I plan on looking into Tiant in detail this year

I think you only have 9 players listed as the "Top 10 Returning Players". Looking at the 2018 results, I believe that Jorge Posada is #10.

EDIT: Oh, and Thank you very much for posting this! I don't want the first post to be a nitpick.

fixed

Might as well get the formality over with now. Have made a few changes since the 2018 results thread post.

Disclosures: am going with Seamheads for Negro Leaguers. Preference for BBRef WAR with some influence of OPS+ and ERA+ for the rest. Am valuing hitting prowess at C, SS, 2B, CF a bit extra. Being best available candidate at your position helps also. Still trying to sort out peak vs. longevity, but often favoring the latter. Fine with giving Negro League credit, not presently giving credit or debit for war, injury, illness, postseason play, or minor league service. Not systematically adjusting for season length, but am giving minimal non-systematic extra emphasis for pre-1961 players. Am currently treating 19th century pitchers pretty much equally as post-1900, but for now tending to discount AA, NA, and UA stats as possibly suspect. Not taken with giving relievers a lot of emphasis. Will dock 1st year candidates who bet on games, threw games, impeded players of color, were caught using PEDs post-2005 (Manny, ARod), and likely used pre-2005 if it looks like they'll get an immediate free pass by BBWAA HoF voters (IRod, Ortiz, Pettitte).

As noted, I am exercising my one-year boycott of Pettitte. He would rank between Rivera and Bridges otherwise.

1. Jim McCormick. Best WAR for starters not in by a mile; even removing all his UA-earned WAR leaves him a point up on Tiant. Short career, but played in NL except for one UA season.
2. Luis Tiant. Best WAR for non-19th century starters.
3. Roy Halladay. Not ranking him quite as high as McCormick and Tiant, but extremely qualified.
4. Buddy Bell. Best WAR at 3B. Currently inclined to trust the metric for him.
5. Andruw Jones. Best CF WAR. Close between him and Bell for me.
6. Jeff Kent. Best WAR at a middle infield position and hit well, can't in good conscience rank him below Helton, Sosa, or Johnson.
7. Todd Helton. Excellent WAR and easily the best qualified 1B.
8. Bob Johnson. Best WAR among available LFs.
9. Wally Schang. Among best C WAR, also hit well.
10. Vic Willis. Good WAR.
11. Sammy Sosa. Better WAR than I remembered. Happy to give him some benefit of the doubt given his treatment by the BBWAA.
12. Ben Taylor. Best NGL position player per Seamheads.
13. Dick Redding. Best NGL pitcher per Seamheads.
14. Vern Stephens. I value hitting at a premium position highly, so I'm ranking him here.
15. Kenny Lofton. Not as much hitting as I'd like, but lots of WAR at a premium position.

16-40. Tommy John, Sal Bando, Mickey Welch, Urban Shocker, Ernie Lombardi, Thurman Munson, Mariano Rivera, Tommy Bridges, Joe Tinker, Jim Fregosi, Bobby Bonds, John Olerud, Luis Aparicio, Bert Campaneris, Johan Santana, Gavvy Cravath, Bob Elliott, Tony Lazzeri, Jorge Posada, Jose Cruz, Fred McGriff, Jack Quinn, Harry Hooper, Lance Berkman, Willie Davis.

Various comments. Phil Rizzuto doesn't do much for me (low OPS+, low WAR, short career) -- and prefer him to Omar Vizquel. Bucky Walters also doesn't sufficiently impress me (good but not top of the line WAR or career ERA+). I do not value relief pitching very highly; Mariano Rivera does have a terrific OPS+ but his WAR number places him 21st on my ballot. Am still struggling with where to place Santana and Cravath, but for now am placing them 30th and 31st.

Ranking by position:

1B. Helton, Taylor, Olerud, McGriff
2B. Kent, Lazzeri, Evers, Phillips
SS. Stephens, Tinker, Fregosi, Aparicio, Campaneris
3B. Bell, Bando, Elliott, Cey, Ventura
LF. B. Johnson, J. Cruz, Berkman, J. Gonzalez
CF. A. Jones, Lofton, W. Davis, Lemon
RF. Sosa, Bonds, Cravath, Hooper, Rice
C. Schang, Lombardi, Munson, Posada, Tenace
P. McCormick, Tiant, Halliday, Willis, Redding, John, M. Welch, Shocker, Rivera, [Pettitte], Bridges, Santana, Quinn, Cicotte

Sorry, can't count. Should read:

"Mariano Rivera does have a terrific OPS+ but his WAR number places him 22nd on my ballot."

Wouldn't Mariano Rivera have an ERA+ number instead of OPS+?

@6: yes. My mistake. Thanks for the correction.

Luis Tiant - year by year
1959 - age 18 - Mexican League, (5-19, 5.92 ERA)
1960 - age 19 - playing in Mexico City 17-7 but with a 4.65 ERA. Assuming a high run scoring environment at altitude
1961 - age 20 - 12 wins in a shortened Mexican League season due to visa issues. Cleveland purchases his contract at the end of the season
1962 - age 21 - pitches well for Charleston in the Eastern League
1963 - age 22 - likely the best pitcher in the Carolina League, 204 IP, 2.56 ERA, 9.1 K/9, 3.6 BB/9, no-hitter
1964 - age 23 - promoted to AAA Portland, pitches 137 innings with a 2.04 ERA, 15-1 record, 10.1 K/9, 2.1 BB/9 then gets promoted to Cleveland where he pitches an additional 127 innings at 128 ERA+. Fellow Cuban Tony Oliva wins ROY. They are among a couple dozen Cubans in MLB.

Would Tiant have made the big leagues sooner if he hadn't taken this route from Cuba to the US? Sam McDowell was ahead of him in the prospect pecking order and they were 1-2 in pitcher WAR for the 1964 Cleveland team.

1965 - age 24 - after pitching 264 innings in 1964 Tiant is plagued with arm soreness in 1965. Still manages a 99 ERA+ in 196 innings with 11 games in relief
1966 - age 25 - starts the season with 3 consecutive shutouts but is demoted to the bullpen in July. Leads the league in shutouts (5) despite only getting 16 starts. If you are using Dan R's pitching WAR his 1966 year does not show up due to the time spent in the bullpen. I had to add it back in manually.
1967 - age 26 - solid season, 213.2 IP at 121 ERA+
1968 - age 27 - awesome season, 258.1 IP, 186 ERA+ (best in league), 9 shutouts, started the All Star Game, finishes 6th in MMP voting and third among pitchers to Gibson and McLain.
1969 - age 28 - Cleveland is last in the AL, Tiant pitches 249 innings with a 101 ERA+. Peripherals all degrade from 1968. Traded in the offseason to the Twins for Dean Chance and future HoM Graig Nettles.

1970 - age 29 - Won his first 6 decisions but leaves due to a nagging shoulder injury May 28. Diagnosed with a broken bone in his shoulder and sits out for 10 weeks. Ineffective the last few weeks of the season.
1971 - age 30 - pulls a muscle during spring training and Calvin Griffith releases Tiant in one of the worst transactions in Twins history. Gets a 30 day trial with Atlanta but they decline to purchase his contract. Signs a minor league deal with the Red Sox and is effective. Promoted to the Red Sox where he plays out the season with a 4.85 ERA but a 3.63 FIP.
1972 - age 31 - begins the season in the bullpen, back into the rotation in August and over a period of 10 starts is 9-1 with six shutouts and a 0.82 ERA, all nine victories being complete games. 179 innings with a 169 ERA+. Wins the ERA title and MLB comeback player award.

1971-1972 are also not included in Dan R pitching WAR due to the relief innings

1973 - age 32 - 272 innings @ 120 ERA+
1974 - age 33 - 311 innings @ 133 ERA+, gets MMP votes, plays in All-Star Game
1975 - age 34 - 260 innings @ 103 ERA+, gets to see his parents for the first time in 15 years, pitches terrific in the postseason 34IP, 3CG, 2.65 ERA. Began the 1975 World Series with a five-hit shutout of the Cincinnati Reds. In Game Four threw 163 pitches in a complete game 5-4 win. Sent back out for Game 6 after a three day rain delay he gave up 6 runs in 7 innings.
1976 - age 35 - 279 IP @ 129 ERA+, gets MMP votes, plays in All-Star Game
1977 - age 36 - contract holdout in the spring, 188.2 IP @ 100 ERA+
1978 - age 37 - last season for Boston and last great season, 212IP @ 126 ERA+
1979 - age 38 - signed by the Yankees, 195IP @ 104 ERA+, last season that contributes to his HoM candidacy
1980 - age 39 - 122IP at replacement level
1981 - age 40 - pitches fairly well in AAA Portland but ineffective for Pittsburgh
1982 - age 41 - 30 bad innings for the Angels

Quotes on "intangibles" from his SABR bio
http://sabr.org/bioproj/person/2212deaf

When the Twins released Tiant, their longtime publicist Tom Mee called the scene in the locker room as Luis said goodbye to his teammates "the most forlorn experience I've ever had in baseball."

After the 1972 season, Red Sox pitcher John Curtis wrote a newspaper story about trying to explain to his wife why he loved Luis Tiant. Dwight Evans would later say, "Unless you've played with him, you can't understand what Luis means to a team."

In the offseason, the Red Sox offered the 38-year-old Tiant only a one-year contract, allowing Luis to sign with the New York Yankees for two years, plus a 10-year deal as a scout. Dwight Evans was devastated at management's ignorance of what Luis meant to the team. Carl Yastrzemski says he cried when he heard the news: "They tore out our heart and soul."

From his Wikipedia entry

Tiant recalled that at Charleston, "I couldn't speak very good English but I understand racism. They treated me like a dog, but when I got to Portland, I didn't have any problems

Tiant broke through in 1968, after he altered his delivery so that he turned away from home plate during his motion, in effect creating a hesitation pitch. According to Tiant, the new motion was a response to a drop in his velocity due to a shoulder blade injury.

excelled in the Venezuelan Professional Baseball League in parts of five seasons spanning 1966–1982, while collecting 37 victories, 29 complete games, a 2.27 ERA, and a no-hitter in 1971

Let's see...

1) Roy Halladay - RIP. Comparable to Newhouser against stronger competition. Also "comparable-but-better-than" to Stieb.
2) Mariano Rivera.

I could write a 1000-word essay on Rivera, but won't. Relief pitching is a specialized role that is less valuable than starting, but to excel in it for as long, by as much, as Rivera puts him here.

