hot topics

Well, just to say something, here's an off-the-cuff prelim I did at some point, without looking at any numbers.

1. Schmidt
2. Brett
3. Boggs
4. Mathews
5. Wilson (Yeah, that's probably too high.)
6. Baker
7. Santo
8. Molitor (Could be a little low)
9. Allen
10. Sutton (Not everybody's going to buy that one)
11. Collins
12. Groh
13. Hack (And I had those 2 the other way around on my Group 3 ballot, so that's changing.)
14. Robinson
15. Beckwith (Too low.)
16. Boyer
17. Evans
18. Nettles

Among non-electees, Bus Clarkson would be somewhere in the 12-14 range, I guess (but I'm his biggest fan), Tommy Leach would be between Beckwith and Boyer, and Ron Cey, Tony Perez and Bob Elliott would be between Evans and Nettles.

I found this info from Chris Cobb's Negro League estimates in my Hall of Merit files on my old computer. Here's the goods on Beckwith and Wilson:


The "pk" column refers to the player's 5-year non-consecutive peak and the "pr" refers to the total of the players WS above 20 in a season (AKA the jschmeagol method). I believe these were the latest available estimates: the John Beckwith WS are dated 2/13/05 and the Jud Wilson WS are dated 3/23/05. I think those refer to the dates Mr. Cobb posted them. . . .

IMPORTANT: I almost forgot to mention, the above estimates are based on the 154 game schedule, not 162. Also, according to the John Beckwith thread (post #311) his estimated time is supposed to reflect his character issues.

Using DanR's WARP, with no adjustments for military service, here are the Pennants Added totals:

John Beckwith (not available)
Ezra Sutton (not available)
Jud Wilson (not available)

And by my salary estimator:

1. Schmidt, $373M
2. Mathews, $273M
3. Boggs, $250M
4. Brett, $244M
(Larry, $200M before having the first half of his life)
5. Santo, $193M
6. Groh, $186M
7. Baker, $184M
8. Allen, $183M
9. Evans, $183M
10. Collins, $180M
11. Molitor, $177M
(Rolen, $176M)
12. Brooks, $170M
13. Nettles, $164M
(McGraw, $164M--can we get him in for 2009 please!)
14. Hack, $157M
(Leach, if he slots here, $155M)
15. Boyer, $148M
(Bell, $148M)
(Cey, $147M)
(Ventura, $146M)

Prelim:

1) Schmidt - Great glove man, strong peak, maintained production late into his career, and (along with Mathews) the best bat among the third sackers.
2) Mathews - As feared a hitter as Hammerin' Hank during their years in Milwaukee.
3) Brett - Despite his batting average, not the hitter Mathews was.
4) Boggs - Could seemingly get on base at will, strong glove.
5) Santo - What more can be said? Among the Hall of Fame's most offensive omissions.
6) Hack - One of the top leadoff men of the thirties, great glove as well. Really bridged the gap between the defense-first 3Bs of the previous generation and the offense-first 3Bs to come.
7) Molitor - Great leadoff man, decent glove at several positions. Couldn't stay healthy for a full season till they made a DH out of him, though.
8) Wilson - A truly great LH hitter, a seemingly adequate glove.
9) Collins - One of the best hitting third basemen of the dead ball era, also a very good fielder.
10) Robinson - Best fielding 3B ever, fair hitter with middling power. Offense was more important from third when he played.
9) Beckwith - A hitter who happened to play third.
12) Baker - Very good peak, but his career was short compared to a lot of the guys here.
13) Evans - Played forever, played consistently well. We all know he's been historically underrated.
14) Nettles - If Brooks Robinson never existed, would be remembered as one of the best fielding third sackers ever. Hit a lot of homers, didn't contribute much else with the bat.
15) Sutton - Seems not to have hit his peak until age 32. Good, long 19th century career. Strong defender and a good hitter, though not the best batsman of his generation.
16) Groh - Good D at a time when it was more important, could also get on base.
17) Boyer - Very good hitter and fielder, shortish career. I wouldn't argue against his enshrinement in the HoM, but I just don't think he was as good as the other players in the Hall.

Whoa, where's Dick Allen on my list? Something seems to have gotten lost. Well, I'd preliminarily slot him at #6, between Santo and Hack; but his shortish career and weak glove may drop him down a few slots (better bat than anyone on the list other than Schmidt and Mathews, though).

10. Sutton (Not everybody's going to buy that one)

I will. The finest third baseman of the 19th century, the guy was still the best career ML third basemen 50 years after he retired (peak is a different matter).

If he winds up near the bottom of the pack, I'm going to be extremely disappointed.

Without reading or considering the comments and discussion so far, which I will before final. There's a couple of tough choices here, nowhere near as tough as SS but probably tougher than at the other positions so far.

1. Schmidt

2. Brett
3. Mathews
4. Boggs

5. Baker
6. Santo
7. Wilson

8. Deacon White? I thought he was on this list.
9. B. Robinson
10. Molitor
11. Groh
12. Sutton

13. J. Collins
14. Beckwith
15. Hack

Not PHoM

16. Boyer
17. Da. Evans
18. Nettles

Not Ranked

Dick Allen, for reasons discussed above. Or, well, it's not discussed. I thought he was a 1B. Slot him in Deacon White's place preliminarily if you must. I might.

8. Deacon White? I thought he was on this list.

He was with the catchers, Marc. Did you vote in that election?

I agree with John regarding Sutton.

When you adjust for season length, you've got a guy with a 10,700 AB career, who put up a 119 OPS+ playing 3B when it was a key defensive position.

When he wasn't playing 3B, he was mostly playing SS, so he gains there.

Molitor had a 122 OPS+ in a similar length career, providing much less defensive value, as an example.

I'll probably have Sutton between 5 and 7 on my ballot.

3B position ballot (prelim)

1) Schmidt - 18th overall
2) Mathews
3) Brett
4) Boggs
5) Beckwith (Bat enough to get him in alone and he played 1/3 @ SS, not 1/3 @ 1B like Wilson or Allen)
6) Molitor
7) Santo
8) Jud Wilson
9) Darrell Evans
10) Heinie Groh
11) Stan Hack
12) Frank Baker
13) Brooks Robinson
14) Jimmy Collins
15) Dick Allen (tied with Collins but Allen loses all tiebreakers)
16) Graig Nettles
(Tommy Leach)
(John McGraw)
17) Ken Boyer
XX) Ezra Sutton - no idea exactly where to place him yet but 10th isn't unreasonable.

Molitor wasn't that great. Really low peak, esp. for those who value durability. He doesn't have a single 6-WARP2 season, which is starting-the-All-Star-game caliber. He wouldn't sniff my PHoM if he didn't add so much value on the basepaths (+76 runs per Fox fourth in the 1956-2007 period, after Rickey, Wilson, and Raines).

DL from MN, what's to love about Hack? How much you loppin' off for the war?

Nothing terrific about Hack, I did notice an error in how I was calculating (getting treated like a pre-shift 3B when he's more modern than that) and he slides below Baker and Robinson. Allen moves up above Collins too.

10) Groh
11) Baker
12) Robinson
13) Hack
14) Allen
15) Collins

My prelim:

1. Schmidt

2. Mathews
3. Brett
4. Boggs

5. Sutton

6. Santo
7. Evans
8. Groh
9. Robinson
10. Molitor
11. Baker
12. Allen
13. Collins
14. Nettles
15. Wilson

16. Hack
17. Beckwith

18. Boyer

It has nothing to do with the voting, but it's All-Star Game time again, so that means its RonStar time!

