26-21-16-13-12-6-4=224

Kiner's totals are better all the way out from his 1st year through his 10th, and nobody adds extra years to put them ahead. As for individual years:

Best of the best--Kiner
Best 2nd best--Kiner
3rd best--Berger and Wilson
4th--Kiner
5th--Berger
6th--Berger and Kiner (Kiner leads 178-175)
7th--Berger
8th--Kiner
9th--Berger
10th--Berger, Kiner and Klein

OPS+ (BA eligible years only)
Kiner 148/183-82-72-54-45-40-31-21-16 (9 years)
Berger 140/177-51-48-43-39-34-21 (7)
Klein 135/168-58-55-49-49-36-28-23-22-20 (10)
Wilson 145/177-60-59-55-50-42 (6)

Best of the best--Kiner
Best 2nd best--Kiner
3rd best--Kiner
4th--Wilson
5th--Wilson
6th--Wilson
7th--Kiner
8th--Klein
9th--Klein
10th--Klein

Raw Numbers
Kiner 6236 AB+BB 369-1015-.279-.398-.548/148
Berger 5598 AB+BB 242-898-.300-.359-.522/140
Klein 7087 AB +BB 300-1201-.320-.379-.543/135
Wilson 5424 AB+BB 244-1063-.307-.395-.545/145

It seems clear that these players are pretty close in their skills and their production rates, but: Kiner was slightly more productive on a rate basis, as little as .003 over Wilson on OBA and SA, but still better. Much better than Berger on both, better than Klein on OPS+. And in more PAs than Berger or Wilson.

Klein is the only one you could make a case for IMO in that he is in the ballpark on rates and with more PAs. And I have Klein generally in the #16-20 range on my ballots, with Berger and Wilson further back in the 50-60 range.

The bigger question for me will be not whether Kiner is better than these guys (he is), but whether he is better than the higher ranking hitters (on my ballot)--Medwick (who may be gone by then), Sisler, W. Brown, Pete Browning, Charley Jones. Those will all be difficult comparisons, of course. But since Medwick and Sisler are currently at the #3-4-5-6ish spots, they are the benchmark. I'll do them later.

Since Doc says Kiner doesn't even make his consideration set (too short of a career), I just thought I'd compare Kiner to some guys ("hitters") on Doc's ballot, strictly for bulk. I mean I can't seriously believe Kiner with his 148 OPS+ and .398 OBA was performing at too low of a rate. So just strictly for bulk.

Kiner BA eligible 9X, 6226 AB+BB, 148 OPS+

2 (on Doc's1959 ballot). Mize 9, 7299, 157
3. Duffy 11, 7708, 121
4. Oms
6. Medwick 12, 8072, 133
7. Cravath 7, 4512, 149
11. Burns 11, 8113, 115
12. Browning 11, 5286 (I adjust to about 7000), 164
13. W. Brown
17. Averill 10, 7127, 132
22. B. Johnson 13, 7995, 139

With a couple-three exceptions (I would call Oms an exception with his OPS+ of about 125 but I don't remember how many PAs he is projected), this is a moneyball ballot. What moneyball fan could not love Ralph Kiner, I wonder?

Well, somebody who wants 11ish years of big league play, including a (possibly) generous assist in the form of extra credit for those with unusual resumes.

Doc, I'd ask you to consider whether Kiner deserves WWII credit as much as Averill and Cravath deserve AAA XC. Kiner was "in the system" before the war and broke into the bigs in 1946 at OPS+ 116, after never playing in the bigs before and barely swinging a stick for 4 years. Then: 172-145-183...

And I'd ask you to consider this: Kiner did play 113 games in a new league (the AL) at OPS+ 116, so in reality he is at 9.5 years, not just 9.

Finally, are you really sure that you want George Burns on your team for 11 years instead of Kiner for 9.5? I'll take my chances with the proverbial replace-mat for 1.5 in the deal. Especially considering that after Burns best 9.5 seasons you're left with OPS+ 88-82-77. This is the bulk Kiner doesn't have. Yippee! And the 9.5 side by side:

Burns 149-48-42-29-26-21-21-7-6-1
Kiner 183-82-72-54-45-40-31-21-16-16

My point is not that Burns is a bum. My point is, Kiner is not even in your consideration set? How many players is that, 100? 50? And you really want 9.5 years of George Burns plus 3 years at 88-82-77 rather than 9.5 years of George Kiner? Wow!

Doc, I'd ask you to consider whether Kiner deserves WWII credit as much as Averill and Cravath deserve AAA XC. Kiner was "in the system" before the war and broke into the bigs in 1946 at OPS+ 116, after never playing in the bigs before and barely swinging a stick for 4 years. Then: 172-145-183...

To start off with, I certainly agree with your point re: Kiner vs. Burns. I also have no idea where Kiner will rank on my ballot; my suspicion is that his putrid D will hold him back some, but that's not the point of this post.

Kiner missed ages 20-22 to the war. Averill gets credit (from most people) for age 26 (maybe back to 24). Cravath is getting a variety of AAA credit, at ages 29-30, as well as some pre-ML credit (ending at age 26, various people have differing opinions on how far it goes back).