3) Wally Schang
4) Adolfo Luque
5) Todd Helton - Little things add up. Helton is akin to Will Clark with a shallower peak/longer prime: a very good on-base offensive threat, defender and base runner.
6) Hilton Smith
7) Jorge Posada
8) Ben Taylor
9) Sammy Sosa
10) Jeff Kent
11) Luis Tiant - Superior to exact contemporary (and twice teammate) Tommy John, with less bulk but a much sharper peak.
12) Johnny Evers
13) Johan Santana
14) Luke Easter
15) Buddy Bell

First ten off: Tommy John, Nomar Garciaparra, Dick Redding, Andruw Jones, Lee Smith, Thurman Munson, Lance Berkman, Lou Brock, Fred McGriff, Andy Pettite.

Lance Berkman compares to "Willie Stargell minus two seasons at Stargell's career averages" and, in a different value format, his contemporary Brian Giles. Hall of Very Good for me.

Andy Pettite is also off-ballot. I'm not boycotting him, but if Pettite, why not Mark Buehrle? He doesn't stand out enough from a gaggle of pitchers we aren't taking seriously.

Lance Berkman compares to "Willie Stargell minus two seasons at Stargell's career averages" and, in a different value format, his contemporary Brian Giles. Hall of Very Good for me.

You have Berkman listed at #22, which puts him in the lower rungs of PHOM territory : )

Andy Pettite is also off-ballot. I'm not boycotting him, but if Pettite, why not Mark Buehrle? He doesn't stand out enough from a gaggle of pitchers we aren't taking seriously.

While Pettitte and Buehrle are comparable in regular season WAR in Baseball-Reference and Baseball Gauge, Andy easily bests Mark in Kiko's W-L, Fangraphs FIP, Baseball Prospectus DRA, WPA, and adds a full season + of regular season level play in the post-season.

And I'd throw in that Buehrle will be deserving of consideration when he joins the ballot. I agree with Bleed the Freak that Pettitte should be ahead of him though.

As an aside, I'd be interested in hearing the argument for Dolf Luque from a Luque voter. He's on a list of players I want to take a closer look at before the 2019 ballot and you seem like a good person to hear the pro-Luque case from. Though if any others have thoughts on the matter, I am interested in those as well.

Comparables for new additions after doing my first calculations

Mariano Rivera = Stan Coveleski great rate but fewer innings
Roy Halladay = Bret Saberhagen+ if you rearrange the years
Andy Pettitte = Chuck Finley + postseason
Lance Berkman = Roy White
Todd Helton = Norm Cash but with less bat than Cash after the standard deviation adjustments. All the modern players had high standard deviations in run scoring.
Roy Oswalt = Dave Stieb - will make my ballot
Miguel Tejada = Jay Bell or Tony Fernandez

2019 Prelim

My list is loaded with pitchers. This is because I target ~30% pitchers in my pHoM and the current HoM makeup is at least 6 pitchers short of that.

1) Roy Halladay - PHoM 2019
2) Tommy Bridges - PHoM 1958
3) Mariano Rivera - haven't computed his postseason value yet - PHoM 2019
4) Luis Tiant - PHoM 1991
5) Johan Santana - PHoM 2018
6) Phil Rizzuto - top position player available, gets WWII credit, PHoM 1967
7) Gavy Cravath - best available OF, 154 game seasons, low run scoring environment (low STDEV), several seasons minor league credit, PHoM 1927
8) Urban Shocker - WWI credit, good hitter, PHoM 1968
9) Tommy John - PHoM 1995
10) Roy Oswalt - could be my #3 PHoM in 2019. It is between Oswalt, Giles and Posada
11) Bus Clarkson - Mexican League, Minor League and Negro League credit, PHoM 1967
12) Bucky Walters - PHoM 1972
13) Bob Johnson - PCL credit, PHoM 1986
14) Bert Campaneris - PHoM 1991
15) Hilton Smith - best available NGL pitcher, PHoM 1987

16) Ben Taylor - PHoM 1973
17) Dave Bancroft - PHoM 1976
18) Brian Giles
19) Wally Schang - best C available, PHoM 1987
20) Norm Cash - PHoM 1997
21) Kevin Appier - PHoM 2009
22) Don Newcombe - PHoM 2004
23) Jorge Posada
24) Johnny Pesky - PHoM 2004
25) Andy Pettitte

26-30) Jeff Kent, Wilbur Cooper, Sammy Sosa, Babe Adams, Burleigh Grimes
31-35) Dave Concepcion, Dick Redding, Tommy Leach, Dizzy Trout, Dwight Gooden
36) Kenny Lofton
61) Bobby Bonds
64) Buddy Bell
66) Todd Helton
82) Andruw Jones
92) Lance Berkman
104) Miguel Tejada

Tommy John vs. Luis Tiant, year by year:

Keep in mind that John is two and a half years younger than Tiant. In high school, he was a baseball and basketball star in Terre Haute, Indiana. He chose to pursue baseball and signed with Cleveland.

1961 (age 18): Pitches 88 innings in Class D (in effect, today's rookie ball). 10-4 record with 6 BB/9 and 10 K/9, crazy in hindsight.
1962 (19): Starts in A-ball, promoted aggressively to AAA towards season's end. 8-10, 4.06.
1963 (20): Big step forward. 15-10, 2.60 in AA and AAA. By now, BB/9 is down to 2.3 and K/9 to 4.8. Allows more than a hit per inning, but only 7 HR in 197 IP. Gets cup of coffee with Indians, allowing 10 runs in 20.1 IP.
1964 (21): 6-6 in AAA, with a worse ERA but better peripherals than 1963. Promoted in mid-season for good - did not pitch in minor leagues again until 1985. Went 2-9, 3.91 as a swingman. Advantage Tiant.
1965 (22): Foolishly traded with Tommie Agee and John Romano to the White Sox in a three-team deal to reacquire Rocky Colavito. Begins in the bullpen, but makes 27 starts. 14-7, 3.09. It's close but John.
1966 (23): Full-season starter for the first time. 14-11, 2.62. Ties Tiant (!) for the AL lead with 5 shutouts, while pitching about 70 more innings. John.
1967 (24): 10-13, 2.47 - remember it's a pitchers' era and the White Sox were a poor-hitting team. This year John misses time and Tiant pitches well over a full season. Tiant.
1968 (25): Again unable to last a full year, making 25 starts. Great when he pitched (10-5, 1.98), but Tiant prevails.
1969 (26): 9-11, 3.25, but a full workload of 33 starts and 232 innings. John.
1970 (27): 12-17, 3.27. More W-L hard luck. 37 starts and 269 IP, a big bump in workload. First of Tiant's two "wilderness years". John.
1971 (28): 13-16, 3.61. 35 starts, but 40 fewer IP than 1970 and worse peripherals. Advantage John.

On December 2, 1971, John was traded to the Dodgers heads-up for Dick Allen. We'll pick up the story there with John leading 5-3.

On the newcomers:

Roy Halladay - Easy selection. Not a great comp, but he ranks somewhere around Juan Marichal for me.

Mariano Rivera - Second on my ballot. Unlikely to move from that slot.

Lance Berkman - I think Ardo makes a good comparison with Berkman and Stargell in terms of their bats - Berkman hit .293/.406/.537 over 7800 PAs, while Stargell hit .282/.360/.529 over 9000 PA. But I'd say that I think Berkman more than makes up the gap in PAs with his better glove. No, Berkman wasn't a great fielder, but he was pretty average in the corners and at first, and he was actually okay in the year he spent in center in his youth. Stargell was a pretty bad fielder his entire career. Berkman will more than likely be in my final 'elect me' slot, although I'm giving hard looks at Tommy John, Kenny Lofton, and Babe Adams as possibilities there as well.

Todd Helton - This year's biggest problem for me. I think he's worthy. It's possible I could change my mind in the future, but that's where I lean now. The problem is how high up the ballot should he go. Before any adjustments, I have him about tied with Jeff Kent for 7-8 range. But two things give me pause. First are his excellent RE24 numbers that are pumping up his batting record. I'm not sure RE24 adjusts correctly for extreme run enivornments, which would explain Helton's very high scores (as well as Giambi's as well, who also sees a huge boost based on the metric). The second thing is his defensive numbers. I believe Helton was a fine defender. - both the numbers and the eye test agree he was quite good. My main concern is how valuable an elite defensive first baseman in 2000-2002 Coors Field was. Until I figure those two out, he'll remain just off ballot.

Andy Pettitte - The question here is how to value postseason pitching. Pettitte is all about volume - 1.5 seasons of postseason innings. But those innings are pretty much identical to his career averages, and it's not as if he was notibly integral to any particular series victory. He pitched well in a lot of games, but not extrordinarily well.

So the question is how to scale value in the postseason. I don't think value scales equally compared to the regular season - I'd evaluate Curt Schilling as a more valuable postseason pitcher than Pettitte, because he was significantly better and more pivotal for his teams, despite having half the innings. Christy Mathewson is better than Pettitte as well - 170 innings is a big difference, but Mathewson was only pitching in the World Series and he was so much better on a rate basis than Pettitte.

But then what about Sandy Koufax? Are 50 excellent World Series innings better than 280 decent innings split between the World Series, LCS, and LDS? The Dodgers likely don't win the '65 Series without Koufax, and he's a big part of the '63 win as well. I'm not sure you can say that about Pettitte for any series win, World Series or otherwise. Mickey Lolich is a step down for Koufax, but he practically won the '68 Series himself, and pitched quite well in the '72 ALCS. Are his 50 innings more valuable than Pettitte's? We're getting down to a point where I'm not sure how to go about it. Maybe some sort of WPA type system that adjusts for the effect on a championship?

Any way, back to Pettitte in particular. Right now, he's in the 16-20 range for me. Postseason adjustments could change that if I give him more credit.

Roy Oswalt - Bret Saberhangen is a really good comp. Saberhagen probably was better at their respective bests and pitched a season-ish longer, but Oswalt has his advantages as well.

My ballot at this point looks something like this
1. Roy Halladay
2. Mariano Rivera
3. Lance Berkman
4. Tommy John
5. Kenny Lofton
6. Babe Adams
7. Mickey Lolich
8. Jeff Kent
9. Kiki Cuyler
10. Bert Campaneris
11. Ben Taylor
12. Bob Johnson
13. Robin Ventura
14. Willie Davis
15. Roy Oswalt
----
16. Todd Helton
17. Hack Wilson
18. Andy Pettitte
19. Jim Kaat
20. Trevor Hoffman
21. Dwight Gooden
22. Bobby Bonds
23. Joe Tinker
24. Dolph Camilli
25. Hugh Duffy

I have a big cluster of pitchers just after that to clear out (Derringer, Bridges, Tiant, Koosman, Santana, Guidry), but that's not my biggest priority. For now, I'm focusing on Helton, Pettitte, and Babe Adams as candidates worthy of long looks.