(Thanks again to Jeff M for the name)

For newer voters, RonStars are people who make the All-Star team (1) for the first time, (2) have no teammates named to the team, and (3) are not starting. As a broad generalization, these are players who are usually helped by the "one-per-team" rule, but not always. Last minute roster changes can and do change RonStar status. As of this morning (after starters are announced), there are only five potential RonStars for the 2008 All-Star Game:

AL

SP - JOAQUIM SORIA, KC. Kansas city is a great RonStar team, contributing such luminaries as Ellie Rodriguez (1969), Amos Otis (1970), Jeff Montgomery (1992), Kevin Appier (1995*), Jose Rosado (1997), Dean Palmer (1998*) Ken Harvey (2004*), Mark Redman (2006*) and Gil Meche (2007*). (* means permanent RonStar, in that they never made another All-Star Game) Mike Sweeney ruined it during the early 90's by making the All-Star team a lot. Soria's chances of getting in the game are slim, as the big four of Nathan, K-Rod, Papelbon, and Rivera will likely get the call before him.

SP - GEORGE SHERRILL, BAL. The Orioles are not a good RonStar team, only having Bob Turley (1954), Ken Singleton (1977) and Melvin Mora (2003). As the only lefty reliever, will probably get in if there is a LOOGY situation.

NL

SP - EDINSON VOLQUEZ, CIN. The Reds have really done well with recent RonStars. In their history, they have contributed Paul Derringer (1935), Grady Hatton (1952*), Adam Dunn (2002*), Aaron Boone (2003*), Felipe Lopez (2005*), and Bronson Arroyo (2006*). Volquez pitched Saturday, so he'll only get an inning. He's too good not to get some action.

OF - NATE MCLOUTH, PIT. Continuing a glorious tradition of Pirate RonStars with Murry Dickson (1953*), Frank Thomas (1954), Hank Foiles (1957*), Ken Brett (1974*), Carlos Garcia (1994*), Denny Neagle (1995), Jason Kendall (1996), Tony Womack (1997*), Ed Sprague (1999*), Mike Williams (2002), Jack Wilson (2004*), and Jason Bay (2005). The only true centerfielder on the NL roster will see some action in New York.

1B - ADRIAN GONZALEZ, SD. Padre RonStar immortals include Chris Cannizzaro (1969*), Cito Gaston (1970*), Nate Colbert (1971), Johnny Grubb (1974*), Randy Jones (1975), Dave Winfield (1977), Mark Davis (1988), Rondell White (2003*), Mark Loretta (2004), Jake Peavy (2005), Since Pujols is the DH, Gonzalez will spell starter Berkman at some point tomorrow night.

As some examples, Washington's Christian Guzman has no teammate in New York, but he made the team before, so he can't be a RonStar this year. Ryan Ludwick and Aaron Cook are also first-time All-Stars, and aren't starting, but they have teammates selected.

I like the "every team represented" rule, and always cheer for RonStars to get in the game, which keeps my interest going during a lopsided game in the late innings. I usually pick one favorite RonStar for each team. This year, its Sherrill and McLouth.

One can lose permanent RonStar status by being selected again. Those who are no longer lifetime Ronstars include:

Scott Kazmir (2006 RonStar with Tampa Bay)
Justin Duchscherer (2005 RonStar with Oakland)
Ryan Dempster (2000 RonStar with Florida)

Just to lend some relevance to this thread, here are Hall-of-Meriters and/or Hall-of-Famers who were once RonStars (but later lost that status):

Satchel Paige, StLB (1952)
Catfish Hunter, KC (1966)
Tom Seaver, NYM (1967)
Nolan Ryan, Cal (1972)
Don Sutton, LA (1972)
Gary Carter, Mon (1975)
Dave Winfield, SD (1977)
Dave Stieb, Tor (1980)

No HOM third basemen have ever been a RonStar.

I have a real time understanding Beckwith/Wilson significantly below Dick Allen. They are very comparable hitters with Beckwith and Wilson both being described in their threads as "Dick Allen clones" but both were considered to have more career value than Allen. I can see a peak only voter bunching them up but I really have a hard time understanding Joe D's ballot placements.

I have a real time understanding Beckwith/Wilson significantly below Dick Allen.

I have never really trusted the MLE's in general, but aside from that, Dick Allen has an argument for being as good as any hitter alive for about a decade or so. If you made a list of the top hitters alive, regardless of league, when Beckwith/Wilson played, how far down the list would they be? In other words, I am expressing doubt that either Beckwith or Wilson were really comparable hitters to Allen.

Interesting... it seems that there is not a single thirdbaseman that made the HoM that didn't also make my PHoM.

Jimmy Collins took 54 years to join the PHoM after the HoM, but he still beats Graig Nettles on this ballot, though Nettles ahd the record the other way (getting in my PHoM 12 years before getting in the HoM.)

Prelim

1. Mike Schmidt
2. Eddie Matthews
3. George Brett
4. Wade Boggs
5. Ron Santo
6. Paul Molitor
7. Frank Baker
8. Darrell Evans
9. Stan Hack
10. Dick Allen
11. Brooks Robinson
12. Heinie Groh
13. Jud Wilson
14. Ezra Sutton
15. John Beckwith
16. Jimmy Collins
17. Graig Nettles
18. Ken Boyer

... as good as any hitter alive for about a decade or so ...

This is from Dizzypaco's comment above. Let's phrase the question this way: suppose either Beckwith or Wilson had traded places with Jimmy Foxx. (The ages and timings aren't perfect, but there's at least an overlap). Foxx as a major leaguer played predominantly 1B, but the hypothetical Foxx-in-the-NgL might have been a 3B, or more likely a catcher. (Of course, Beckwith was a SS.) I don't know where all that leads us, but it seems to be part of the frame of reference for Dizzypaco's comment.

Well, Jud Wilson was a pretty good 3B, I mean with the glove, and Beckwith was a SS/3B. Somebody please refresh my memory about Allen's defensive value.

It looks to me after X number of ballots that the easiest thing to do with the 19C and NgL once again is just to say, ah, the hell with 'em.

Joe and John, I think you are overrating Ezra Sutton. Here's why:

Joe sums up Sutton's case like this:

When you adjust for season length, you've got a guy with a 10,700 AB career, who put up a 119 OPS+ playing 3B when it was a key defensive position.

When he wasn't playing 3B, he was mostly playing SS, so he gains there.</i>

Sutton's raw career OPS+ is helped by the fact that his late peak came in the longest seasons of his career. If you adjust his seasons to 162 games individually to weight his offensive performance equally for each season, his career OPS+ is only 110. Graig Nettles (OPS+ 110) is thus a much more appropriate offensive comp than Molitor.

Third base was a key defensive position, but offensive performance by third basemen was not horrible during Sutton's era: shortstop was much much lower. For his career, Sutton had only 445 RCAP. To put this in context, Deacon White had 711 RCAP: that is what a great hitter at C/3B looks like in this era. Hardy Richardson had 408 RCAP, in a much shorter career. He's a better comp for Sutton in terms of offensive value.

So both by properly weighted OPS+ and RCAP, Sutton's production is comparable to players in the lower echelon of the HoM, not to the higher echelon guys.

(I'll also note that he really fattened his OPS+ in years with weak competition: 1871 and 1875, which were the NA's weakest years, and 1884, which was the NL's weakest year b/c of competition with 2 leagues, are 3 of his 5 big offensive seasons. I don't dock him for that directly, but I don't give him the benefit of the doubt when it comes to assessing his offensive value.)

Moreover, Sutton's defense was not outstanding. He played some shortstop, yes, but WARP1 does not see his defense as exceptional: He is earns 80 season adjusted FRAA for his career at third base, but he was below average at all the other positions he played, so his total FRAA is more like 40. Compare his OPS+ comp Nettles, with 135 FRAA.

Now, 3B was a more important defensive position during Sutton's era, so maybe that's a wash, but I don't see Sutton being comparable to guys like Santo and Molitor, who sit in the tier below Matthews/Brett/Boggs. He doesn't have the offense to match with them, nor the defense, in Santo's case. He is more comparable to the career candidates down in the lower part of the list: Robinson and Nettles. It's not time-lining that puts him there, it's the total package on offense and defense that he brings.