Now, this is not a disqualification to Kiner WWII credit. In fact, I'd be interested in seeing his minor league numbers before he left. But surely we should be a little more cautious in assigning extra credit to 20-22 year olds vs. guys in their mid 20s?

Now to the real nut of it--Kiner against his real eligible comps:

WS
Kiner #18 LF--242/37-35-30/155/26.6 no WWII credit
Medwick #13 LF--312/40-36-33/157/25.5
Sisler #87 1B--292/33-29-29/135/23.0

Oh, no, Sisler is #24, Bill James' bad. Still, advantage Medwick, Sisler just depends on your career/peak orientation.

WS Peak
Kiner 37-35-30-30-23-23-19-19-15-11=242
Medwick 40-36-33-24-24-24-24-22-21-19-19-9-5-5-4-3=312
Sisler 33-29-29-27-25-24-22-19-16-15-13-11-11-10-8=292

Advantage Medwick

OPS+ (BA eligible years only)
Kiner 148/183-82-72-54-45-40-31-21-16 (9 years)
Medwick /179-57-49-40-38-33-29-28-22-18-15-97 (12)
Sisler 124/179-69-63-59-53-37-33-9-0-98-90-84 (13)

Advantage Kiner for 3 years, though it is slim for 1 year, then significant, then Sisler for 2 years, then Kiner, then Sisler for 2 more years, then Medwick finally has the better out years and stays above average through year 11. This is a bit of a muddle, so I would go back to Kiner's superior career number.

This is obviously a closer call than G. Burns, where you're getting 2 extra years (Burns) and sacrificing 9 years of 25-30 point OPS+ differentials to get them. Here you're getting 3 extra years (Medwick) and only sacrificing--well, in this case, a variety of differentials all the way from 4-5 to 25ish.

I wonder what Pennants Added would say about that?

Raw Numbers
Kiner 6236 AB+BB 369-1015-.279-.398-.548/148
Medwick 8072 AB+BB 205-1383-.324-.362-.505/133
Sisler 8739 AB+BB 102-1175-.340-.379-.468/124

Some people will say, well, Ducky wasn't a moneyball player, didn't walk enough. And Kiner had too short of a career. And I don't recognize Sisler's career shape. I'm gonna hold out for Ted Williams, among the "hitters." I see all 3 of them at the top of the second tier and if I had to elect one...

Well, nobody I can see had 9.5 years like Kiner's, and I'd be hard pressed to trade for 2 or 3 extra out years from anybody. And if there's anybody, I guess it's Ducky since Sisler's out years aren't worth that much.

Michael, age is a factor. But after the war at age 23 Kiner broke in at 116-73-45-82...

Averill, whatever his age was, is argued as "shoulda been in the bigs" and then broke in with 134-131-137. Does this mean he was better in AAA at age 26 than Kiner was in the miltary at age 22? No. Does it prove that Averill was better at age 26 than Kiner at age 22? No.

In Caravath's case, he spent 1910-11 in AAA (age 29-30) then broke in with 118-169-157-170, suspisciously similar to Kiner's start. Does it matter if he was 29 and Kiner 22? Maybe, maybe not. There's not much there to say that Cravath was better in 1911 than Kiner "was" in 1945, however.

I was just pointing out that Doc's ballot is shaped to a much greater extent by "extra credits" than the average voter, and that he oughta consider Kiner's case a little further. I just can't believe a guy with OPS+ of 183-82-72-54-45-40-31-21-16-16 not making a top 50, and if there's even one more year of MLE value there, maybe he cracks the magic number, whatever it is.

Jingoist here again, lobbying hard for my buddies Duffy, GVH and Ryan.

I need some help uderstanding how Chris J can rank Ryan @ #5, GVH @8 and Duffy nowhere in his top 15? I love the fact that 2 out of the 3 got top ten billing; just confused why Duffy doesn't make the grade.
Duffy scored and drove in 2854 runs/rbis on just 2,282 hits.
Not many of the greatest players ever scored/drove in 570+ runs over their hit total.

His fielding stats are quite good regardless of OF position played showing excellent range and fielding ability, better than either Ryan of GVH.

However it's sliced these 3 guys are better HOM candidates than Cuyler who the electors saw fit to induct.

ps Jake Beckley and Earl Averill deserve slots as well.

Keep up the good work; I remain a big fan of your collective efforts.

I agree with Michael Bass on the war credit issue re Kiner. I would likely be persuaded by a good argument, though.

Before I start out here, I should point out that in 1988 or so I was at 3-Rivers for Ralph Kiner night and I've got two Special-Collectible(TM) Ralph Kiner Night Programs to prove it!

I look at the top 25-ever guys per defensive position (and about 75-90 pitchers give or take) and I compare peaks, primes, and careers. Kiner is literally right on the edge.

I'm a WS guy; I adjust for schedule length.

NAME   PEAK3 PEAK5 PRIME10 EXPRIME15 CAREER
--------------------------------------------
KINER    107   163     254       254    254
BURNS    105   161     279       314    314
MEDWICK  114   165     280       321    324
CRAVATH* 105   165     281       325    325
BROWNING 110   173     276       295    295
AVERILL*  98   157     290       326    326
DUFFY    104   167     290       338    339
*Includes minor league credit.