My one Tiant question - does anyone know what is up with Baseball Prospectus's numbers for him? 15.4 WAR for his career, a -5.1 (!) WAR season in 1975. BPro for pitchers is a bit eccentric usually, but this seems more like a glitch than anything else. I'm not using BPro to evaluate him at the moment, but I'd like to know if anyone has any insight into those numbers.

Tommy John vs. Luis Tiant, year by year (Part 2):

1972 (29): 11-5, 2.89 as the Dodgers' third starter behind Don Sutton (who had arguably his best season) and Claude Osteen. Tiant wins his second AL ERA title. Advantage Tiant.
1973 (30): 16-7, 3.10. Bumped to 4th starter with the arrival of Andy Messersmith. Adjusting for home ballpark, Tiant is the better pitcher and works 55 more innings. Tiant.
1974 (31): Back to #3 after off-season trade of Osteen to the Astros for HoM CF Jim Wynn. On July 17 (13-3, 2.59, leading the NL in wins), he gets hurt. His season - and, apparently, career - is over. Advantage Tiant, who has his best season in Boston.
1975 (32): Undergoes his eponymous surgery with Frank Jobe (actually two surgeries, on 9/25/74 and 12/15/74). By August, he's able to pitch batting practice. Makes five appearances, working his way up from three to seven innings, in the Arizona Fall League. Another win for World Series hero Tiant.
1976 (33): 10-10, 3.09. Wins Comeback Player of the Year, just as Tiant had four years prior. Tiant better by both rate and bulk.
1977 (34): 20-7, 2.78. First 20-win season. Runner-up to Steve Carlton in NL Cy Young voting. Tiant shows eroding skills at age 36. Advantage John for the first time since 1971.
1978 (35): 17-10, 3.30. A step back from 1977. Advantage Tiant for the final time.
1979 (36): Signs with Yankees as a free agent, as does Tiant (!!). Triumphant season: 21-9, 2.96, career-high 276 IP. Second in AL Cy Young voting to Mike Flanagan (who went 23-9; count the winz!), although John had a far better season. Close contest between John, rotation-mate Ron Guidry, Dennis Eckersley, and Jerry Koosman for best pitcher in AL. John.
1980 (37): 22-9, 3.43. Starts Game 3 of ALCS behind Guidry and AL ERA titlist Rudy May. Tiant pitches poorly as Yankees' 4th starter in his last full season. John.
1981 (38): 9-8, 2.63. Deserves strike credit and postseason credit (26 IP, 7 RA). John.
1982 (39): 14-12, 3.69. August rental to Angels - Yankees get back Dennis Rasmussen. Splits two starts in Angels' ALCS loss to Milwaukee. Tiant retires. John.

From then on, in his 40s, John pitched 1000.2 IP with an ERA+ of 92, above replacement but below average.

The final score is John 12, Tiant 8 - but Tiant's wins, on balance, were by wider margins.

Jaack ballot representation by position

P 40%
1B/OF 44%
2B/SS/3B/C 16%

OF is our most represented position already. C and 3B are our least represented positions.

C and 3B are our least represented positions.

There are a couple of worthy 3B, but I'm having a hard time coming up with a deserving catcher.

Schang isn't really borderline, IMHO. Other than him, who else is there?

Nice analysis, Ardo! It's a nice way to think about the comparison between those two. Personally, I value the extra length of John's career, but I don't think there's necessarily a "right" answer between the two of them and I'm fairly sure they'll both be on my ballot.

As for the BPro numbers in #19 - wow! That has to be some kind of error in their formula. Both 1975 and 1969 are bizarre - Tiant wasn't great in either year, but he was basically a league-average innings-eater both seasons and BPro rates those seasons at -5.1 (1975) and -2.3 (1969) WARP!? His FIP was quite high in 1969 (4.83 vs. ERA of 3.71), so I could see getting that season down to "below average", but not two full wins below replacement level, and his 1975 FIP was better than his ERA (3.62 vs. 4.02). Eyeballing the surrounding years, it looks like BPro is using a higher replacement level than BB-Ref (& Fangraphs), which is fine, but not THAT much lower. Damn!

By position, already inducted (I'm missing someone)

C - 23
1B - 26
2B - 24
3B - 21
SS - 23
LF - 25
CF - 26
RF - 24
P - 74

That's 27.8% P, 34.2% C/3B/SS/2B, 38.0% 1B/OF

Just using WAR is NOT sufficient to guarantee fairness across positions

There are a couple of worthy 3B, but I'm having a hard time coming up with a deserving catcher.

Schang isn't really borderline, IMHO. Other than him, who else is there?

Personally, I give Schang a bit of a mental boost for being the best catcher of his time. I don't have Player won-lost records for most of his career, but I'm guessing that especially his hitting stands out dramatically compared to his peers.

Other catchers worth considering would include Jorge Posada (who is the #10 returning candidate, so he's a required disclosure anyway), especially if you're as skeptical as I am about the pitch-framing / game-calling numbers. Thurman Munson has his supporters (I'm not really one of them). My system likes Darrell Porter, although I think he's still fairly far off-ballot for me.

Beyond that, it does kind of come down to how many catchers you think there should be in the HOM - i.e., how much are you willing to lower your normal standards to get to that next tier of catchers (Lombardi?, maybe Lance Parrish, maybe Bob Boone or Jim Sundberg if we want to reward defense and career length?). I can see both sides here - the HOM has fewer C than any other position and maybe the default expectation should be the same number at every position. But individual catchers really do accumulate less value than individual first basemen, both in-season (fewer games per season) and in their careers (shorter careers).

I can't count! The final score should be John 10, Tiant 9.

Fun stuff: Take the "winning season" all 19 times, and you get a record of 296-185 in 4372 IP - comparable, but superior, to Tom Glavine's career.

Take the "losing season" all 19 times, and you get a record of 176-146 in 2801 IP. The winning record is deceiving; it includes John's well-supported years with the Dodgers, and the ERA+ wouldn't be much better than average. Career value is in the vicinity of guys like Jim Lonborg, Burt Hooton, and Andy Benes.

"Schang isn't really borderline, IMHO. Other than him, who else is there?"

I have Munson slightly ahead of Schang. Both are borderline IMHO, but I'd be on the 'in' side of the line. That said, I doubt I'd ever be upset if they don't get elected.

As an aside, I'd be interested in hearing the argument for Dolf Luque from a Luque voter. He's on a list of players I want to take a closer look at before the 2019 ballot and you seem like a good person to hear the pro-Luque case from.

Thank you, Carl! I'll have a Luque post ready in the next week or so.

I'm not overly concerned with positional distribution unless it's egregious - I vote for the players I consider to be best qualified. If there are 23 qualified catchers and 29 qualified RF, it's not that big of an issue for me. I also think it's imporatant to look at the representation in context to see if a player actually helps/hurts this issue.

Take catchers for example. THe HoM has elected 23, and I agree that those 23 are the most qualified eligible candidates. I rank Thurman Munson 24th, and below my borderline, but he's close enough that I'll admit I may be wrong about him and he could be deserving of ballot space. But There's an enormous gap between the bottom of the HoM catchers (Bresnahan and Freehan) and the best of the rest after Munson (for me I rank Ernie Lombardi and Jim Sundberg as the next two best, but they are well below any reasonible borderline). That being said, this will change in the future, as I rank the recent set of catchers rather highly - Mauer and Posey are both already qualified, and Yadier Molina, Brian McCann, and even Russell Martin rank better than any eligible catcher aside from Munson. 2B is a similar thing - the only inducted player I disagree with is Nellie Fox, but aside from Jeff Kent, there aren't a lot of strong candidates.

Then there are situations like third base and center field. Third Base is underrepresented, yes, but that's mainly due to the slim pickings from 1920-1960. You have Eddie Mathews, Jud Wilson, and Stan Hack, as well as have of Joe Sewell for 40 years. The next best choice is Bob Elliott, but he's still not a great candidate. I think Sal Bando or Buddy Bell are better candidates (although not great ones). But third basemen from their era are quite well accounted for, so I don't feel any need to bump them up because of an unrelated 40 year positional drought. If there were a better candidate from the underrepresented era, I'd give more consideration to giving them a boost.

Similarly for CF - While there may be too many inductees from the 20s and 30s, there aren't that many pure CF from the 60s, 70s, and 80s - Willie Mays and Jimmy Wynn are there. The second half of Mickey Mantle's career, but he's really a 50s player. Reggie Smith and Andre Dawson were tweeners and Robin Yount is really a SS who spent his decline in center. A raw count gives you 6 for 30 years, but in terms of actual time spent in center and the quality of that time, it's more like 3 and a half. I don't think inducting Davis would add to any problem.

Overall, I don't think positional counts are all that useful in a vacuum. If there is something egregious.

So maybe I should give Munson a longer look.

I have Munson slightly ahead of Schang. Both are borderline IMHO, but I'd be on the 'in' side of the line. That said, I doubt I'd ever be upset if they don't get elected.

Munson's career is just so short. Unless you give him some sort of post mortem credit (which I don't see the justification for), I don't see how he has a really strong case. Too short a career with not a high enough peak to offset it, IMHO.

Schang's more interesting because of the era was rather devoid of worthy Cs. And if you adjust his WAA to account for the 154 game schedule and being a catcher, then he's not that far away from being at a level where most guys who have good cases are. So I can kind of see it if you squint real hard. As previous poster said, a "mental adjustment" given the scarcity of the position for his era may be appropriate.

I'm sympathetic to Jaack's perspective that positional balance is a not a necessary condition for a balanced ballot. I'd add that era balance needn't be either, given that guys who have been around for 100 years have been pretty well picked over (and if they haven't been included, it's because they weren't consensus picks). There can and perhaps should be a "recency" bias on forthcoming ballots.

I haven't thought all that much about it before, but Jim Sundberg is probably worth a deeper dive too. The glove is clearly there - I rank him 4th all time in catcher defense after Ivan Rodriguez, Bob Boone, and Gary Carter. Pudge and Carter could actually hit, and Boone really couldn't. But Sundberg could hit okay. Career wRC+ of 91 is certainly on the low end of things as far as the HoM goes, but it's not abjectly awful. And Sundberg was pretty durable - catching 140-150 games a year.

Probably not a HoM, but maybe someone to look a little harder at.

The Win Value stat I developed analyzes a starting pitcher's game-by-game performance and estimates how much his performance during a game improved his team's chance of winning the game (based upon runs scored and runs allowed and innings pitched) relative to a league average starting pitcher.