Third Base preliminary ballot.

From one of the strongest positions (SS), we turn to one of the weakest (3B). Schmidt would rank second if he were ranked against the shortstops, but he is comparable to Lloyd, Ripken, and Vaughan, not to Hans. Mathews, Brett, and Boggs are comparable to the top 10 shortstops, but the quality falls off rapidly after that. Santo, who is #7 here, would be at #20 my shortstops list, as I see it. Lou Boudreau and Pee Wee Reese are his best comps, I think.

I. All-Time Top 10

none.

II. Inner-Circle HoMers
1. Mike Schmidt. Total = 492. Another election in which the top of the ballot is very easy. Mike Schmidt, while not an all-time top ten player, is comfortably in the inner circle. No one else is close.

III. Among the best players of their generation
2. Eddie Mathews. Total = 393. Hit like Schmidt, but a mediocre fielder.
3. George Brett. Total = 387. Brilliant when healthy: a player I was always excited to watch. As a hitter, he could do anything.
4 Wade Boggs. Total = 384. Nearly Brett’s equal in quality, and was a hit-machine during his peak, but not nearly as exciting a player as Brett.
5. Jud Wilson. Total = 330.5. Like Buck Leonard at first base, he got a late start in the Negro Leagues. I think he was probably closer to Mathews, Brett, and Boggs than my system shows.

IV. Obvious HoMers
6. Paul Molitor. Total = 309. George Brett lite, but his injury problems reduced his value considerably.
7. Ron Santo. Total = 295. Great prime; marvelously durable for the hot corner.
8. Frank Baker. Total = 265. Fabulous peak. If he hadn’t left the game twice before finally retiring, he would probably rank higher. I rank him above my system’s total because the missed time saps his career value.

V. Solid HoMers when you read the fine print
9. Heinie Groh. Total = 266. Peak was not far behind Baker’s. Unjustly forgotten.
10. Brooks Robinson. Total = 267. Top modern defensive third baseman.
11. Dick Allen. Total = 277. Could mash, but a defensive liability in a short career. I’d rather vote for him as a first baseman, but since he’s in this pool, here’s where I rank him at third. I don’t buy much of the “he hurt his teams” argument, but really, I’d rather have had Brooks Robinson as my team’s third baseman—hence the lower placement of Allen.
12. Darrell Evans. Total = 275. Odd career shape lowers him in the rankings somewhat, as does the uncertain value of his defense.
13. John Beckwith. Total = 264. The Dick Allen of the 1920s.
14. Graig Nettles. Total = 254. Almost as good defensively as Brooks Robinson; a better hitter than is usually recognized.
15. Jimmy Collins. Total = 249. Top pre-modern defensive third baseman. Just enough career to be a solid HoMer.
16. Ezra Sutton. Total = ???? Probably the best third baseman before Collins and McGraw, and managed a longer career at the position than anyone else in the nineteenth century, but didn’t put together a sustained peak, wasn’t a great fielder, and tended to have his strongest years as a hitter in seasons when league quality dropped. I support his inclusion in the HoM, but I think he is often overvalued. See my prior post for a more detailed accounting.
17. Stan Hack. Total = 246. Great at getting on base. Just enough defensive value to be a solid HoMer.

VI. Almost as good an argument to be out, as to be in
18. Ken Boyer. Total = 232. Nice prime, but short career. This placement is without military service credit, which he might deserve. Better than Bob Elliott and Ed Wiliamson, so not a bad choice at a weak position, but not as good as John McGraw.

Chris Cobb, why so high on Molitor? No peak, couldn't stay on the field, all those years at DH...what's pulling him up so high on your system?

Chris Cobb, why so high on Molitor? No peak, couldn't stay on the field, all those years at DH...what's pulling him up so high on your system?

Well, despite all that, he still had lot of very valuable seasons--I count 9 seasons of 4.2 WAR or better, and another 4-5 more in the 3+ range, and that's before an adjustment for playing in the DH league. That adds up: my system does like career value, and if there's enough of it, a player doesn't have to sport a great peak to do quite well. As so few of the third basemen who were really offensive forces had long careers, Molitor rises high.

Really, I think it's more an indication of how weak the competition is here. Molitor would be around 18 if he were dropped into the shortstop rankings.

His high ranking definitely isn't affected by my use of WARP1 and WS as well as WAR: if I used only the WAR score, he would still rank just behind the Big 5.

"Sutton's raw career OPS+ is helped by the fact that his late peak came in the longest seasons of his career. If you adjust his seasons to 162 games individually to weight his offensive performance equally for each season, his career OPS+ is only 110."

Wow, I had never considered that factor seriously for anyone (peak in the longest seasons). Thanks, Chris, as always.

We've had 9 lists, 2 of which didn't include the NeLers or Ezra Sutton. Still, pretty clear tiers are already evident. The rating is just the sum of the rankings. A low number is better. Wilson, Sutton and Beckwith just get their average for their missing values.

1. Schmidt 9--the Honus Wagner of 3B

2. Mathews 21
3. Brett 27
4. Boggs 33--the Lloyd, Ripken and Vaughan of 3B; no votes outside of 2-4

5. Santo 54--no votes outside of 5-7, the Appling of 3B

6. Molitor 73--I agree, not that good, probably a tier too high
7. Baker 78--a little under-rated but probably in the right tier
8. Wilson 79--under-rated but probaby in the right tier, big range from 5-15, which I understand, but he certainly oughta be closer to 5 than to 15

9. Allen 91
10. Groh 92--big range from 6-16, why?
11. B. Robinson 96
12. Da. Evans 97--big range from 7-17, which I understand

13. J. Collins 104
14. Sutton 106--I suppose Sutton and Collins could be a separate tier; it would be too much to bump them up with Allen and Groh
15. Beckwith 116--huge range from 5-17, which I, well, I don't understand a 5, were you thinking of Wilson?
15. Hack 116--range from 6-16, hard to see a #6 for Hack

17. Nettles 138
18. Boyer 150--obviously a big mistake

Wilson, Groh, Beckwith, Hack and Evans seem to be the tougher choices (bigger range) so far.

This probably won't make any difference to anybody, but check post #368 in the Beckwith thread. The MLEs, based on Holway, see 1922 as an off-season for him, when it was in fact one of his best seasons--.388/.435/.652 in 52 games vs. NNL opponents, .357/.415/.589 in 66 games against all top black opposition. He led NNL third basemen in both range factor (3.53, NNL ave 3.17) and errors (.913 fielding pct, .938 NNL ave).

Wow, I had never considered that factor seriously for anyone (peak in the longest seasons). Thanks, Chris, as always

In not considering that factor, I don't think you have missed much. I had been curious about this effect for early players for a while, so when the HoM-not-HoF ranking project gave me cause to look again at 1870s and 1880s players, I worked up the numbers. Sutton was, I think, the only player whose pro-rated OPS+ differed from his raw OPS+ by more than 2 points. But in that one case, it's a meaningful effect.

Offense-only chart. My usual system, based on RCAA. No extrapolation for season length. MLB players only.

An apology and a few notes that I remember from old analysis:

1. I apologize for not understanding how forgiving this group is about philosophy. I thought the group was MUCH more strict about evening out the decades than it seems now to be. So I apologize for obsessing when I got to George Wright. I'm still going to vote that way, because I think it's right, but I do understand that my opinion is not a hard consensus here. I badly overestimated the comments about hating timelining.

2. I don't dock Baker, Collins or Groh for playing time. They all three played about 15 years, and there's no one out there who's playing much more. Third base seems to have been an unusually physically demanding position in the early 20th century, and 15 years is about all anyone could get out of his body. I treat the three of them, when comparing to modern third basemen, as if they had played 18 years or so. Well, let's say 16 or 17 for Baker.

3. There's no one really like Home Run Baker playing third base anywhere near his time. I give him a lot of credit for that. Right now, my prelim list has him #2, ahead of even Brett. I do understand that this will probably be his highest vote.