Ralph vs. Not-Tioga. The difference in their peaks is minor, less than a WS a year after 162ing it. Where the difference becomes apparent is in the prime/extended prime where Kiner's the only guy whose prime is under 275: by 20 WS.

I would rank Kiner much closer to Charlie Keller or Bobby Veach than I would the other guys. But remember, we're not talking about a big difference here either. I've got Burns as the 45th or so best corner outfielder in history by my system. So it's my opinion that Kiner's around #50.

OK, so that means it's metadiscursion time. Burns is my 45th best-ever corner OF. And he's on my ballot. That's roughly the same as being the 22nd best player at another position in my system. So who's nearest to 22 at the other positions (that is among eligible, elected or soon to be eligibles)?

                                  #ELIGS
                                   AHEAD
NAME     POS     3   5  10 15  TOT @ POS
------------------------------------------------
CRAVATH# CROF  105 165 281 325 325  5
BURNS    CROF  105 161 279 314 314  5
RYAN      CF    98 151 259 342 368  9
RIZZUTO*  SS    90 141 258 318 318  2
SISLER    1B    96 150 255 311 311  0/1**
ELLIOTT   3B    87 137 242 295 295  1
DOERR     2B    85 138 252 319 319  3
SCHANG    C+    77 123 218 294 314  3
+Includes catcher bonus
*Includes war credit 
#Includes minor league credit
**Depends on how I feel about Luke Easter, he is probably ahead of Sisler.

I suppose this explains why I've got a lot of OFs on my ballot. They look better by comparison to players at other positions. But it also shows that I'm not being terribly inconsistent either. Sisler, Schang, Ryan, Doerr, Elliot, and Rizzuto would all be off of my ballot if we voted on them today. Burns/Cravath might be an inconsistency, but I'm not certain because they've simply got gaudier numbers. Maybe I should be docking corner OFs the same way that I boost Cs?

I guess I'd better duck for cover.I have a feeling I'm going to catch some flack.

I need some help uderstanding how Chris J can rank Ryan @ #5, GVH @8 and Duffy nowhere in his top 15?

Easy - those 16 guys I posted were all that I had done the math on. Duffy's always been off my horizon. Maybe I'll plug him into the system at some point.

Couple quick things on Burns vs. Kiner.

Burns's glove makes up some ground for the differences in their bats.

NAME   FWS    G  FWS/1000
----------------------
Burns 46.8 1844  3.30
Kiner 25.8 1382  2.20

There's not too much benefit to all of Burns's SB because he stole at a terrible rate (just over 50%), however, Kiner was top-ten in GIDP four times and finished with 126. No data exists for Burns, but as a speedy player, I don't think it's unfair to say he would have many fewer.

Finally, the seeming closeness of the PAs doesn't take into account the differing offensive levels of the game. Kiner had more opportunity to accrue PAs than did Burns. I took a quick look at 1917 and 1949 (to pick two years). Both leagues batted about 42000 times, but the 49ers walked 1350 more times than the 17ers. True, they did sacrifice less often and probably GIDPed more often, but on the other hand, deadball baseball saw more CS as well. That and Burns playing in 1918 and 1919 help explain why the difference in PAs isn't wider between them.

Which is all to say that there are small little things that begin to even things up a bit. I might be wrong in having one and not the other, (I probably am!), but it's not like OPS+ is the only contributing factor here, and because of the deadball era's different style of play, it may not be the best cross-generational comparison tool either.

Doc, I guess if it's all WS all the time... but I just can't look at all the other numbers and come up with Kiner's peak being no better than Burns et al. Does not compute.

Let's see what WARP says.

NAME WARP1 WARP2 WARP3 EQA EQA2 FRAA FRAA2
----------------------------------------------
KINER 74.9 70.9 72.5 .319 .319 -70 -79
BURNS 96.3 68.0 70.6 .285 .274 84 59

So depending on your take with WARP, the little things add up. Now WARP2/3 with its steep timeline adjustment agrees with you by virtue of its heavily penalizing the early-1900s NL, to the tuneof about 25% of Burns's value. That seems a bit oppressive to me. However, its not-bad fielding metric also supports the idea that Burns was a good fielder and Kiner was, well, able to stand up in the outfield for 25.5 outs at a time.

I personally am skeptical about WARP2/3's timeline adjustment (it's one of the reasons I don't like WARP and I use WS). But in the same way that I'm upping Johnson and Elliot because of the disparity between their WARP and WS, I'm inclined to find support for my placement of Burns between his WS and his WARP1 scores (that is his placement viz Kiner).

Sorry everyone, here's the chart from my last post, this time with formatting (and feeling!!!!).

NAME   WARP1  WARP2  WARP3  EQA  EQA2 FRAA FRAA2
----------------------------------------------
KINER  74.9   70.9   72.5  .319 .319  -70  -79
BURNS  96.3   68.0   70.6  .285 .274   84   59

It comes down once again to an extra 2 years of a .285 EQA guy, and all you have to sacrifice for that is 9 years of .319.