I am not sure if I have ever posted Luis Tiant's season-by-season Win Value figures for his career:

Year   Win Value
1964     1.81
1965    -0.12
1966     1.72
1967     1.51
1968     6.13
1969    -1.21
1970     1.01
1971    -1.16
1972     2.25
1973     3.71
1974     4.71
1975     1.85
1976     3.51
1977     0.72
1978     2.82
1979     0.71
1980    -1.00
1981    -0.75
1982    -0.07
TOTAL   28.15

Tiant's 28.15 career Win Value total is very impressive and is one of the reasons Tiant has been fairly high on my HOM ballots.

Rob, can you provide a reference point or two? Like who are some pitchers who have 26-30?

Third Base is underrepresented, yes

Has anybody done a more detailed analysis that splits players across the positions they played? One thing about third base is that it feels to me like it's a transitional position for a lot of Hall-of-Meriters who played some significant portion of their career there, but you wouldn't consider it their primary position: Cal Ripken, Edgar Martinez, Jim Thome, Pete Rose, maybe Dick Allen (or maybe he's counted as a 3B here), Joe Torre. Even Johnny Bench played a couple of years at 3B at the end of his career (although, realistically, if you just excluded those two seasons, I doubt Bench's HOM-worthiness changes at all). But maybe I'm wrong about that and/or that's true of other positions as well.

To my mind, the big under-representation in the Hall of Merit is pitchers. I think my personal HOM would be something closer to 35% pitchers. And that discrepancy is going to really show up when I share a preliminary ballot this year (which I may not do for a couple of weeks). I think my first pass at such a thing had 9 pitchers in my top 10 - partly that's because the top two newcomers (Rivera and Halladay) are pitchers; actually, top 3 newcomers in my opinion (Pettitte - although I think he falls into that "I would replace 20 outfielders and relief pitchers with 20 starting pitchers in my pHOM" grouping), but basically, because my pHOM would already have fewer outfielders than the actual HOM does, you've already elected a bunch of the guys I'd consider borderline outfielders (e.g., Vlad Guerrero).

Tiant's 28.15 career Win Value total

This is centered on average, right? So, you're saying Tiant was below average in 1969 and 1971, but not necessarily below replacement-level?

Has anybody done a more detailed analysis that splits players across the positions they played?

Yes, in one of the previous year's discussion threads. Maybe start looking for it in the 1980s threads?

slim 3B pickings from 1920-1960

One obvious guy there is Pie Traynor who has a Brooks Robinson like anecdotal reputation on defense and an RField of 0. If he really was a plus-plus glove then he's close.

The guy on my ballot is Buster Clarkson who is likely more 3B than SS and consistently hit better than OF Willard Brown in the same leagues.

Regarding the 1970s third basemen (Buddy Bell, Sal Bando, Ron Cey) the problem is there are just too many of them. If you look at BBREF WAA for Buddy Bell it's 32.4 for his career. If you look at Dan R's WAPA (positional average) Bell scores a 17.7. This says the average 3B is better than the average player and those positive RPos values for 3B in the 1970s don't make sense. We've already inducted Brett, Boggs, Schmidt, Nettles, Darrell Evans, Dick Allen, Ron Santo, Joe Torre, Paul Molitor and Harmon Killebrew who appear at 3B in the 1965-1985 time frame. Should we add Bell, Bando and Cey to the list too? Meanwhile Bert Campaneris scores 29 wins above the average SS in that era.

Here are the top-50 starting pitchers from 1941-2017 (the Retrosheet play-by-play era) according to Win Values.

Rank Pitcher          Win Values
 1   Roger Clemens      82.37
 2   Greg Maddux        65.23
 3   Tom Seaver         55.96
 4   Randy Johnson      53.96
 5   Pedro Martinez     51.60
 6   Jim Palmer         45.67
 7   Warren Spahn       44.57
 8   Bob Gibson         40.40
 9   Mike Mussina       39.00
10   Curt Schilling     38.41

11   Tom Glavine        37.95
12   Clayton Kershaw    37.90
13   Gaylord Perry      34.67
14   Bert Blyleven      34.64
15   Fergie Jenkins     34.40
16   Roy Halladay       34.18
17   Whitey Ford        34.16
18   Steve Carlton      33.57
19   Nolan Ryan         33.09
20   Kevin Brown        32.06

21   Sandy Koufax       31.01
22   John Smoltz        30.84
23   Tim Hudson         30.65
24   Robin Roberts      30.07
25   Zack Greinke       29.54
26   Don Sutton         28.72
27   Bret Saberhagen    28.28
28   Luis Tiant         28.15
29   Roy Oswalt         28.13
30   Johan Santana      28.02

31   Felix Hernandez    27.50
32   David Cone         27.22
33   C.C. Sabathia      27.12
34   Cole Hamels        27.04
35   Kevin Appier       26.82
36   Hal Newhouser      26.49
37   Andy Pettitte      25.90
38   Juan Marichal      25.64
39   Don Drysdale       25.56
40   Jimmy Key          25.31

41   Ron Guidry         24.55
42   Billy Pierce       24.54
43   Jon Lester         24.53
44   Phil Niekro        23.66
45   Tommy John         23.65
46   Justin Verlander   23.56
47   Chuck Finley       23.47
48   Max Scherzer       23.39
49   Dave Stieb         22.84
50   Rick Reuschel      22.01

Yes, these are measured relative to league average. I am fairly confident that Tiant was above replacement in both 1969 and 1971.

Has anybody done a more detailed analysis that splits players across the positions they played?

The good doc from Post 261 from the 2018 hall of merit ballot discussion thread:
http://www.baseballthinkfactory.org/hall_of_merit/discussion/2018_hall_of_merit_ballot_discussion/P200

261. Dr. Chaleeko Posted: December 03, 2017 at 01:48 PM (#5585129)
I've updated my position counts for the HOM. It now includes all appearances data from the NLDB, as well as some from select players' minor leagues time (Campy, Jackie, Irvin, Grant, Minoso, and Trouppe). I've seen a couple different ways that people prefer to assess position, so here's a few ways we could look at it. The totals below are all rounded.

PRIMARY = The position the guy played the most (i.e.: Ernie Banks is at 1B)
SUM ALL = The sum across all HOMers at every position of the percentage of appearances (i.e.: Ernie Banks counts 51% toward 1B, 45% toward SS, 3% toward 3B, and 1% toward LF)
SUM 33%: Similar to SUM ALL but only counting positions where a player made at least 33% of his appearances (i.e. Banks only counts 51% toward 1B, 45% toward SS, and everything else is ignored)
SUM 25%: Ditto but with a 25% threshold

C 1B 2B 3B SS LF CF RF DH SP RP
====================================================
PRIMARY 20 29 22 17 26 24 25 24 3 69 4
SUM ALL 18 29 22 20 23 27 26 24 5 58 12
SUM 33% 17 23 20 15 21 20 21 18 2 57 4
SUM 25% 18 24 20 16 22 22 21 18 2 57 6
----------------------------------------------------
AVERAGE 18 26 21 17 23 23 23 21 3 60 6



Based on any of these means of reckoning position, we remain well short at catcher and third base, and mildly so at second base. If we were to elect Chipper, Rolen, and Thome as three of our four, which seems pretty likely, it would have a very positive effect at 3B, of course. Here's how it would play out by each of the four methods above:
-PRIMARY: Adds 2 3Bs, total of 19 3Bs
-SUM ALL: Adds 2.02, total of 22
-SUM 33%: Adds 1.82, total of 16 (rounding error, goes from 14.6 to 16.4)
-SUM 25%: Adds 1.82, total of 18

-PRIMARY: Adds 2 3Bs, total of 19 3Bs
-SUM ALL: Adds 2.02, total of ..22

Thanks, Bleed - and Dr. C!

So, my hunch was correct: adding in partial play does boost 3B, although it remains low (although in the same ballpark at 2B and SS it looks like).

I do think that my ballot is going to reflect my dissatisfaction with this. I'm not sure how many, if any, 3B there'll be, but, as I already noted, I think it's going to be very pitcher-heavy, and it should have at least a few up-the-middle defenders - Posada, Kent, Vern Stephens, probably Wally Schang. As for 3B, I think DL makes a good point in #38 that we aren't really short on 3B since at least 1970, maybe even the 1960's (Brooks, Santo, Dick Allen, late Eddie Mathews).

I'll take DL's suggestion (in #37) and take a look at Bus Clarkson. And maybe I will take a look at Pie Traynor. My recollection is that my system actually kind of likes him, although he hasn't popped up on my ballot yet, but I'm still missing two full seasons of his career - plus a few missing games within some other seasons (although Retrosheet has pretty strong coverage of the Pirates in the 1920s). [As a side note, not to get peoples' hopes up, but Dave Smith suggested to me that he's hoping to release at least 1923, 1924, and the 1921 AL (NL is already there) this summer. So, I could have Pie Traynor's full career by the time we vote next Fall.]

Do we have an Andruw Jones thread? There has been so many good threads on the site lately about his HOF worthiness, it seems like we should be linking them. I think those threads have made good arguments for pumping the breaks on Lofton and Jones. Not that they are unqualified, just that their value is so tied to iffy defensive metrics.

"Should we add Bell, Bando and Cey to the list too? Meanwhile Bert Campaneris scores 29 wins above the average SS in that era."

I'd only argue for Bell myself. Plus Ray Dandridge. And I think Campaneris deserves a good long fresh look this year. He's on my list of returning guys to give that to.

"To my mind, the big under-representation in the Hall of Merit is pitchers. I think my personal HOM would be something closer to 35% pitchers. "
I tend to agree. I'd wager most of our PHOMs include a higher pitcher % than the overall HOM. The problem is that your analysis of pitchers is greatly influenced by the method with which you choose to evaluate as well as your stance on career vs peak. Some solid options left from the Negro Leagues also complicate matters. Everybody seems to have their own 'pet' pitcher or couple of pitchers. I myself like a lot of the pitcher backlog and feel several belong in the HOM, but I don't feel super strongly about any single one to be a hard core advocate for him. I guarantee we have a lot more consensus on position players which makes them easier to actually elect.

Rob,
I'd generally like to understand more about what that Win Value number means. In a given game, is Tiant's win value computed on a WPA type basis with each batter faced? Or is his total game taken after the fact and compared to what an average pitcher would theoretically have done? I'm definitely interested in learning more as this affects not only Tiant, but also Oswalt.

http://www.baseballthinkfactory.org/hall_of_merit/discussion/andruw_jones

Win Values are calculated at the end of each game. The core data underlying the estimation of how much a starting pitcher contributed (positive or negative) to his team winning the game are (i) the last inning pitched and (ii) the score of the game at the end of the last inning pitched by the starting pitcher. Win Values do not use play-by-play data and they do not take into account the quality of the team defense behind the starting pitcher.