4. One fun thing I did when I got the Historical Abstract was to take Bill's "overrated" players and look at who he had ranked above them that were either contemporary or earlier. George Sisler, for example, is ranked 24th at first base, but there's no one above him until you get back to the 19th century. So when the HoF voters decided he was the greatest first baseman of the 20th century, that was fact at the time. Well, the same thing applies to Jimmy Collins, who is also generally credited with inventing modern third base play, and is true of Pie Traynor (who isn't even on this list), unless you count either Stan Hack, who is actually later, or Home Run Baker. I give Collins credit there, and would give Traynor credit if he were here. On the other hand, the inability of HoF voters to understand just how great Baker was indicates that, early in the century, defense was valued very very highly. Everyone knew he could hit like no other third sacker.

5. What happened to Martin Dihigo? I just assumed he would end up here, although it is very well documented that he could literally play anywhere.

I have a real time understanding Beckwith/Wilson significantly below Dick Allen. They are very comparable hitters with Beckwith and Wilson both being described in their threads as "Dick Allen clones" but both were considered to have more career value than Allen. I can see a peak only voter bunching them up but I really have a hard time understanding Joe D's ballot placements.

Adding Dick Allen to the Negro League chart above and doing a Q&D;adjustment to 162 games shows that, according to the MLEs, while Beckwith's & Wilson's careers were longer, and their defense better, Allen was just in another class offensively than the two Negro Leaguers and it shows, particularly in the difference between peak values.

Comparisons between Major Leaguers and Negro Leaguers were never meant to be more than creating a generalized impression concerning ballplayers who we knew so little about. Both Beckwith & Wilson were big bat infielders not known for their glove. Both were known for their temper--Beckwith to the point where Holway has stated (according to the NBJHBA) "his character deficiences often negated his performance value." Combine that with a short career and Dick Allen seems a fairly good comparison, but like Dizzypaco said it would take a lot to equal Dick Allen's offense and the MLEs are only an educated guess, which at this point don't point to Dick Allen, IMO, but maybe a poor man's version.

Also, Beckwith may have played SS/3B, but the MLE grade for his defense is a D, I believe; however, I have no idea how accurate that is since it was entirely based on his reputation at the time the MLEs were made. There was also a lot of discussion about how much of a clubhouse cancer Beckwith really was; lots of information can be found in the rather large thread dedicated to Mr. John Beckwith.

Well, Jud Wilson was a pretty good 3B, I mean with the glove, and Beckwith was a SS/3B. Somebody please refresh my memory about Allen's defensive value.

My impression from the discussion on Jud Wilson was that he was an offensive guy first and that his glove wasn't something to brag about (like Beckwith). In that file I mentioned earlier his defensive grade is marked as "C+,B+". I'm not really sure what that means other than that one grade refers to 1B and the other to 3B. I guess I could go back and read his thread, but it's getting a bit late so I'll refresh my (hazy) memory concerning Negro League defensive analysis some other time. . . .

O's chart is interesting to say the least. Can't wait to see your ballot. Maybe Mike Schmidt won't be unanimous after all.

Brock, nice post. We all oughta remember that 3B was a lot different before 1930 than it has been since. My early-3Bmen rankings may be higher than most others, altho I'm not quite bold enough to slot F Baker at #2.

re: the case for/against Molitor. For a guy who 'wasn't durable', & got 'hurt a lot'; he sure had a career ANYWAY, didn't he? Name all of the men who played more games at "throwing infielder" positions who crossed home plate more times in their careers than Paul Molitor. Hint: Criag Biggio, limping along to his final career ##s, doubled the total. The man created runs for his team, and was a good glove while he played. Oh, and then you have his post-season stats: an OPS of about 1050 in 29 games. He scored 28 runs in those 29 games, and drove in 22. "Mister October"'s post-season stats look pathetic compared to Molitor.

The real question isn't if Paul belongs in the upper half of our group - it's more like should Molitor take the #4 spot form Boggs, a man who got on base extremely well but was a lousy runner and thus in rea life generated far fewer runs for his team than the RC formulae indicate.

re: Ezra Sutton - so underrated by the masses that we're now in danger of overrating him. Unless we go all Chris Cobb and underrate him :) A cursory look at BP's 'translated stats' show him to not be an offensive force, but he had a long career at a tough position and flashed a good glove. Really, I'm very glad we discussed him significantly, and my understanding of 19th century ball is greatly enhanced by finding this gem of a player. He won't make my top 7, but he seems a good comp for J Collins.

re: Beckwith and Allen. Y'all may as well get ready now to howl at my placements of the troblemakers.

Thanks for the discussion, it is clear I have been overestimating Beckwith's career length and slightly underestimating Wilson. Wilson seems comparable to Gwynn and Clemente offensively. I'm still having a really hard time (as a career voter) understanding either player below Dick Allen. Allen was a pretty terrible defender, even considering his time at 1B.

I can't get Beckwith below the Evans-to-Hack grouping so he'll end up somewhere in there. Wilson is going to move up to claim the #5 slot I had for Beckwith. Chris has me convinced that Ezra Sutton should place right behind Jimmy Collins at this point (ahead of Nettles).

This whole exercise has been fruitful. The only two guys I still don't understand at all how to place are Lip Pike and Al Spalding.

O's chart is interesting to say the least. Can't wait to see your ballot. Maybe Mike Schmidt won't be unanimous after all.

On the one hand, I think there is a slight edge of underappreciation of just how good a hitter Mathews was. Well, among the mainstream, it's more than slight. Mathews had to creep up on the HoF ballot for several years before he was elected; he should have been first ballot by acclamation. When some writer lists or mentions great third basemen, Mathews is likely to be the one he forgets about.

But against that are two other considerations:

The first, and bigger, consideration is defense. Schmidt was an outstanding defender. Mathews wasn't. In an earlier generation, Mathews wouldn't have been a third baseman at all - he'd have been LF or 1B.

The second consideration is Dan R's favorite point: it was easier to achieve offensive dominance in the 1950's than in the 70's-80's. Mathews wasn't as good an offensive player as Mickey Mantle; Schmidt doesn't have a Mantle to look up to. Of course that argument shouldn't be about Mantle, as there's no reason for players at that level to be evenly distributed in time. But there's still a point there. (Mathews did play in the stronger league.)

As for comparing Schmidt to Brett and Boggs: if you consider prime and start looking at, say, years 4-10 on my chart, Schmidt does pull a clear offensive edge on those two, and also has his defense. And recall that I didn't extrapolate for season length - I don't think that matters much for 154 vs. 162, but it does matter for 1981, considering what Schmidt was hitting that year.

I will put Schmidt first on my ballot - but putting Schmidt first is not the kind of no-thought-needed exercise that putting Wagner first on the shorstop balot was.

Chris Cobb, why do you make an adjustment to my WARP for playing in the DH league? Those adjustments are already baked into the replacement levels (0.6 WAR per full season lower in the AL than in the NL) and the standard deviation regression. No further effect is necessary...

Whoa, TomH, where is this coming from? Who cares about Runs? This isn't the 1880's and Harry Stovey, it's the 1980's...don't we have some moderately more sophisticated tools at our disposal? You're certainly right that after factoring in baserunning, Molitor created more offensive value above average than Boggs did. Boggs is ahead by 8 wins on raw hitting, but Molitor is a massive 13 wins ahead on baserunning, leaving an advantage of about 5 wins to Molitor.

However, that spread isn't anywhere close to as big as the gap in career defensive value between them. First, Molitor was not "a good glove when he played," he was an average one. He is -9.7 runs for his career in SFR and +10 by TotalZone. Boggs, by contrast, was a *damn* good glove: +91 runs according to SFR, +61 per TotalZone, and +48 just for the 1987-98 period covered by Chris Dial's Zone Rating numbers. So that's about +70, which is around 7.5 wins over Molitor given the run environment. Total value above average is +2.5 to Boggs.