For any NYCrs, over in primer, Chris Dial's flying up to NYC and is calling for a primer meetup on 9/14. Thread. If you've never met Dial, you should, it's an enjoyable experience.

Man, I'd love to go. The 14th? We'll see if work will let me.
That thing on the 24th with Dimino and allies is looking promising, though!

Two quick comments on Ralph Kiner:

1 - I don't see giving him war credit. In 1946 at age 23, with a 117 OPS+ and a slug in the outfield (I can't believe the Pirates actually let him play 76 games in center field), he was a below average outfielder. Maybe without the war he would have developed earlier, but that's too much speculation for my taste.

2 - The comparable player who does deserve WWII credit is Charlie Keller. A better hitter than Kiner, and a _much_ better fielder, if you add the 200-250 games that Keller is missing during his age 27 & 28 season due to the war, his career is almost as long as Kiner's. It's hard to see anyone who gives war credit ranking Kiner ahead of Keller.

Prelim

1. Jennings
2. Newhouser - I compare him to Vance in some ways, but with a better peak. A WWII discount takes him down a notch, however.

3. Ferrell
4. Childs
5. Medwick - It is between he and Lyons for my PHOM, Ducky is more likely to get the honor.

6. Duffy
7. Redding
8. Walters
9. Keller - I agree with Brent, he was better than Kine with war credit
10. Averill
11. Moore
12. Grffith
13. Rixey
14. Trouppe - For a catcher he has av ery big peak, playing 140 games at the positin will do that for you.
15. Gordon

16-20 Doerr, Browning, Dean, GVH, Sisler

Quick comments from me on 1960 itself...

- Newhouser is an exceedingly obvious #1 to me, probably better than even Mize. I give very little war discount to him: his 44-45 fit in very well with his 46-49. Simply an amazing peak candidate.

- Reynolds and Day are not serious candidates.

- Nor, IMO, is Pesky, though I could see a peak voter who gives very liberal war credit rationalizing him onto the ballot.

So I actually get to add someone from the backlog onto my ballot. Yay! The lucky winner this time is Bob Johnson.

Birdie: Confessions of a Baseball Nomad. Page 69.

"Newhouser got into the Hall of Fame by begging to get into the Hall of Fame. And Newhouser was no more a Hall of Fame pitcher than was Schoolboy Rowe or Tommy Bridges. There was a guy named Ed Head who had pitched a no-hit, no-run game for Brooklyn and he was inducted into the army in Texas. I had the army post team, and we faced Ed Head in a tournament down there after his no-hitter and beat him rather soundly. And then Howie Pollett, who was one of the really great pitchers, came into Texas to pitch in our semiprofessional atmosphere, and our army post team beat him. Then they inducted Tex Hughson, who won 18 and lost 5 before he was inducted. An oil company paid him a lot of money to pitch a game against the Waco Army Air Base and we beat every one of those guys. So I must say that if Hal Newhouser were pitching in Texas semipro or Texas service ball he might have had a little trouble winning. He was a good pitcher, but he was not the greatest left-handed pitcher in the world. And there are people I have caught whom I consider more qualified to be in the Hall of Fame then Hal Newhouser, because his record was almost completely a wartime record."

Sunnyday2,

Depends on what you think of WARP1 v. WARP3 (IIRC, you're not a WARP3 believer, but I might be confusing you with someone else). WARP1 is giving him nearly 20 more wins over those 2 extra years of .285 EQA (plus well above average fielding), while WARP3 with its nutty timeline makes them out as about equal.

20 wins in two years is two MVP-level seasons. That's not as trivial as you're making the difference between them to be, and it's parralleled by the gap between their extended primes in WS.

Birdie Tebbetts, I presume. Tebbetts in the Army would have to be within the years 1943-45. Maybe the fact that he had only 87 games in 1946 represents some time away, but mostly I think it just means he was splitting time at C with Paul Richards. (Neither one of them could hit.)

I had the army post team, and we faced Ed Head in a tournament down there after his no-hitter and beat him rather soundly.

The timeline is wrong - Head pitched his no-hitter in 1946.

Head did get off to a fast start in 1944 - 9 G, 63 IP, 131 ERA+, so perhaps that's the year he was drafted. But the no-hitter came later.

...Howie Pollett...

Although Howie Pollett had looked very good in limited use in 1941-42-43, he had fewer than 300 ML innings pitched when he went away to war. He came back to establish himself and have an excellent career after the war - 131-116 record, ERA+ 113. Tebbett's wartime sample of Pollett is one game.

Then they inducted Tex Hughson, who won 18 and lost 5 before he was inducted.

Tex Hughson was indeed a superb pitcher. Just two things to say. First, his 1944 record (ERA+ 150 in 203 innings) is not out of line with his whole career and certainly not out of line with his 281 IP of ERA+ 144 in the far less diluted 1942 season. Second: the quality of the AL in 1944 included that it had Tex Hughson for most of the season.

Tebbetts says of Newhouser, "his record was almost completely a wartime record." That's a distortion - it's not really true of Newhouser, and it's not really true of some of the other pitchers who were good in the 1945 AL, such as Benton, Wolff, Leonard, or Newsom.