Here is a link to a Win Values write-up the good folks over at Retrosheet were kind enough to post.

http://retrosheet.org/Research/WoodR/WinValues.pdf

Interesting. Thanks for the post.

Thanks Rob. So looks like more or less the same scale as WAA (which makes sense, since they're both relative to league average).

Looking forward to reading through that PDF when I get a chance.

In terms of the positional balance, I just don't think it's as big an issue as some others. I'm even less concerned about positional balance within an era (in part because how an era is defined is rather arbitrary). Treating all eras equally (i.e., the HOM constitutional provision) doesn't require some sort of affirmative action on the part of positional identity.*

*-I'm saying positional identity because a lot of C/3B spent a portion of their careers at other positions, not just to be glib.

There were many more great 3B relatively speaking in 1960-present and not a lot of them prior to 1960. I'm not sure why Buddy Bell (32.4 WAA) should be less deserving than Pie Traynor (10.2 WAA) just because there happened to not be a lot of great 3B in the 1920s and 1930s.

The opportunity to balance positions within the past era was when these guys were the only ones on the ballot. Close to 100 years have come and gone and Traynor's not been elected, which to me indicates that he's not deserving. Now it's the turn for the more deserving, more recent players. And to the extent that we're concerned about positional balance, then future inductees should represent the best players at that position. Bell and Bando are the best qualified 3B who aren't already enshrined, IMHO. Now they're borderline, but in a weak year they might slip through (like Guerrero just did).

There were many more great 3B relatively speaking in 1960-present and not a lot of them prior to 1960. I'm not sure why Buddy Bell (32.4 WAA) should be less deserving than Pie Traynor (10.2 WAA) just because there happened to not be a lot of great 3B in the 1920s and 1930s.

Here's where it gets interesting though. Buddy Bell gets a 32.4 WAA from WAR and a 17.7 Wins Above Positional Average from Dan R. Traynor gets a 10.2 WAA but a 29.9 WAPA from Dan R. This says the "average" 3B was one helluva lot better when Buddy Bell played than the average 3B when Pie Traynor played. If you look at Baseball Reference RPos (runs from positional scarcity) you see 1-5 RPos for Traynor and 1-4 for Buddy Bell. Dan R clearly thinks there was a 3B drought during Traynor's era and a 3B glut during Bell's but bWAR somehow thinks every 3B throughout history has the same positional scarcity.

Grading defense, Bell gets a 174 RField from bWAR and 18.2 FWAA from Dan R; agreement. Traynor gets a -32 RField but a 9 FWAA from Dan R; a difference of 12 WAR! Traynor was considered by many of his peers to be the "greatest fielding 3B of the 1920s", which due to the Bill Doak glove would likely make him the best fielding 3B of any era before 1930.

I have no doubt Traynor was able to make plays that players just 10 years earlier would not have been able to make simply due to the advancement of fielding gloves. Most of his peers thought he was best but the bWAR and fWAR calculators think he was barely adequate. Joe Cronin agrees with Fangraphs that Ossie Bluege was the best glove 3B of that era. Chief Bender thinks it was Willie Kamm and Fangraphs has him in the top 5. What was probably happening is every team was seeing the "best 3B in the history of the franchise" simply because the glove technology took a leap forward. Still, I don't know how the anecdotal evidence could be THAT wrong. It's like Eric Hosmer times 10.

This is pretty quick-and-dirty, but it's the level of data I can work with. I have a spreadsheet with all of the HoMers listed by season and position. I just use the most-played position for each year. This does include some garbage (i.e., Larry Jones' 3 games in 1993 count as 1 SS season), but I don't think it makes a major impact on what I did.

I figured out the percentages of total seasons for all of the positions. So, 7.25% were C, 10.45% were 1B, etc. (I did adjust for DH and non-DH seasons.) Then I figured out for each year, given the number of HoMers playing that year, how many should be at each position, by the historical average, and compared it to the actual number. So there were 32 HoMers in 1960, which should translate to 2.36 C, 3.41 1B, and so on. Since there was 1 catcher that year (Yogi), I get a -1.36 for that year. Then I put together the 10-year averages for each position to look for trends.

Obviously, since this is based on what we've already done, it doesn't address any overall trends about not having enough third basemen or pitchers, or whatever. But I thought it was interesting, so here are the top and bottom of the lists. (I also did a breakdown by Pitchers/Hitters/Fielders, which I'll put in the next post.)

Most under-represented position-decades
SS 1960s -3.188
1B 1910s -2.472
P 1930s -2.469
P 1870s -2.365
CF 1970s -2.334
P 1890s -2.333
SS 1970s -2.247
P 1880s -2.236
CF 1980s -2.215
1B 1950s -2.179

Good points for Bert Campaneris & Ben Taylor. I suspect the 1930s pitchers are more about too many hitters from that era, and the 19th Century pitchers shouldn't be compared 1-for-1. 1970s CF, I guess the best candidate is Cesar Cedeno.

Most over-represented position-decades
CF 1920s 3.549
SS 1900s 2.543
P 1980s 2.462
CF 1910s 2.461
P 1970s 2.407
P 1960s 2.304
1B 1930s 2.229
CF 1950s 2.068
C 1930s 2.029
3B 1970s 1.976

I suspect the 1920s CF are in part because Cobb & Speaker weren't moved off the position when they normally would've been. Then again sometimes you just have clumps. This is a point against any more 1970s 3Bmen. But also against Tommy Leach as a 1910s CF, who's on the bottom of my ballot.

The opportunity to balance positions within the past era was when these guys were the only ones on the ballot.

It is fairly well balanced, all things considered. Way better than the actual HoF. We need to be diligent going forward to make sure things don't get imbalanced. Look over your top 50 players. If you don't have any catchers or left side infielders in your top 50 you need to tweak things. Your distribution should toss a catcher about every 10-15 players. If not, you're skewing to favor one position over another.

I also very much disagree that we should stop inducting players simply because they weren't inducted yet. The borderline "backlog" players should be pretty much randomly distributed throughout baseball history if we balance eras. For older players we are also relying on much more imperfect data than with current players. However, the play-by-play and Negro League data is improving all the time. We need to make sure we continue to consider players from all eras.

But also against Tommy Leach as a 1910s CF, who's on the bottom of my ballot.

Except he's also a 1910s 3B.

And here's the Pitchers/Hitters/Fielders breakdown. Hitters = 1B, LF, CF, RF & DH, Fielders = C, 2B, SS, 3B.

Most under-represented
Hitters 1980s -4.494
Fielders 1960s -3.171
Hitters 1970s -2.830
Fielders 2000s -2.525
Pitchers 1930s -2.469
Pitchers 1870s -2.365
Pitchers 1890s -2.334
Pitchers 1880s -2.237
Fielders 1950s -2.057
Pitchers 1940s -2.002

I will say one part of why the 1980s hitters are so low is they get a lot from the DH (-1.760). The 1970s had a big clump of guys ending their careers (Aaron, Killebrew, Frank Robinson, Billy Williams and more), and in the 1990s you get full-time DHs in addition to the end-of-career guys. The 80s had a few years each for Yaz & Reggie, and that was it. I'd tend to treat the 70s and 80s hitters as relatively equally overlooked.

Most over-represented
Pitchers 1980s 2.462
Pitchers 1970s 2.406
Pitchers 1960s 2.304
Fielders 1940s 2.153
Fielders 1980s 2.032
Fielders 1870s 1.976
Pitchers 2000s 1.708
Pitchers 1990s 1.670
Hitters 1880s 1.653
Hitters 1890s 1.431

1990s & 2000s numbers aren't really complete yet. This would argue against Tiant, Tommy John and Rizzuto (and I guess Bus Clarkson as well). I'm honestly not sure why 1940s fielders score so high.

The discussion of positional balance is a fascinating and important conversation, but I want to clarify a rule. This mostly relates to Carl's suggestion in #43 that most of us probably have more pitchers in our personal HOM's, but we disagree on who the "missing" pitchers are.

If I were to construct a personal Hall of Merit, I am 99% sure that it would include both Tommy John and Luis Tiant. I very strongly believe that Tommy John is the better candidate of the two - his additional career bulk more than makes up for the fact that Tiant had a handful of better individual seasons. I believe that this is a minority position among HOM voters.

I just want to clarify: if I really truly believe that Tommy John is a better HOM candidate than Luis Tiant, it would be a violation of the rules for me to nevertheless place Tiant higher on my ballot in an effort to get him elected. Correct?

Also, I want to say that I very much agree with DL's point in #49. It seems to me that BB-Ref has set its positional averages such that the average third baseman in the 1970s was above average and the average third baseman in the 1920s was below average. Which I believe is nonsensical. Yes, there will be star clusters at various positions at various times. But ultimately a player's value has to be measured against his contemporaries first and foremost and if everybody has a good third baseman, then there's simply not as much value in having a good third baseman.

The 1940s numbers: C=1.41, 2B=1.10, SS=0.94, 3B=-1.29.

Catchers get a boost from some generational crossover early in the decade, plus losing less players to WWII - Gibson, Trouppe and Campanella all stuck around. And Campy gets more big league seasons than he would've later on.

Second Base just has a lot of guys, with Gordon, Doerr, Herman plus some of Gehringer, Jud Wilson, Robinson and Fox. The '30s had a little bit less, the '50s had a lot less.

SS and 3B pretty much stay at the levels they had in the 30s.

if I really truly believe that Tommy John is a better HOM candidate than Luis Tiant, it would be a violation of the rules for me to nevertheless place Tiant higher on my ballot in an effort to get him elected. Correct?

Here's the rule:

Voters agree to take the voting seriously and to put in sufficient time in researching the merits of the players and in filling out their ballots. In addition, voters pledge to refrain from “strategic” voting; that is, manipulating one’s ballot (i.e., so it does not reflect one’s own beliefs regarding the relative merits of the players) in an attempt to achieve a more desirable group ranking. Voters should simply vote for the 15 best eligible players, ranking them from 1 to 15. Even if it appears a player won’t be elected, you should still vote for him if you feel he is worthy.

http://www.baseballthinkfactory.org/hall_of_merit/discussion/our_constitution

Thanks, DL!

Thanks for discussion on positional balance--I need to digest it some before commenting further.

But with respect to era balance...

It is fairly well balanced, all things considered. Way better than the actual HoF.

Absolutely agree.

We need to be diligent going forward to make sure things don't get imbalanced. Look over your top 50 players. If you don't have any catchers or left side infielders in your top 50 you need to tweak things. Your distribution should toss a catcher about every 10-15 players. If not, you're skewing to favor one position over another.