But then comes the real kicker: Boggs was a real third baseman! A replacement player, in Boggs's playing time (2,215 games at 3B, 108 at DH, and 67 at P, with strike adjustments), would have been 31 wins below a player hitting and fielding at the league average. By contrast, a replacement player in Molitor's playing time (1,174 games at DH, 791 at 3B, 400 at 2B, 197 at 1B, 57 at SS, and 50 in the outfield, with strike adjustments) would have been just 21 wins below a player hitting and fielding at the league average. The result is a very beefy 12.5-win career advantage for Boggs. That's really big.

Then, if you care, obviously the peak is no comparison. Boggs had seven straight MVP-caliber seasons, while Molitor never once had a year good enough to justify starting the All-Star game. I just don't see how an honest voter could have Molitor even within sniffing distance of Boggs.

Not a voter, but again, there is an argument here which interests me. I'm not sure I buy the arguments that Brett deserves to be ranked above Boggs.

Some peak analysis...

Brett's best OPS+s:
203 (117 games, 112 at 3B)
178 (155 games, 152 at 3B)
158 (123 games, 102 at 3B)
153 (142 games, 1 at 3B)
149 (157 games, 0 at 3B)

Boggs's best OPS+s:
173 (147 games, 145 at 3B)
166 (155 games, 151 at 3B)
156 (149 games, 149 at 3B)
151 (161 games, 161 at 3B)
150 (153 games, 153 at 3B)

I leave this for people to see for themselves, but superficially, Brett looks like the better peak hitter, but several of his best seasons (including his superlative 1980) happened in years in which he missed quite a few games, and others happened in the latter part of his career by which time he'd been moved away from third base. For those people who care about consecutive-years peaks, subsets of Boggs's 1982-1990 will likely (once games played are taken into consideration) beat any section of similar length in Brett's career (especially if you restrict Brett to his seasons at 3B).

Career length:
Brett played 300 more games overall, but Boggs played 600 more games at 3B. (Brett spent roughly 35% of his career at 1B and DH). Brett has a higher OPS+ (135 to 130) but was (possibly) not as good a fielder (witness Boggs's Gold Gloves and the fact that Boggs was never moved off third, unlike Brett). Brett last played 3B regularly in 1986 at the age of 33. I submit that we shouldn't give Brett a ton of credit for his years after that when ranking him against Boggs at 3B.

All of the above could be gone into in more depth. For now, I'll strongly suggest that people who rank Brett over Boggs re-examine their numbers. I think Brett tends to get a boost in these sorts of discussions (perhaps too much of one) based on his reputation (gamer, clutch), but is not properly debited for his injuries, his (possibly) inferior defense, and the time he spent playing first or DHing. I can't argue Boggs/Brett over Matthews or Schmidt, but I'm thoroughly unconvinced that Brett belongs above Boggs.

Chris Cobb, why do you make an adjustment to my WARP for playing in the DH league? Those adjustments are already baked into the replacement levels (0.6 WAR per full season lower in the AL than in the NL) and the standard deviation regression. No further effect is necessary...

My understanding is that the DH adjustment is something you added into WAR version 2. I am still using WAR version 1 numbers, because I got a long way with them and have no time to update everyone to the new numbers. These posit a single major league replacement level. Therefore, I apply a DH adjustment for AL post-DH players. If I am mistaken in my understanding that adjustments for DH were not built in to WAR version 1, then I should obviously adjust my treatment of Molitor and a host of other post-1972 AL players.

Third Basemen ranking. MVP/AS are my own calculations, but for Negro Leaguers I use some combination of Holway/James MVP and AS selections. Monster = 15.0 WARP1 or WARP2 and above. Great=10.0 WARP1 or WARP2 and above. Interesting that no third baseman has ever had a Monster WARP1 or WARP2 season.

1. Mike Schmidt. 26.3 bws/700PA, 9 MVP, 14 AS. No Monster, 12 Great (1974, 1975, 1976, 1977, 1979, 1980, 1981, 1982, 1983, 1984, 1986, 1987). High – 12.9 W1 (1980).

2. Eddie Mathews. 26.8 bws/700PA, 9 MVP, 14 AS. No Monster, 7 Great (1953, 1954, 1957, 1959, 1960, 1961, 1963). High 13.3 W2 (1959).

3. George Brett. 22.4 bws/700PA, 6 MVP, 12 AS. No Monster, 5 Great (1976, 1977, 1979, 1980, 1985). High 11.7 W1 & W2 (1980).

4. Wade Boggs. 20.8 bws/700PA, 6 MVP, 12 AS. No Monster, 6 Great (1983, 1985, 1986, 1987, 1988, 1989). High 11.8 W2 (1989).

5. Jud Wilson. 22.7 bws/700PA (MLE), 1 MVP, 11 AS.

6. Paul Molitor. 21.2 bws/700PA, 4 MVP, 13 AS. No Monster, 1 Great (1982). High 10.0 W2 (1982).

7. Frank Baker. 24.7 bws/700PA, 4 MVP, 10 AS. No Monster, 6 Great (1909, 1910, 1911, 1912, 1913, 1914). High 14.4 W1 (1912).

8. Ron Santo. 19.0 bws/700PA, 4 MVP, 8 AS. No Monster, 6 Great (1963, 1964, 1965, 1966, 1967, 1968). High 14.1 W1 (1964) He also had 14.0 in 1967 as the only 3B to crack seasonal 14 WARP1 twice.

9. Brooks Robinson. 14.7 bws/700PA, 2 MVP, 11 AS. No Monster, 3 Great (1962, 1964, 1967). High 12.1 W1 (1964).

10. Dick Allen. 29.9 bws/700PA, 6 MVP, 9 AS. No Monster, 4 Great (1964, 1965, 1966, 1972). High 13.5 W1 (1964).

11. John Beckwith. 23.0 bws/700PA (MLE), 2 MVP, 7 AS.

12. Heinie Groh . 20.0 bws/700PA, 3 MVP, 8 AS. No Monster, 6 Great (1915, 1916, 1917, 1918, 1919, 1920). High 13.7 W1 (1917).

13. Jimmy Collins. 17.2 bws/700PA, 3 MVP, 10 AS. No Monster, 5 Great (1897, 1898, 1901, 1903, 1904). High 12.1 W1 (1898).

14. Darrell Evans. 19.2 bws/700PA, 4 MVP, 11 AS. No Monster, 2 Great (1973, 1974). High 12.7 W1 (1973).

15. Stan Hack. 20.6 bws/700PA, 3 MVP, 11 AS. No Monster, 1 Great (1945). High 12.0 W1 (1945).

16. Ezra Sutton. 18.4 bws/700PA (after 1875), 2 MVP, 10 AS (starting 1871). No Monster, 1 Great (1884). High 10.8 (1884).

17. Graig Nettles. 15.7 bws/700PA, 1 MVP, 8 AS. No Monster, 1 Great (1976). High 11.3 W2 (1976).

18. Ken Boyer. 17.9 bws/700PA, 1 MVP, 8 AS. No Monster, 3 Great (1956, 1961, 1964). High 10.6 W1 (1964).

Chris Cobb, to be totally honest with you, I don't remember. But there's a real easy way to check: take two players with a SFrac of 1.00 at the same position at the same year in different leagues. Do they have the same Rep2 value? If so, then you need to adjust for the DH (0.6 wins/162). If not, then the adjustment is already baked in.

I just realized I left a line off of #31, because I didn't even have him listed with the third basemen:



The summary number I extract out of that puts him about even with Boggs offensively. I agree that he has to rank below Boggs becuase he had considerably less career defensive value.