And there are people I have caught whom I consider more qualified to be in the Hall of Fame then Hal Newhouser

You know, Tebbetts caught an awful lot of very fine pitchers. With Detroit, besides Newhouser, he had Bridges, Neswom, Trout, Rowe - all of them at least marginal candidates for the HoM, and Bridges in particular has significant support. Moving to Boston, he caught Mel Parnell, as well as the aforementioned Tex Hughson. He finished his career with Cleveland and Lemon, Wynn, Garcia, and Feller.

Obviously, Tebbetts knew Newhouser a lot better than any of us do - but he doesn't have to have been right about this.

Birdie was just talking big in a book called Confessions. If Bob Feller hadn't come back with a mean streak and struck out every mother he faced in '46, Newhouser would've had a pitching Triple Crown. Hal simply had the misfortune of having his personal performance peak during the war - he threw 54 complete games in those 2 years, so clearly his arm was humming.

On a completely different note, Newhouser quit his special scouting job with the Astros because they went with Phil Nevin over The Jeter. Whether this is a plus or a minus in your book is up to you.

Doc,

I for one don't timeline to the degree that WARP does but I would find it hard to believe that the 1910's NL was as strong as the post war NL. I think that Kiner does deserve some edge there versus Burns.

This, by the way, is one reason that I like to use both WARP3 and WS. One timelines too much, the other has no strength of league adjustment. I think the truth is somewhere in between.

jschmeagol,

If it's true that the 1910s NL wasn't as good as Kiner's NL (which I agree may be true), what's the appropriate discount rate? WARP3's discount rate is 26.7%. Put another way, WARP3 believes that the 1910s NL was roughly 73.4% as good as the 1940s NL. That would peg the 1910s NL as being less than a single-A league as compared to the 1940s NL. That seems wildly inaccurate.

For perspective, I, like Chris Cobb, discount the NgLs at 10% for AVG, 18% for SLG. If we simply discounted Burns at 10% for argument's sake, he'd be at 86.67 which is still 14 wins (or two 7.1 win seasons) ahead of Kiner.

Again, two 7.1 win seasons is not insignificant.

Here’s another look at the Medwick/Johnson business, with Averill and Sisler included as well, to bring in all the top “hitter” candidates in the major-league pool, according to the support they received in 1959. (I hope to present Oms, Brown, and the 19th century hitters in comparison to this group next week).

I wanted to get several views of this group of players across a set of seasons. I looked at career, totals for each player’s first 10 years, since Medwick and Averill both had primes of about 10 years, and at five-year consecutive peak, since all of these players had peak stretches of about that length. I avoided grouping together non-consecutive seasons or groups of seasons smaller than five, since that could exacerbate the effects of good/bad teams and over/underperformance vs. pythagorean expectations. First, a view of 5-year peaks:

Medwick      Averill     Johnson      Sisler
1935-39      1930-34     1936-40      1918-22
3328 PA      3378 PA     3138 PA      3206 PA**
761 g        752 g       731 g        716 g**
152 OPS+     138 OPS+    139 OPS+     161 OPS+
.324 EQA     .307 EQA    .312 EQA     .327 EQA
-5 FRAA      20 FRAA     -5 FRAA      61 FRAA
98 FRAR      170 FRAR    94 FRAR      142 FRAR
332 BRAR     279 BRAR    274 BRAR     350 BRAR
137.9 bws*   113.4 bws   93.4 bws     129.7 bws
17.4 fws     29.2 fws    15.0 fws     11.8 fws (0.3 pws)
*Medwick’s ws are for 33-37, because it sees the peak there, rather than 1935-39 as OPS+ and WARP see it.
**Sisler’s PA and G for 1918-19 have been prorated to 154-game seasons to calculate rate stats for the period.  WARP and WS totals have also been adjusted.

Clearly Sisler and Medwick have the superior peaks as hitters. OPS+ and EQA show Sisler as superior, as does BRAR. WS shows Medwick with the better total, which is due to his superior durability, as the two have exactly the same rate of bws/game. Interestingly, OPS+ and EQA show Johnson as the superior hitter to Averill, but WS has Averill far ahead.

On the defensive side, both systems agree that cf Averill is the most valuable defender by a considerable margin. WARP1 and win shares see the rest of the group quite differently. WARP1 sees Sisler as a brilliant defensive first baseman, far above average, and with the greater value of first-base defense in the deadball era, he places far ahead of the slightly below average left-field defense of Johnson and Medwick. WS sees Sisler as a good, but not great, first baseman, and his value trails significantly behind both Medwick and Johnson. WS sees Medwick as an excellent corner outfielder, Johnson as a good one.