This strikes me as something akin to positional affirmative action. Should I really give <insert whoever is the best catcher> a boast just because my ballot happens to not include a catcher? For the record, my 2018 ballot included:
- (4) 3B (Jones, Rolen, Bando, Bell)
- (4) SP (Tiant, Santana, Appier, Shocker)
- (4) OF (Lofton, Bo. Bonds, Guerrero)
- (2) 1B (Thome, Chance)
- (1) 2B (Tinker)

I promise that I'll reconsider Bando and Bell for the next ballot. But on its face, I don't see anything wrong with the positional composition. 4/15 were pitchers (1 short of 33%). 4/15 were outfielders (about the right percentage given 3/8 positions are outfielders and 66% target for position players). Two 1B and 1 2B seem reasonable and I think it's defensible that it lacked both a SS and C given the field.

In the Top 25, there was 1 SS (Rizzuto) and 3 C (Posada, Schang, Munson). I considered all of those players, but decided against including them. Maybe I'm wrong, but in my mind:

1) Rizzuto requires a lot of WWII credit to merit inclusion, to the point where I think it requires speculation beyond what the data support. If you take the average of his 1942 and 1946 seasons, then you get 2.1 WAA. Multiply that by 3 for 6.3 additional WAA and add that to his total of 21. 27.3, even with adjustment for 154 game schedule, still has him on the outside looking in. Borderline, but I think that it's justifiable to not include him on the ballot (only 6 of 30 included him).

2) Posada's defense (-60 RField) detracts from his overall value to the point where he only has 17.3 WAA. Even if you adjust the standard to account for catchers having decreased playing time and award him a bonus for postseason play (solid, but unspectacular), I think that it's defensible to leave him off of one's ballot (only 7 of 30 included him).

3) Schang, I'll take another look at in light of the discussion about position relative to standard for the era. Maybe if you award him bonus points for playing a weak position for the era and adjust for reduced playing time (154 game and position), then maybe he gets there (only 8 of 30 included him).

4) Munson just had too short a career with not a high enough peak. I don't think he should get a post mortem bonus or whatever. Had he played another few above-average seasons, he might have a case. But he didn't (only 6 of 30 included him).

So I'll take another look at Schang and reconsider in light of the discussion on positional averages across eras. Maybe he's more deserving of a spot at the back of my ballot than Frank Chance (#15 on last year's ballot).

For next year, I'll have at least two more pitchers (Rivera and Halladay, possibly Petitte) and I'm going to read Kiko's book to see if he can convince me about Tommy John (only 6 of 30 ballots).

I also very much disagree that we should stop inducting players simply because they weren't inducted yet. The borderline "backlog" players should be pretty much randomly distributed throughout baseball history if we balance eras. For older players we are also relying on much more imperfect data than with current players. However, the play-by-play and Negro League data is improving all the time. We need to make sure we continue to consider players from all eras.

Well I respectfully disagree with the claim that the distribution of backlog players will be random across baseball history at this point. Imagine that there's three private ponds of stocked fish. One you go to every day starting in April. The other one you start going to at the beginning of May. Now it's the start of June. If you want to bring home a fish, aren't you most likely to find some in the pond you've never fished in and more likely to find some in the second pond than the pond you've been visiting since April?

It's a dopey analogy, but really there just aren't very many 19th century players left or Negro Leaguers, etc. worth considering. To return to my analogy, those ponds have visited a lot over the years to the point where those who are there really aren't the cream of the crop for their respective eras (or they would have been elected). Going forward, the vast majority of players on future ballots will be more recent players. The HOM as a whole should be well represented across eras (by design); however, individual ballots in the 21st century need not be to achieve that objective given that existing HOM inductees are already biased toward the earlier players given that they've been subject to more elections than more recent players.

I agree that the top of the ballot should almost always be new players. The players left are borderline by definition. However, once you get past the top 20-25 the eras should be fairly well distributed.

It's a dopey analogy, but really there just aren't very many 19th century players left or Negro Leaguers, etc. worth considering.

I tend to agree regarding the 19th century. But with Negro Leaguers, I think we need to be aware that as more and better data become available, worthy players may well emerge that the HOM missed on due to a lack of data more than said players having been judged unworthy. I think that's what we're seeing, for example, with Ben Taylor.

The players left are borderline by definition. However, once you get past the top 20-25 the eras should be fairly well distributed.

I started voting here maybe three years ago and my expectation was along the lines of 6-4-3. Basically, as you all know, my data doesn't go back to the start of MLB. But, I figured the missing time period had already been well picked over, so it was okay; very few players from that time period were worth considering anyway.

But the reality is more as DL suggests. I think once you get past the obvious guys - Chipper Jones, Jim Thome, Roy Halladay, as recent examples - the next tier of your ballot is going to be guys who look better in your system than in other peoples' systems - your "pet" players, so to speak. And there, I do think it makes sense to expect your personal "pets" to be spread through history - for me this includes Tommy John (1960's - 80's), Vern Stephens (1940's), Orel Hershiser (1980's). Down-ballot or just off-ballot, I get guys like Gil Hodges (1950's), Dolph Camilli (1940's), Toby Harrah (1970's - 80's). And I find that, because of this, I do need to go back and try to identify the players that I think my system would have liked even in the early years before I have full data - e.g., Wally Schang (1910's - 20's), Kiki Cuyler (1920's), et al.

Although this is probably due in part to the fact that my system is pretty idiosyncratic relative to everybody else - although I'm hoping this will change over time as some of you start incorporating my data into your ballots. But if you jump in and start using BB-Ref's WAA/WAR, then, yeah, you're probably going to bump against a consensus that might make your backlog skew more modern if more of your old-timey borderliners line up with the old-timey borderliners who were inducted back in the old timey.

Does Johnny Evers have a discussion thread? I wanted to talk about him some but didn't necessarily want to clog up this thread - which I'll do if he doesn't have a thread (that can be easily found - I tracked down a list of discussion threads for 20th-c candidates and he wasn't there).

But the reality is more as DL suggests. I think once you get past the obvious guys - Chipper Jones, Jim Thome, Roy Halladay, as recent examples - the next tier of your ballot is going to be guys who look better in your system than in other peoples' systems - your "pet" players, so to speak. And there, I do think it makes sense to expect your personal "pets" to be spread through history - for me this includes Tommy John (1960's - 80's), Vern Stephens (1940's), Orel Hershiser (1980's). Down-ballot or just off-ballot, I get guys like Gil Hodges (1950's), Dolph Camilli (1940's), Toby Harrah (1970's - 80's). And I find that, because of this, I do need to go back and try to identify the players that I think my system would have liked even in the early years before I have full data - e.g., Wally Schang (1910's - 20's), Kiki Cuyler (1920's), et al.

Okay, following up on this paragraph of mine in comment #60, I realize that I haven't necessarily done as thorough an evaluation of the full time period preceding my data as I might. In particular, my "pet candidates" go no earlier than Wally Schang, who debuted in 1913. In effect, I don't really have any players who fully predate my system.

So, living on the north side of Chicago, a team that has always fascinated me is the 1906-10 Chicago Cubs. By record, this was the greatest team in (white) major-league history. One can certainly timeline that away, but the spirit of the Hall of Merit is to evaluate the 1906-10 Cubs within their time. And for all of their greatness, somebody correct me if I'm wrong, but that team has only two players in the Hall of Merit (3-Finger Brown and Jimmy Sheckard). One reason for this is that these teams won a lot more games than they should have based on their raw statistics. BB-Ref's WAR says the 1906 Cubs should have won 101 games (56.0 bWAR plus an expected 45 "replacement" wins - replacement level is .294, which translates to a 45-109 record in a 154-game schedule); they actually won 116. The 1907 Cubs beat their BB-Ref WAR (92) by 15 games as well. In 1908, they only beat bWAR by 12 (99 actual wins).

So, in thinking about - who might my system pick up in the years before I have data - the obvious answer is that, the key to my system is that I tie my pWins to team wins. So, the players on a team who regularly win more games than WAR says they should have won would probably show up more favorably in my system.

Having set it up, I'll continue this in a separate post.

The specific guy on those Cubs teams who has always fascinated me the most was Johnny Evers. Because Evers did his thing in Chicago, but then moved to Boston, and, BAM!, the Braves go out and win the pennant and Evers wins himself an MVP award. And if you look at the 1914 Braves, you see the same thing as what you saw with the 1906-08 Cubs - bWAR says the Braves should have only won 81 games. They actually won 94!

In fact, here's how every team for which Johnny Evers played a full season (and more than one game) did vs. bWAR (and, in parentheses, vs. their Pythag record):

1902 CHN -12 (-5) (Evers played only 26 games)
1903 CHN +3 (+4)
1904 CHN +13 (+6)
1905 CHN +1 (-12) (Evers only played 99 games)
1906 CHN +15 (+1)
1907 CHN +15 (+5)
1908 CHN +12 (+1)
1909 CHN +10 (-5)
1910 CHN +14 (+3)
1911 CHN +8 (0) (Evers only played 46 games)
1912 CHN +12 (+8)
1913 CHN +12 (+2) (Evers was the manager - and played fulltime)
1914 BSN +13 (+5)
1915 BSN +4 (+2) (Evers only played 83 games)
1916 BSN +10 (+1) (Evers only played 71 games)

That seems fairly remarkable. And it's not a Pythag-v-actual difference. It's WAR-vs-actual.

Two more posts with some sensitivity analysis of these results.

My first sensitivity analysis. Maybe the issue is that bWAR breaks down for extremely good teams. Although, the 1914 Braves (and 1915-16 Braves and 1904-05 Cubs) weren't "extremely good" just (mostly) pretty good or good. Anyway, I did a similar analysis for several great teams through history. I picked four teams that set their league victory record, the teams with the top regular-season record the last two years, and the 1969 Baltimore Orioles, who I believe had the most wins of any team that under-performed its Pythag record. Same numbers: Wins minus WAR (Wins minus Pythag)

1927 Yankees -2 (+1)
1954 Indians +12 (+7)
1969 Orioles -1 (-1)
1998 Yankees +3 (+6)
2001 Mariners +1 (+7)
2016 Cubs -2 (-4)
2017 Dodgers +8 (+2)

A bit of a positive bias, but that's basically being driven by 2 of the 7 teams (Indians, Dodgers). Certainly nothing that could explain the numbers for Johnny Evers's career.

The other sensitivity test, then, was Evers's teammates on the Cubs.

Start with the other Hall-of-Famers.