Preliminary, with the two Negro Leaguers not placed as yet:

1. Mike Schmidt
2. Eddie Mathews - The gap between #1 and #2 is not as large as I once thought, and almost all of it is glove-related. Schmidt leads in OPS+ 146-145, OWP .740-.727, Mathews in EQA .318-.317. Schmidt’s lead is much larger in the metrics that encompass fielding.
3. Wade Boggs - Has the edge on Brett in several metrics (except WS). His real edge, for me, consists of 6 MVP-worthy seasons to Brett’s 4.
4. George Brett - I once thought more highly of Brett but his career at 3B was not all that long and, defensively, not all that glorious, despite one Gold Glove.
5. Ron Santo - I’m still hopeful about Cooperstown, but meanwhile…
6. Paul Molitor - Certainly the career numbers allow him to be placed this high, but I’m not certain it is warranted by other considerations.
7. Home Run Baker - Truly outstanding during his time, a shame it was disrupted.
8. Brooks Robinson - Seems to me his overall value is over-rated, although his defense was truly great. A lot of other Orioles had something to do with the team’s success in Brooksie’s years.
9. Dick Allen - I don’t usually think of him as a third baseman (652G).
10. Stan Hack
11. Eddie Collins
12. Heinie Groh - Numbers 10-12 don’t arouse any strong feelings in me, other than to say that they all seem worthy of inclusion in the HoM.
13. Graig Nettles - Should have had a few more GG. His 1971 season was probably the best ever for defense among third basemen.
14. Darrell Evans – I’ve always felt that his peak seasons were too spread out to qualify him as a really great player. Just too many so-so seasons mixed in.
15. Ken Boyer - Another good but not great player.
16. Ezra Sutton – I just can’t see him in any higher place on this list. A couple of good years but they came when the talent was thinnest.

(Jud Wilson and John Beckwith not yet rated.)

> Eddie Collins

Wrong Collins...

"Who cares about Runs...don't we have some moderately more sophisticated tools at our disposal?"

Given the choice between the non-sophisticated metric of runs scored, and your subjective 'all-star games started', I and many others might feel the former a better stake in the ground :)

Look, my point for mentioning runs was to quickly and simply refute the "Molitor was always hurt" argument, which doesn't hold much water if you look at his career totals.

I think Boggs is underrated by MLB fans; the fact that he did not even make the final 100 on the all-century team is lousy. How people have Brooks Robby ahead of him I don't know. I was a Red Sox fan in the 1980s, and appreciated those .440 OBAs. But....

Defensive value: OK, Boggs' career at 3B vice Molitor's all-over-the-diamond map might give Wade a 10-win edge. That's giving zero points for Molitor's flexibility, in say, moving to a new position (3B) for the good of the team, and then having them spank the league for their first and only pennant ever, and come one poorly-timed injury away from a deserevd World Series title. Is that worth ANYthing? That's also giving zero points for the tougher job, batting-wise, for being a DH. A few studies have shown baters hit slightly less well when DHing, and I think our 'replacement-level DH' models do not account for this very well.

Boggs took possibly more than his share of advantage of Fenway; .354 at home, .302 on the road, with 50% more wall-doubles at home. He may have been less valuable in other places.

Defense: BP and WS do not show Boggs significantly better than Molitor. Their reps while active seemed about equal as well. I will defer somewhat to the pbp methods smart people have developed for 1987-current time, but I think the ##s you gave overstate the case somewhat.

Win Shares has Molitor ahead on career, 412 to 294. I use a homemade WSAB, "base" being 13 WS per 162 games. They come out even using my WSAB. And again, WS replacement level for DH is pathetically low, AND it does not adjust for baserunning. If I only used WS, Molitor would come out ahead.

You have Molitor 13 wins ahead on baserunning. That's believable.

So, Molitor ahead on batting + baserunning, close on fielding, Boggs ahead on replacement level, Molitor small edge in flexibility, Molitor far ahead in post-season performance, which your methods didn't seem to account for. Seems close to me.

Yes, Boggs has a better peak argument. Even for a career voter like me, that puts Boggs on top of Molitor on my ballot.

"I just don't see how an honest voter could have Molitor even within sniffing distance of Boggs."
So that makes me...?

When evaluating Boggs, its worth keeping in mind that his primary offensive value was getting on base (which would typically lead OPS+ to underrate his contribution). However, I think Boggs' lack of speed is more relevant for him than for a typical slow player, given that getting on base was his primary value, and that he batted at the top of the lineup. He scored a lot of runs, but not as many as you'd might think given his on base percentage and the lineup batting after him (in Fenway Park), and he sure didn't drive in many. Jim Rice's enormous GIDP totals are partially a result of Boggs' baserunning, IMO.

I'm questioning how many actual runs he (or a player of his full range of abilities) was responsible for. Typically, I'd be very sympathetic to the durability argument against Brett, but in this case, its not hard for me rate Brett over Boggs, and I'm not sure Boggs isn't being overrated in general. He had a very good peak, but I don't think he really contributed that much during the 1990's, and I don't even think I would consider him a devastating player even in his best years.

A few studies have shown baters hit slightly less well when DHing,

The problem is how to know when a hitter is DHing because he's recovering from injury and when he'd DHing for the heckuvit. Manny Ramirez might well hit better as a left fielder than as a DH, but that may be because he's more likely to be DHing when his hamstrings are bothering him, and so forth.

I didn't say actual All-Star games, I said "good enough to start the All-Star game," which is 6 WARP2.

No, the "Molitor was always hurt" argument has nothing to do with his career value. It matters in terms of peak/pennants added, because he didn't contribute nearly as much value in any given discrete season as your average HoM'er due to his perennial missed games.

No, I don't think Molitor's willingness to shift to a new position is worth anything at all. If he HADN'T been willing to do what his manager asked him, I think there would be fair grounds to dock him for that. But I certainly don't see why he somehow deserves brownie points just for following orders. Does Banks get extra credit for moving to 1B? Will ARod get extra credit for moving to 3B? This is ludicrous.

I also see absolutely no reason to give a DH extra batting credit. The fact is that the minor leagues are full of guys who could be 100 OPS+ DH's in the major leagues (see Nate Silver's study). That is the appropriate baseline to measure DH batting against--period. Furthermore, for what it's worth, I think those studies are BS, because position players are likely to be nursing nagging injuries if they're put at DH.

Yes, Fenway was a dream park for Boggs. That won games for the Red Sox. Do you dock Mel Ott for playing in the Polo Grounds?

I don't see why we would pay the slightest attention to FRAA and WS now that we have stats using Retrosheet data, which show a far, far higher correlation to today's PBP metrics. SFR, TZ, and DRA are very, very good, and if they see things one way and FRAA/WS see them another, I'm going to weight FRAA/WS precisely 0 in my analysis (since they come out with a 0.00 weighting in the multiple regression analyses I've done). (DRA isn't available for Molitor because he so rarely played 130 games at one position in a season, so I left that stat out).

What, exactly, is Win Shares bringing to the table that my analysis is leaving out here? What additional information does it include that makes it relevant to this debate?

It makes you very, very misguided. :)

Dizzypaco, again, we have precise measures for this stuff, there's no need for more guesswork! Boggs' baserunning cost his teams 45 runs below average over the course of his career. You should knock about 3 points off his career OPS+ for it (but that's partially counteracted by needing to adjust for the OBP-heavy shape of his production).

Typically, I'd be very sympathetic to the durability argument against Brett, but in this case, its not hard for me rate Brett over Boggs

Why? You say that Boggs doesn't deserve as much credit for his OBP because (essentially) Jim Rice kept hitting into double plays. Ok, for the sake of argument, we'll fault Boggs for this. How much do we fault Boggs? And for any non-base stealer, how do we know they wouldn't have been just as much "at fault" if they'd been on base for Rice?

Back to how much do we fault Boggs. How many runs do we subtract from him for his (alleged) baserunning problems? And is that enough to make up for the (possibly) superior defense, superior OBP (possibly more important than SLG), and more games spent at the position in question? And are we yet properly taking credit away from Brett for his years spent extending his career as a DH/1B (which may or may not have kept him more healthy, which may or may not have allowed him to concentrate more on his hitting, and which almost certainly pads his OPS+ numbers relative to third basemen unfairly since he wasn't actually a third baseman for the last third of his career)?