WARP1 orders their peaks
Sisler 492 F+BRAR
Averill 449
Medwick 430
Johnson 368

WS orders them
Medwick 155.3
Averill, 142.6
Sisler 141.5
Johnson 108.4

Now for first 10-years (primes)

Medwick      Averill      Johnson     Sisler
5901 PA      6647 PA      6323 PA     6105 PA*
1476 g       1485 g       1459 g      1374 g*
142 OPS+     136 OPS+     137 OPS+    142 OPS+
.308 EQA     .307 EQA     .308 EQA    .305 EQA
-14 FRAA     15 FRAA      -8 FRAA     53 FRAA
179 FRAR     309 FRAR     183 FRAR    206 FRAR
--           --           --          30 PRAR
515 BRAR     548 BRAR     528 BRAR    509 BRAR
213.7 bws    215.0 bws    189.5 bws   209.8 bws
34.8 fws     51.9 fws     28.2 fws    19.1 fws (7.5 pws)

In moving out to a ten-year period, Sisler’s batting rate stats suffer the most, as two post-infection years are included in his ten. Medwick’s rates also drop significantly, while Johnson’s drop slightly and Averill’s don’t drop at all.

Both WARP and WS agree that Averill is the best of the group. His slight advantage in playing time leads him to top the group in both bws and BRAR, and his fielding value is far ahead of everyone else’s. Interestingly, Sisler actually has the highest rate of bws/g of the four, though his playing time trails the other three.

The striking feature of this set of figures is Johnson’s performance in win shares. He appears about equal to the others in OPS+, EQA, BRAR and playing time, but he lags far behind the others in bws. I find it difficult to credit that Johnson really could have been 10% less productive offensively over a ten year period than players with offensive stats so nearly identical to his as Medwick’s, Averill’s, and Sisler’s.

WARP1 orders their primes
Averill 857 B+FRAR
Sisler 745 B+P+FRAR
Johnson 711
Medwick 694

WS orders their primes
Averill 266.9
Medwick 248.5
Sisler 236.4
Johnson 217.7

Looking at peak and prime together, I would say that WARP and WS together show Averill as the top player. They agree he had the second-best peak, and they agree he had the best prime. In WARP his competition is Sisler, in WS Medwick. The difference between the two systems’ evaluations of these two players is primarily in defense: all systems of measure surveyed (OPS+, EQA, BRAR, and bws) see them as nearly identical hitters over 5-year and 10-year stretches. WARP likes Sisler’s defense and places considerable weight on first-base defense in his era. WS likes Medwick’s defense and doesn’t place much weight on first-base defense in any case. Johnson clearly lags in a peak/prime assessment in comparison to this group, but WARP has him close while WS has him a distant fourth.

Career comparisons will appear in the next post.

Now, career. This comparison is trickier, of course, because their career lengths aren’t the same and there’s the big issue of minor-league credit for Averill. But here are the numbers:

Medwick      Averill      Johnson        Sisler   
8142 PA      7215 PA      8047 PA        9013 PA
134 OPS+     133 OPS+     138 OPS        124 OPS+
1984 g       1668 g       1863 g         2055 g
.300 EQA     .303 EQA     .308 EQA       .291 EQA
1 FRAA       7 FRAA       14 FRAA        2 FRAA
254 FRAR     317 FRAR     256 FRAR       215 FRAR
--           --           --             33 PRAR
624 BRAR     560 BRAR     677 BRAR       596 BRAR
267.4 bws    223.9 bws    250.8 bws      260.2 bws
44.4 fws     54.5 fws     36.6 fws       24.6 fws (8.3 pws)
1899 adj. g# 1761 adj. g  1848 adj. g    2075 adj. g
311.8 ws     308.4 adj.ws 287.4 ws       300 adj. ws
             (add 100 RAR)               (add 25 RAR)

#adjusted games makes a variety of adjustments to career. For all players, it converts pinch-hitting appearances into games played at a rate of 1 game per 4.2 ph apps. This adjusts game totals downwards for players, like Medwick, with sig. Ph apps in late career. For Averill, adj. games also adds in his 1928 PCL season, as converted by Brent. For Sisler, adj. games prorates his 1918-19 seasons to 154 games. The adj. ws totals for Averill and Sisler reflect these changes. Medwicks’s and Johnson’s totals are unaffected by the playing time adjustment.

Let’s start with the totals for each system.

WARP1 (adj.)
Averill 977 RAR (877 unadj.)
Johnson 933
Medwick 878
Sisler 869 (844 unadj.)

WS
Medwick 311.8
Averill 308.4 (278.4 unadj.)
Sisler 300.0 (293.1 unadj.)
Johnson 287.4

Whether one gives Averill a minor-league season makes a very big difference in his win-share career evaluation. With it, he probably has the best career value by win shares, since he has almost the same career ws total as Medwick in a season’s fewer games. In WARP1, he has the best career with the minor-league season, the second-best career without it, since his career WARP1 total is about equal to Medwick’s and Sisler’s in fewer games than either.

It’s interesting that Johnson does not make up more ground on Medwick in win shares. Johnson’s career rates rise slightly going from his first 10 years to his full career in 404 games, while Medwick’s OPS+ drops by 8 and his EQA drops by 7 as he adds 423 games to his career length. Yet Johnson adds 69.7 win shares over this stretch while Medwick adds 63.3. Sisler, in his decline adds about the same number of win shares (65.4), but takes 700 games to do it, while his OPS+ drops by 18 and his EQA drops by 14. Ouch.