Joe Tinker became a regular one year before Evers - in 1902 - which, as I showed in #63, was unremarkable. The big over-performance of the Cubs started in 1904 - Evers's second full season and Tinker's third. Tinker left the Cubs are 1912. But, of course, the 1913 Cubs matched the 1912 Cubs at +12. Tinker went to the Reds (and managed them). The Reds won one fewer game than would have been expected by either bWAR or Pythag. Tinker played one more season, with the 1914 Chicago Whales. They did win 11 games more than you'd expect from bWAR.

But overall, when Evers and Tinker were separated, the over-performance seemed to follow Evers more than Tinker.

What about "The Peerless Leader" - Frank Chance. Frank Chance only played 100 or more games six times in his career, from 1903 - 1908. In 1909, he was down to 93 games played; in 1910, he was down to 88 games played, and by 1911, he was down to 31 games played. But the Cubs out-performed their bWAR by double digits in all of those seasons except for the one where Evers missed significant time (1911). Again, the over-performance seems to track better with Evers than with Chance.

So, who else could it have been?

Catcher Johnny Kling became a regular the same year as Tinker and left a year earlier. The team started winning fewer games after Kling left, but they continued to out-perform bWAR by similar margins as before.

The only Hall-of-Merit position player on these teams was Jimmy Sheckard, who joined the team in 1906. And his arrival certainly helped the 1906 Cubs to a much better record, but, relative to bWAR, the 1906 Cubs were comparable to the 1904 Cubs. Before joining the Cubs, Sheckard was in Brooklyn, where he played for pretty bad teams the two previous years. The 1905 Superbas only won 48 games (they weren't very superb(a)); their bWAR would translate to 50 wins; the 1904 Superbas won 10 fewer games than their bWAR translated into.

The 1906 Cubs also added third baseman Harry Steinfeldt, who had a career year in 1906 and, as such, was a huge reason why the Cubs won 116 games. But bWAR recognizes Steinfeldt's huge season (7.0 bWAR). The two years before that, Steinfeldt's Reds teams beat their bWAR by 6 and 4 wins, respectively. Nice numbers, but nothing like what we see in #63.

Final conclusion: Johnny Evers appears to have had a knack for helping his team win games in a way that is missed by bWAR. As such, I suspect that when I finally get play-by-play data for Evers's career that he is a player who will do extremely well, especially in my pWin measures. How precisely to translate that into a specific ballot slot here, however, is something that I haven't quite worked out just yet.

Any thoughts, comments, critiques of what I've done here?

yes, I have posted before about the "Evers Effect" as well.

I think some of it has to do with the relevance of 2B defense in the deadball era.

was Evers a "pouncer" like 1980s Keith Hernandez from 1B?

I think that has been researched.

I don't know how else to explain the 1914 Braves - or Post 63

I think some of it has to do with the relevance of 2B defense in the deadball era.

was Evers a "pouncer" like 1980s Keith Hernandez from 1B?

My initial assumption was that this was defense. The Cubs had great defense, especially great infield defense, for their time. But in theory, that's something that should be captured within BB-Ref's WAR. BB-Ref keys its pitcher WAR off of actual runs allowed. They adjust pitcher WAR based on the team's fielding WAR. But even if they're mis-valuing the fielding WAR for Evers (and Tinker and Chance), they'd just be introducing an offsetting error in pitcher WAR for 3-Finger and Orval Overall and the other pitchers. They do some weird stuff, so it's not a perfect one-for-one tradeoff of WAR between fielders and pitchers. So that could be some of it, but I would still think that by starting with actual runs allowed in their calculation of pitcher WAR, there's a limit as to how much effect any error on the defensive side could show up at the team level.

So, I think if BB-Ref WAR is missing something, it would have to be either something on the offensive side - the Cubs were, perhaps, scoring more runs than you'd expect from their raw numbers? - or in the translation from runs to wins - which would also show up in Pythag. And while Evers's teams tended to over-perform their Pythag by a bit, they over-performed bWAR by a lot more.

One possibility is that Evers was an outstanding baserunner - going 1st-to-3rd, scoring on shallow fly balls, the sorts of things that you can only measure with play-by-play data. Although I have a lot of trouble seeing how Johnny Evers's baserunning could possibly be worth 10+ wins per year.

That said, the game being played in the early 20th century was sufficiently different from the game played even 20 years later, that I'm not sure how comfortable I am with extrapolating what I/we know based on more modern data to that era. Between the spread in player talent and the importance of infield defense and baserunning, maybe it really is possible that Evers was contributing that much in those aspects?

A few hypotheses on Evers

The obvious one is defense. The two main defensive metrics that go back that far have both Tinker and Evers as exceptional defenders, and Evers shor stop partner with the 1914 Braves was Rabbit Maranville, also an excellent defender. My thinking is that dWAR might not capture the value of having two excellent defenders up the middle in a heavily ground ball oriented era like deadball.

This explains why Tinker alone wasn't worth so much - in 1913 his primary 2B was Heinie Groh, who wasn't really a second baseman. In 1902 it was Bobby Lowe, who I'm not super familiar with, but he seems to have been a fine 2B, but certainly on the downside of his career.

That being said, if this is true, I'm not sure who to give the credit to. It's hard to say either one of them deserve extra credit defensively because their partner was good, but at the same time, that credit should go somewhere if it is real value. Has anyone done research on defensive positioning in this era? That could be something.
---

The second possibility is WAR underrates Evers offensive contributions.
Evers may have been caught stealing much lower than the average rate. We only have his caught stealing from 1913 and 1915, and the rate is not good, but it's from the end of his career. If Evers was rarely caught stealing in his prime, that would explain a lot of the missing value. This sounds plausible but the limited evidence we have definitely does not point in that direction.

It also could just be better context hitting. But there's not really any evidence of that.

Well now I look dumb for not refreshing the page....

I think I may have an unexciting explination for the bWAR0wins descrepancy.

Actual Wins vs bWAR Wins for the Pirates

1900 PIT +3
1901 PIT +6
1902 PIT +8
1903 PIT +12
1904 PIT +10
1905 PIT +14
1906 PIT +4
1907 PIT +8
1908 PIT +13
1909 PIT +16

I thought I was on to something the the two good infielders thing - The Pirates had Honus Wagnaer and Claude Ritchey from 1900-1906. Neither was as good as either of Tinker or Evers, but still a good pair. The pattern seemed to hold... but after Ritchey left and Wagner was left with pretty average gloves next to him, the Pirates still outplayed their WAR.

Next lets look at the Giants from the same period, generally

1903 NYG +5
1904 NYG +9
1905 NYG +6
1906 NYG +5
1907 NYG +1
1908 NYG +2
1909 NYG +6
1910 NYG +2
1911 NYG +2
1912 NYG +11
1913 NYG +11

Now the Giants were not notably good up the middle - Their shortstops were regularly good (Bill Dahlen, Art Bridwell, Art Fletcher) but they were mostly starting Larry Doyle at second, who was not a good defensive player. But they still overwhelmingly beat out their bWAR - not quite as compellingly as the Cubs or Pirates, but they still did it every year that they were good ( I excluded the pre-1903 Giants for being just awful).

If you look at the stadning for this period, the Cubs, Giants, and Pirates basically dominated the NL. From 1903-1913, the to two in the NL were always two of these teams, and only twice did a different team sneak into 3rd. On the other hand, Brooklyn, St. Louis, and the pre-Evers Braves were consistently terrible in this period.

While the differences between bWAR-wins and actual wins isn't quite as great, I found that the Cardinals and the Superbas (Brooklyn) did regularly underplay their bWAR - the only years that they didn't in my sample were St. Louis in 1910 and Brooklyn in 1906 and 1911, which were all better than usual teams (65 win teams as opposed to 55 win teams). I didn't check Boston, but I'd assume it's likely the same story.

While there may be something about Evers in particular, I think the most likely explanation is that the replacement level is likely off. The good teams are consistently better than they 'should' be and the bad teams are consistently worse.

I'm not sure that explains the 1914 Braves - they might just be a fluke, a good team that had a lot of luck. Or maybe there's still a replacement level issue. Either way, withut more data I think the issue with replacement level is more plausible than Johnny Evers being a lot better than he's been given credit for.

Hi all, room for a new voter ? I am working on a system for ranking players for HOF. It's designed to give bumps for peak, and also situational performance, (WPA) for both regular season and post season. One of the things I don't like about the JAWS system is how it double counts peak seasons. So I am separating the seasons , i.e., top 5, next best 5, balance of career, etc, and weighting differently. Still playing with the those weights. Guys with high peaks and big post season moments will get a bump. Long career compilers without much peak get knocked down a few notches.
What I'm trying to get to is the "feels like a Hall of Famer", but with a numerical system instead of pulling it out of my arse.

As I understand from reading HOM threads in the past, all I need to do is present my work and thoughts here in this thread, and then when the actual voting opens up in separate thread, cast my votes there. Is that still the correct process ?

Also, other than the top 10 returning vote getters and the newly eligible, is there any other list I should be referencing ?

Thanks in advance for response.

Jaack, thanks for running those numbers in #70.

Either way, without more data I think the issue with replacement level is more plausible than Johnny Evers being a lot better than he's been given credit for.

I think it's probably a bit of both. But certainly the numbers that I show in #63 are too big to be entirely due to Johnny Evers. They imply that bWAR is underrating Evers to the tune of 10 wins per year, which would push him past Honus Wagner on the career leaderboard - unless you also gave Wagner another 6-8 wins per year based on the numbers in #70 - which, in turn, would make Honus Wagner the greatest player in (white) major-league history (without timelining).

shoe (#71), welcome!

I am working on a system for ranking players for HOF. It's designed to give bumps for peak, and also situational performance, (WPA) for both regular season and post season. One of the things I don't like about the JAWS system is how it double counts peak seasons. So I am separating the seasons , i.e., top 5, next best 5, balance of career, etc, and weighting differently. Still playing with the those weights. Guys with high peaks and big post season moments will get a bump. Long career compilers without much peak get knocked down a few notches.
What I'm trying to get to is the "feels like a Hall of Famer", but with a numerical system instead of pulling it out of my arse.

I really like the concept here. It sounds like a lot of work, but that's up to you (says the guy who built his own statistic from scratch - but, having done so, I create my rankings by basically running a single query - albeit a fairly complicated one).

Also, other than the top 10 returning vote getters and the newly eligible, is there any other list I should be referencing?

When I first started doing this, my starting point for putting together a list of players that I needed to look into was the final ballot results from the previous year. How deep down that list you want to go is up to you. There are pretty recent Negro League MLE's floating around, so you may want to go fairly deep down the ballot there, because as I think I mentioned in an earlier comment, I think there may be some players there who haven't been "rejected" by the electorate so much as they were overlooked due to a lack of data back in the day.