Offense-only chart. My usual system, based on RCAA.

Brett  74 73 57 52 52 52 41 38 38 30 27 25 21 20 18 12  2  -5 -6 -19
Boggs  72 68 64 61 58 45 34 31 29 29 21 19 14  8  5  4 -4 -10 

And here's a BRAA chart, restricted to seasons in which the player in question played a majority of his games at third base. (I'm using baseballprospectus's numbers, incidentally).



Notes:
Brett's 22 came in a strike year (1981). So did Boggs's 25 (1994) and, to a lesser extent his 18 (1995).
Boggs played a few more games/innings at 1B than at 3B in 1982 so I did not include it.

Are there voters who give credit for time missed due to managerial incompetence? (See Edgar Martinez and Wade Boggs for two, certainly; possibly Willie McCovey -- I don't know his story particularly well; arguably some black ballplayers as well. Not saying that Boggs's missed years are morally equivalent to Satchel Paige's or Minnie Minoso's, but I'm not sure how voters account for this).

"No, I don't think Molitor's willingness to shift to a new position is worth anything at all."

It's not willingness. It's ABILITY. A team without a player who has the ABILITY to be a utility infielder would be pretty hamstrung if they had to carry three backups for 2B, 3B and SS, none of whom could play anywhere else, would they not? The team is the sum of its parts, and ya know, sometimes WAR/TR/EqA does not comprehensively catch how to add to the sum.

Pete Rose moved to 3B for Cinci in 1975. He was uniquely qualified to do so. This got some guy named Foster in the lineup. Sorta helped the team win 108 games. That's worth nothing?? Any non-misguided analyst should be able to see this :)

What does Win Shares uniquely bring to the table? What if a team constantly under-(over-)performed their RC projections, or their W-L pythags? WS apportions credit for this. In Boggs' case, as others have said, he seemingly contributed to the team at times not equating to the sum of their parts, as a slow OBP guy stood on first abse in front of GIDP RH suggers. Jim Rice gets all the credit in bizarro-fan-world for his RBI. In bizarro-saber-world, he merely gets all the blame for the GIDP. In this case, I happen to think saber-world is closer to the truth, but the split ain't 100-to-0.

I still don't agree, Tom. Virtually every major league shortstop could be at least an average fielding 2B if they had to. How do you know that Boggs couldnt've shifted somewhere else if he had been asked to?? Yes, positional flexibility has a value in that it enables the team to effectively expand the roster (carry an extra reliever or whatever), but a) it's pretty small and b) it's not fair to just assume that guys who weren't asked to do it wouldn't have been able to.

Did Boggs's teams consistently have worse records than their OPS+/ERA+ would indicate, especially *after* deducting for the 45 runs--not more and not less--that we know Boggs cost them on the basepaths? If that's the case, it's news to me.

Dan R, the data I'm working with don't actually include SFRAC: just WARP1, WARP2, pennants added, rate, and salary. A separate file contains SDev for each league for each season and rep level for each position for each season. So SDev is broken out by league, but rep level isn't. AFAICT, there is no DH adj. to rep level in the iteration that I am using.

If I have time, I'll download the newer numbers and look at the third basemen in light of those, and see if it changes anything.

No, the rep level chart doesn't include the DH adjustment. The rep level chart is for the NL post-1973, there's just an 0.6 wins/162 correction added in to the actual calculations.

Moving from OF (or 1B) to 3B sucessfully is NOT common. Unsuccessfully, yeah. When a team has a 'hole', along with an extra good player on the bench, and a man fills the hole by moving to a more (or at least as) demanding a position (not just from SS to 2B), and the team is helped, I give the hole filler some extra credit. YMMV.

Well, that's not relevant in this case, since Molitor was moving to a less demanding position (3B) from a tougher one (2B).

so, you're saying you'd agree with my point if it WERE the case he was moving to a tougher one?

I wouldn't give him credit for the "flexibility" per se, no. But I would consider giving him credit for being "wasted" by his manager at a less demanding position for the years that he actually had the ability to play a more challenging one. Along similar lines, I could see giving Molitor a little extra credit for his first few years at 3B, because presumably he still would have been able to play a decent 2B if his manager had put him there.

Right about now I need Harvey's Wallbangers to give me an early 80s Brewer history short course.

Best as I can recall/put back together, in 1980/81, Jim Gantner came up, and he could really pick it at 2B. Molitor put in some time at SS in those years, but I guess was not gonna cut it as a fulltime SS. And they had this Yount guy anyways. So by 81, Molitor is mostly a CFer. Gorman Thomas is a huge bat and can also play CF, but he plays RF when Paulie is in center. They don't have a 3Bman in 81 - Money (late career, on decline?) and Howell are not cutting it. So in 82, Molitor goes to 3B from CF. This allows Money to DH, Thomas to CF, Charlie Moore of the rifle arm in RF. Success, all is well.

Could Molitor still have played 2B in 1981ff? (Could he have played SS for a MLB team that did not have Yount? I dunno!) Stats seem to say he was good at 2B in 80/81. Of course, maybe he was moved off for injury protection; would not be the first time THAT happened to a good-hitting 2Bman, and since he lasted long enough to collect a bajillion hits, ya can't argue it.

I perceive his flexibility helped the team, and that he may have been slightly underused. Not sayin everyone has to agree.

Just a quick reminder: Molitor didn't quite go directly from 2B to 3B. There's some OF in there.

1978-1979: primarily 2B, with occasional games at SS (but with Robin Yount developing, there was no need to have him play SS)
1980: hurt some. Played 2B when he played, but Gantner gets in a full year split between 2B and 3B.
1981: mostly hurt. Mostly played CF when he did play, with Don Money returning to 3B. Gantner established at 2B.
1982: 3B; completely healthy. Scores 136 runs.
1983: 3B. Healthy, mostly.
1984: Hurt. Misses nearly entire season.
1985: 3B. Plays 140 games.
1986: 3B. Misses a third of the season.
1987: Plays 118 games, split nearly evenly between DH and 3B/2B (yes, there were some games back at 2B). Scores 114 runs in 118 games.
1988: Healthy. 2/3 3B, 1/3 DH. Scores 115 runs.
1989: Healthy. 80% 3B, 20% DH. Towards the end of the year gets moved from leadoff to #3. Doesn't score as many runs.
1990: Misses a third of the season. Plays more 2B than anything else. (Gantner also hurt?)
1991: Becomes DH/1B. Remains primarily that for the rest of his career.

In the early 80's. Molitor would probably have been a better CF than Gorman Thomas. Whether Molitor in CF, Thomas in RF, and Money/etc. at 3B would have been a better alignment for the team ... eh, probably not.

Harveys has posted on this. I believe his line is that Molitor was good at both 2B and 3B, but kept getting hurt when he played the field.

Prelim:

Let's see if I'm any closer to consensus than I was on the shortstops...

1. Schmidt
2. Mathews (not far behind)

3. Baker
4. Brett
5. Boggs (these three are packed pretty tight, not certain about the order, though clearly Baker's peak speaks to me more than to most)

6. J. Wilson
7. Santo
8. Allen
9. Groh
10. E. Sutton
11. J. Collins
12. Beckwith
13. B. Robinson
14. Molitor
15. Hack
16. Da. Evans (this whole group is reasonably tight, though I can probably find some separation before my final ballot. Probably movement to come here before then)

17. Nettles (not in my pHOM)

18. Boyer (really not in my pHOM)

Presumably Molitor gets credit for the games he played at 2B and he gets credit for the games he played at 3B. And he gets credit for the games he played elsewhere. Prtesumably he doesn't get any EXTRA credit for playing anywhere or everywhere or nowhere. Next question.

sunnyday, that's how I vote too. But TomH is arguing for extra credit for Molitor on the grounds of "flexibility."