I give Averill a year of minor-league credit, and I dock Medwick and Johnson a bit for their WWII years in a way not reflected in these numbers, so I see Averill’s career value as the best of the group of four. Johnson’s career value in WS just seems unbelievably low in relation to his production: I favor the WARP evaluation over the WS one, so I place him second in career value. Medwick is third, and Sisler fourth, as he trails Medwick in both WARP and ws. Clearly his last 700 games do little to improve his HoM case.

Since a combination of peak/prime in WARP and ws also favors Averill, the outcome of this study is to see Averill as the top player in this set. Sorting out the other three is harder, as the two systems return such different views of the three.

I no longer think that win shares overrates Medwick, but I think that it underrates Sisler by undervaluing his defense, and I think it underrates Johnson because of the bad teams effect. I give Sisler a slight edge over Medwick, and Medwick a slight edge over Johnson, in both cases because their peak edge over Johnson is greater than his career edge over them.

So, in my own rankings, Averill moves ahead of Sisler, who drops a bit, and Johnson moves much closer to Medwick. I’d hate to have to draw the all-time in/out line between any of these guys.

Thanks Chris. I hate to do it too...but that's why they pay us the big bukcs fo rthis, right? :)

Good job, Chris Cobb.

**Sisler’s PA and G for 1918-19 have been prorated to 154-game seasons to calculate rate stats for the period. WARP and WS totals have also been adjusted.

Are the 5-yr and 10-yr rate stats all weighted means of the single-season rates?

(I hope to present Oms, Brown, and the 19th century hitters in comparison to this group next week).

Remember, everyone hopes you will do an infinite amount of work on MLEs by 1962 if not earlier.

Are the 5-yr and 10-yr rate stats all weighted means of the single-season rates?

Yes. It was my inference that weighted means would be accurate for these stats, unlike DERA+ or ERA+. If my inference was incorrect, the value of those numbers would need to be rethought :-).

Remember, everyone hopes you will do an infinite amount of work on MLEs by 1962 if not earlier.

Well, we'll see. Infinite amount of work, finite amount of time. . . . I'm hoping that my work on Brown and Oms will help people to use the MLEs that we already have more effectively.

WS
Medwick 311.8
Averill 308.4 (278.4 unadj.)
Sisler 300.0 (293.1 unadj.)
Johnson 287.4

But...

WS/162 Games
Averill 28.37
Medwick 26.60
Johnson 25.19
Sisler 23.42

Clearly, Averill is the king of that group, while Sisler is on the outside looking in, relatively speaking.

Anybody not using WS/162 Games in conjunction with straight WS will not get a true picture of quality.

Chris C.,

How much defensive value do you believe WS ought to reapportion to a deadball 1B? And from what positions would that credit be drawn? 2B? RF?

Also what seasons throughout baseball history should be so reapportioned?

Is it OK if I rejoin? This was a busy summer for me and I'll be glad to be invited back.

Nice to see you back, Ardo!

How much defensive value do you believe WS ought to reapportion to a deadball 1B? And from what positions would that credit be drawn? 2B? RF?

Also what seasons throughout baseball history should be so reapportioned?</i>

I increase first-base fielding win shares by 60% up to 1890, by 40% 1890-1919, and 20% 1920-1929.

These are very large percentages, but considering that first basemen are seldom earning more than three win shares, the increase is never more than 2. It is approximately twice the percentage adjustment I give to other infielders.

I don't try to balance win shares adjustments at the team level, but conceptually these win shares are coming from the pitchers.

Yesterday I emailed Clay Davenport for an explanation of his WARP1 to WARP2/3 discounts and I got an answer. I am in a bit of a rush right now but I will paste the email here sometime either tonight or tomorrow morning.

However, he says that the discount rate for Burns was 88% instead of the 73.4% (or you could say 12% instead of 26.6) stated buy Doc C. above. How did you get the 73.4 percent Doc?

I don't try to balance win shares adjustments at the team level, but conceptually these win shares are coming from the pitchers.

[This whole line of discussion may have happened during a HOM thread of another generation, and if so, just direct me there instead of answering this question...which by the way is really picayune and probably doesn't make much difference in most cases anyway....]

An average team throughout history is supposed to get about 40 WS as a fielding unit, while the remaining 400 WS are divvied up among hitters and pitchers at some pre-determined percentage. I know it's just a few FWS, but now you're upping the number for fielders to, say, 42 which shifts that pre-determined percentage and moves everyone off of equal footing.

Don't you "have to" take it out of the hide of some fielder to make it conceptually balanced/correct?

Burns has 96.3 WARP1s.
Burns has 70.6 WARP3s.

70.6 / 96.3 = 73.3%

No matter what Clay claims the discount rate for the league might be, Burns ends up with 73.3% the WARPs he notched in iteration 1.

If the discount were 12%, Burns would ultimately end up at 84.7 WARP3.

Granted, I could be misunderstanding where and how that discount is applied and what effect it therefore has on Burns's ultimate WARPage, but the bottom line is that Burns suffers pretty mightily by the hand of WARP3.

One other point, however, is that even if 12% is the discount, I have to ask: Is that reasonable?