Good luck! I look forward to seeing your preliminary ballot.

Thanks for the info and advice Kiko, especially re Negro League MLE's.

When is this thread close and when does voting thread open ?

The thread will remain open indefinitely and the voting begins sometime in November, I believe.

Welcome shoewizard!

BWAR also has issues with the runs-to-wins formulas in extreme low and high scoring run environment. Not sure if that is happening here.

I just want to say that the recent Evers comment series here is BBTF at its best. a legit topic is revisited, many thoughts are expressed, more input is welcomed, minds are open, the journey continues.......

Hi all, room for a new voter ?

Sure. I would recommend looking at the top 50 in the previous election as well as the newcomers. Voting will happen in December but discussion is ongoing until then.

http://www.baseballthinkfactory.org/hall_of_merit/discussion/second_basemen

There is some info on Evers there. I think his link is dead. Joe D mentioned that as we add new threads now the old threads get pushed off. If someone wants to help archive threads it would be appreciated

for showewizard:

I am not endorsing any of these guys per se, and our numbers experts haven't yet fixed my (presumably very minor) errors, and of course the older you debuted the more chances.... but maybe glance also at guys who in some cases got close to being elected before being drownded by the tide

Most Votes Points, but not chosen (points in 2018)
DUFFY...... 28053.5 (54)
VAN HALTREN 26923.5 (x)

MWELCH..... 18921 (38)
REDDING.... 18841 (86)
TLEACH...... 16142 (48)

CRAVATH.....12752 (57)
JRYAN....... 12678.5 (x)
WALTERS.....12194 (54)

BJOHNSON....10007 (89)
GRIMES.......9876 (x)
LDOYLE....... 9816 (x)

BMONROE......7947 (x)
SCHANG.......7169 (86)
VWILLIS.......7125 (102)
NWILLIAMSON...7035 (x)

Thanks all for the welcome, and also tips and direction. Got about month of work to do. Will be back before then with more questions though.

Interesting stuff in the WAA adjustments from Binkley. The average player is actually 0.15 WAA throughout history. Some positions (LF/RF/1B) average out to zero but have wide swings throughout history.

DH has always been penalized where the average DH is -1 WAA from 1970-1990 and -0.5 WAA from there forward.
C WAA was well below average from 1883 to the 1960s (between 0.5 to 1 win) and then again 1990-2005 (-0.5) but just a little below average in between.
1B was well above average from 1880 to 1895, 1930-1945, and basically average the rest of the time with a dip below average recently.
2B has been below average forever with the worst stretch 1955-1981 but is near average now
SS took a dive in 1920 below -0.5 WAA and made it back near average 1940-1960 but then dove again down to -1 until the 1990s.
3B has been mostly near average except below average 1920-1935, above average 1970-1985 and 2013-2018 above average
RF has been the closest to average throughout history with a blip above during Babe Ruth's career.
LF has tanked lately but otherwise has been basically average throughout history except for 1890-1905 well above average
CF has been above average up to 0.5 win from 1910 until the present with blips close to average during the 1930s and 1950s. Centerfielders are overrated by WAA except 1925-1945.

Just one quick point of clarification: The WAA adjustments are based upon positional medians, not averages. For example, 1b in the 1880s and 1930s were above average not because the ABC boys or Gehrig/Foxx/Greenberg skew the data, as they would with positional average, but just because the rest of the 1b pool was pretty strong, too.

Tinker played one more season, with the 1914 Chicago Whales. They did win 11 games more than you'd expect from bWAR.

B-R adjusts for league quality in WAR calculations; I suspect that this might be responsible for at least some of this discrepancy, since the Whales were of course a Federal League team. (Actually, it would be worth looking at whether that adjustment is responsible for some of the measured effect overall.)

Just a quick check - the teams in the 1914 NL totaled 227.8 WAR in 615 games (1230 team games); the 1914 FL totaled 110.3 WAR (1218 team games).

Checking a little more:

1914 NL: 227.8 WAR (615 games)
1914 AL: 288.4 WAR (613 games)
1914 FL: 110.3 WAR (609 games)

So WAR thinks the AL was a vastly stronger league in 1914, to the tune of about 8 wins per team season.

Quick check for overall timelining...

1954 NL: 272.5 WAR (616 games)
1954 AL: 233.6 WAR (616 games)

That's 506 WAR between the NL and AL in 1954, and 516 in 1914; no obvious timeline adjustment between those two years at least. (I assume there were more ties in the 1914 leagues, and statistics from ties count; I figured total games by just adding up team wins.)

Let's do another Evers year without the Federal League mucking things up:

1908 NL: 237.7 WAR (616 games)
1908 AL: 268 WAR (612 games)

Obviously this isn't conclusive, but it looks like B-R thinks the AL was the stronger league during that stretch. (Just basing it on pure WAR totals among position players, it looks like the pass came in 1903; the NL had more total WAR in 1902. But then, the AL had more total WAR in 1901. Which seems... odd.)

Anyway, since the total number of wins available in each league was roughly the same year-to-year (both went to 154-game schedules in 1904, so no schedule-length issues), the league that earns fewer total WAR will necessarily outperform its WAR in terms of total wins. The Cubs may outstrip the difference in the Evers years, but it should at least be accounted for.

Eric, thank you very much for #84-#86! So, what I need to do is go year-by-year and calculate the implicit replacement level by league by season and then repeat my analysis. Hopefully, I'll have some time to do that at some point this year.

#86:

(Just basing it on pure WAR totals among position players, it looks like the pass came in 1903; the NL had more total WAR in 1902. But then, the AL had more total WAR in 1901. Which seems... odd.)

The National League played with the newly-created foul strike rule in 1901 and 1902 while the American League didn't adopt it until the 1903 season. This should partially explain the AL's higher offensive numbers in '01, but how did the NL more than catch up in 1902? Was the NL reclaiming some of the talent that had jumped to the AL for 1901???

The National League played with the newly-created foul strike rule in 1901 and 1902 while the American League didn't adopt it until the 1903 season. This should partially explain the AL's higher offensive numbers in '01, but how did the NL more than catch up in 1902? Was the NL reclaiming some of the talent that had jumped to the AL for 1901???

WAR should be adjusted for the offensive environment of the individual leagues. (If it's not, there's a bigger problem.)

As a quick check on this - in 1936, the AL's run environment was about a run higher than the NL's (5.67 R/G to 4.71), thanks to differences in the balls used by the leagues. The NL's position players still out-WAR'd the AL's, 158.2 to 140.9.

1964 (21): 6-6 in AAA, with a worse ERA but better peripherals than 1963. Promoted in mid-season for good

Cleveland promoted pitchers pretty aggressively in the 1960s. Tommy John and Sam McDowell both reached the majors at a younger age than Tiant.

WAR should be adjusted for the offensive environment of the individual leagues. (If it's not, there's a bigger problem.)

I believe the average R/G is normalized for BBREF WAR but I know it doesn't consider the standard deviation in run scoring. Dan R's WAR is the only one I know of that is average AND standard deviation adjusted.

Dan R's WAR is the only one I know of that is average AND standard deviation adjusted.

For my player won-lost records, positional averages are calculated empirically every season (see here) and replacement level is set one standard deviation below average, where standard deviations are also calculated empirically every season (see here).

So, I think I'm finally ready to jump into the voting pool. I'm working on my ranking system, which is rWAR-centric right now. Also a fairly big boost for catchers, to put them on the same rWAR level as other positions. I may incorporate other WAR systems before the vote, depending on time. Here's what my top 25 looks like right now:

1 Roy Halladay
2 Mariano Rivera
3 Johan Santana
4 Gene Tenace
5 Urban Shocker
6 Kenny Lofton
7 Kevin Appier
8 Luis Tiant
9 Thurman Munson
10 Andruw Jones
11 Sal Bando
12 Roy Oswalt
13 Bobby Bonds
14 Andy Pettitte
15 Bob Johnson
16 Chuck Finley
17 Ernie Lombardi
18 Frank Chance
19 Chet Lemon
20 Dwight Gooden
21 Robin Ventura
22 Buddy Bell
23 Lance Berkman
24 Eddie Cicotte
25 Todd Helton

I know I'm missing the popular Negro League guys like Taylor and Redding. Not sure what to do with them, beyond more research. I'm just looking for input from people, to see if there is anything I should look at/incorporate/change.

The one thing that jumps out is your ranking of Gene Tenace. I'd assume his placement is the result of your boost to catchers, but he played only abut 60% of his innings there. He only once started 90 games at catcher in a season, so it's not an Ernie Banks/Rod Carew type situation where he was moved to first later in his career. He probably only deserves a partial catching boost.

It is hard to know what to suggest unless I know what you value in a player. Are you peak or career?

It does look like you have several modern 3B and CF which is what I would expect from rWAR from the posts I have already outlined. Who are your top SS and 2B candidates?

Yeah, Tenace stuck out. I'll have to add a bit of finesse to the catcher numbers.

I'm trying to value peak and quality over career, but still giving career results some respect. Topped ranked SS is Nomar at 37, Tinker not far behind. Top second baseman is Lazzeri at 49.

I'd suggest taking a closer look at Wally Schang. I feel he should do well with a good catcher boost.

Wally Schang is 35th right now.

I updated the catchers data to reflect time spent catching, and Tenace drops to 17th. Feel a bit more comfortable with that. Munson becomes the top ranked catcher at #10, and Lombardi drops down to 25.

Jimmy Key is in my top 25, but I drop him to add Helton. Mostly because I'm surprised he does so good in my system. Of course, once I add the NeL players, they'll both be off.

cookie - I think looking at the JAWS leaderboards would help you with your ballot. You want to use BBREF WAR and are more peak inclined. Those should be your kind of players. Don't forget to war adjust players like Rizzuto and Shocker and strike adjust 1980s players like Hershiser. Also, seasons of 154 games should be length adjusted by 105% to account for the fact that WAR is a counting stat.

P (post 1900) - Halladay, Vic Willis, Tiant, Shocker, Cicotte, Uhle, Appier, Finley, Hershiser, Walters, John, Tanana, Santana
C - Munson, Tenace, Posada
1B - Helton, Olerud, Tony Perez
2B - Kent, Fox, Lazzeri
3B - Bell, Bando, Ventura, Cey
SS - Fregosi, Campaneris, Garciaparra, Tinker, Bancroft
LF - Bob Johnson, Berkman, Jose Cruz, Luis Gonzalez, Veach
CF - Lofton, Andruw Jones, Willie Davis, Pinson, Cedeno
RF - Sosa, Bobby Bonds, Giles