I don't think it probably is. That suggests that George Burns, Zack Wheat, Pete Alexander, and friends were playing double-A baseball in the teens, relative to Kiner (whose leagues are roughly equivalent to today's leagues by WARP). I've got quibbles with that on all kinds of levels, including but not limited to:
1) a pennant's a pennant's a pennant
2) they were the highest major leagues of their time
3) if they had the training and equipement of 2005 via time machine
4) these same guys were around in the 1920s and some in the 1930s when play appears to improve, etc etc etc....

Dr. Chaleeko, I can't see how a 12% discount would be equivalent to playing AA ball.

12% of 96 WARP (wins) is 11.5 wins, or about 1 win per year.

1 batter win per year is about 10 runs, or turning 13-14 outs into hits, which is .022 in batting average.

No way a player hitting .272 in AA would only drop to .250 in the majors; most studies show a much bigger difference than this. That is closer to the triple-A-to-majors difference.

An average team throughout history is supposed to get about 40 WS as a fielding unit, while the remaining 400 WS are divvied up among hitters and pitchers at some pre-determined percentage. I know it's just a few FWS, but now you're upping the number for fielders to, say, 42 which shifts that pre-determined percentage and moves everyone off of equal footing.

Don't you "have to" take it out of the hide of some fielder to make it conceptually balanced/correct?

Not if your premise is that the win shares system basically makes a mistake in its assessment of the pre-lively-ball game by fixing the pitching/fielding split at 67%/33% of the defensive pie. You can tinker with that built-in feature of the system just as you can tinker with any other built-in feature. Win shares is obviously way wrong for pre-1893 pitchers: it is less obviously but still significantly wrong for pre-1920 pitchers, in my view.

I figure that the actual pitching/fielding split reaches 67/33 at about 1930: before that I figure it gradually shifts towards an approximately 33/67 split in the 1870s.

If I were trying to construct my own comprehensive metric or study success at a team level I would need to balance everything out. Since I'm trying only to evaluate some great players fairly, I build some simple fixes into the existing system. I don't try to move win shares back and forth between pitchers and fielders on individual teams: I calculate my own pitching win shares in a simpler way that deals with fielding differently, and I give fielders a percentage boost.

It's not rocket-science sabermetrics, but I like the results, generally.

Doc,

I looked up Burns DT card and you are right, it has 70.6 as the WARP3 total. However, Davenport, in his email to me gave Burns over 80 WARP3.

Here is the letter:

Mark,

There are two seperate conversions going on to get from WARP1 to
WARP2, and a third one to get to WARP3.

First of, let me say that I don't understand how you came to the 73%
number. I see

Burns 96.3 warp1, 84.8 warp3, warp3/warp1 = .881
Kiner 74.9 73.2 .977

That's 10% less, not 27%. Anyway, to the reasons.

Each season in history has been rated for difficulty. The approach is
similar to Cramer's from the Hidden Game, but with a couple of
significant differences - the biggest one being that the time between
between the two leagues is part of the weighting function. So where
Cramer counted matched Ted Williams seasons from 1939 and 1957 at full
value, I would count them at just 1/19 of their value, because there
is an 18 year difference, and I use (time difference+1) in the
weighting function.

I then have an estimate of what each league should have been rated,
using the slope of difficulty taken from the integrated era, 1950 to
2000. Projecting that line backwards in time, there is a steep dropoff
during WWII and prior to 1940.

Taking a year from the middle of Burns' career, 1917, the year is
rated at 1.414 (the scale is proportional EQR), and 1917 has a par
value (from the regression equation) of 1.572. That means that 100
runs in 1917 are only treated as being worth 89.9 runs
(1.141/1.572=.899) , adjusted for difficulty. All of Burns' run
values, and hence his win values, are going to drop by about 10%. The
diffference between AAA and the majors today is around 25%, so the
difference in Burns' day (which I think is a result as much of not
having organized farm systems as to segregation) is not as large as a
full level today, yet still quite noticeable. If you go to pages like
this one (you may have to manually change the year in the URL):

http://www.baseballprospectus.com/statistics/eqa1917.php

and look at the league totals, the last column shows the difficulty
rating I used for that year, relative to the par. I'm not sure why
that page says .901 instead of .899; I'm not sure how long ago I made
that page, and I may have made some (obviously small) change.

Kiner's era is much closer to the standard. 1950 rates at 1.764, with
a par of 1.773, so each run is still worth .995 runs after difficulty
is factored in - and again the website is .002 higher than what I have
on my system right now.

That is far and away the largest difference; had Burns been an
infielder, the second effect would have been more important. The
allocation of defensive runs between pitching and fielding has changed
over time, with pitching getting a larger share over time.That will
show up in the values for defensive runs and in WARP-1. Warp-2
standardizes the defensive values of positions. In Burns' career, an
average left fielder was worth 27.7 runs above replacement, per 162
games; in Kiner's career, they were worth 17.6. Those both get
adjusted to the "recent" value of 15.0. That is clearly going to cost
Burns more than Kiner.

The third factor, going from warp-2 to warp-3, is only about playing
time, expanding the 154-game season (and in Burns' case, a 128- and a
140-game season, for 1918 and 1919) out to 162 games. Slight edge to
Burns there.

Clay


He seems to indicate that AAA baseball is actually 75% Of MLB. I take it this is in wins or runs, not in AVG/SLG.