Exactamente.

glad we all agree on what we mean at least :)

If I have time, I'll download the newer numbers and look at the third basemen in light of those, and see if it changes anything.

I've now done this, and it appears that the old numbers I was using from Dan R's WAR were already DH-adjusted, so I was adjusting twice. Removing that adjustment will lower the totals for all the post-1972 AL players a bit. That'll shuffle the rankings some, too. Molitor and Nettles will be down a spot or two; Brett and Boggs may swap spots. Boggs' value numbers are a bit better, but, really, if I were building a team that isn't the Fenway Red Sox, I'd take Brett over Boggs every time.

Sutton's raw career OPS+ is helped by the fact that his late peak came in the longest seasons of his career.

I don't believe this is true - his OPS+ from 1871-77 was 122, from 1878 on was 117. It all depends on where you draw the cutoffs.

If I get a chance I'll do a weighted average of his OPS+ based on the PA in the B-R adjusted stats. Don't have time now though.

I just don't see Beckwith and Wilson as Dick Allen hitters. As was said, Allen was the best hitter on the planet for a few years. They were not. It's not an 'aw, hell with them' thing at all. I've been one of the biggest proponents of not doing that. You are barking up the wrong tree there.

I was not a big supporter of either player when they were on the ballot.

I don't believe this is true - his OPS+ from 1871-77 was 122, from 1878 on was 117. It all depends on where you draw the cutoffs.

If I get a chance I'll do a weighted average of his OPS+ based on the PA in the B-R adjusted stats. Don't have time now though.

Well, maybe there was an error in my calculations or my data entry, but I got the 110 number from a weighted average. You won't find the effect by splitting his career in half, because he has a big trough in the middle. It's the 1883-85 peak in 98-112 game seasons that distorts, I think (plus 1875 was a big year in the NA and also its longest season). But by all means check my math.

prelim

1 schmidt

2 mathews
3 brett - I WANT brett to be #2. It's close, but my head wins over my heart.

another tight grouping:
4 boggs
5 molitor
6 santo
7 baker

8 brooks
9 sutton
10 collins
a gap after the above top 10

11 groh
12 hack
13 wilson- it's been so long since these threads, it's tough to recall all I learned about these guys, and at times the spreadsheet and notes I have do not serve as the world's best memory tools. I have Wilson lower than many; I'll look again.

14 beckwith - as with Allen below, the name of the game is to help your team win. 99% of the time, individual stats capture accurately a player's contribution.
15 boyer
16 evans
mcgraw oughta go about here :)
17 nettles
18 allen - I still support Allen for the HoM, as I do the other 17 men above him.

Amazing to me how many dissing comments I see about Boyer. It's not like I feel I have to defend someone I rank 3 spots from the bottom, but here's what I see:

96 WARP3 over 11 seasons. Five seasons of 10 WARP3 or more.

Played in one of the toughest competition leagues in history.

Was Nettles/Robinson with the glove in his prime. Except for the year he went out and played like Willie Mays in CF.

Had a career year in 64; EXACTLY what the doctor ordered to help the Cards win a pennant by one game that their pythag says they didn't deserve to win. Who drove in most of those clutch runs? Hmm... Then he drove in 6 runs and scored 5 in the Series that his team barely prevailed in.

Finished in top 20 of MVP voting many years running, often higher.

Sure, he wasn't AS good as some, or played as long as others. But Boyer settles in nicely in my barely-worthy HoMer catgeory.

I just ran the weighted average on Ezra Sutton and it turns out it's 117.46.

This uses the B-R normalized AB+BB, with OPS+.

I would also say that the big peak in the mid 1880s came as the game was vastly improved over the 1870s version. I have no reason not to think it was a very legitimate second peak.

A 117.5 OPS+ over a 10990 PA career is immensely valuable, especially at key defensive positions.

Santo had a 125 at a less important defensive position (3B in the 1960s is nothing like 3B in the 1870s and 1880s) in a career that's only 9350 normalized PA, or 3 seasons shorter.

It's not like Sutton didn't have a peak either, his best years include OPS+ of 165, 162, 149, 148, 144, 135, with 6 other seasons between 99 and 115.

Santo's best years include 164, 161, 153, 146, 139, 131 and he gains some ground with 3 years in the 120s.

But Sutton had a comparable offensive peak to Santo, was more valuable on defense and had more in-season durability (2643 neutralized games, vs. 2254, despite the fact that he only played two more season as a regular).

I had Sutton 5th and Santo 6th. I could possibly see flipping them, if it can be shown that Santo was more valuable defensively. But I can't see going much slower. Their careers have different shapes, but the end results were very similar.

As someone who runs a Hardball Dynasty (What if Sports) team, you get to see some intricacies that you don't get running a Diamond Mind type team, for example.

You don't have ANY future knowledge. There is a fictitious universe of players and you run the entire franchise from Rookie League to the Majors. Long-term contracts, waivers, options, Rule 5 draft, they really cover most of the bases. You can't get 162 games out of 150 game players, like you can in most replay leagues either. Almost every player needs rest from time to time.

I'm 7 seasons in now, and I mention this in light of the Molitor 'flexibility' question that's been raised.

I can say, unequivocally that having a player around who can play multiple positions (think Gil McDougald or Paul Molitor) is very, very valuable. I purposely have held onto a guy that is really only a league average or slightly below hitter, and paid him $7 million this year, because he can basically play anywhere but SS, and could even fill in there in a pinch if really necessary.

So there is definitely something to the ability to be moved around has value concept, IMO.

Of course the player (who had a 99 health rating, which controls injuries but not durability) was injured in the 7th game of the season and lost for the season.

While this stunk for my team, I was able to see just how valuable he was, losing him, despite having a decent bench was a gaping hole. The flexibility I had with platoons was now limited.

It got to the point where I was so used to being able to drop this guy in wherever I needed him, that I had to trade for another 'overpaid' $6 million dollar player I had just recently traded away (because of the injured guy's flexibility) who was an even worse hitter, just to get some of the flexibility back.

Sorry for the long-winded response, but in a somewhat real world type situation, I can definitely vouch for the value of being able to play multiple positions depending on the team's needs.

BTW, 99 is a 'great' not 'awful' durability rating, I really got screwed there!

TomH, that's a *heck* of a ballot. I'm glad you posted it as prelim, because I'd like to offer a few challenges.

1. Dick Allen last??? Come on. What is your concrete evidence that Allen's statistics did not produce as many pennants for his teams as we would expect? This isn't the Miss America contest, it's the Hall of Merit. I have a hard time believing that vote is constitutional. I could perhaps see his bad reputation being used as a tiebreaker, but LAST?

2. Boyer and Hack ahead of Darrell Evans?? Is that a typo? They're two of the absolute worst players in the HoM! Just my basic charts of the three here, standard deviation-adjusted...

(Note: My war adjustment is to add 10 to the player's PF and subtract 2 runs/162 from his fielding for 1943, and to add 20 to the player's PF and subtract 4 runs/162 from his fielding for 1944 and 45).

Hack



3-year peak: 19.2
7-year prime: 39.4
Career: 57.1


Boyer



3-year peak: 20.5
7-year prime: 41.7
Career: 53.1


Evans



3-year peak: 21.7
7-year prime: 41.9
Career: 66.1

Like, this just seems open-and-shut to me. Evans has the highest peak, the highest prime (by a nudge), and by *far* the most career--so much more career that it's hard to see how he's in the same breath as the other two guys. Where are you coming from with this?? The only things I can imagine are

a. that you just don't buy the defensive numbers on Evans. OK, I think that was, if you'll excuse the pun, a defensible position when we were just talking about FWS and FRAA. But TotalZone has Evans at +52 for his career. DRA has him at +98. These are not dubious statistics--they're based on play-by-play Retrosheet data, and they have a .8 correlation to today's PBP metrics. There are no two ways about it--the guy could pick it.

b. that you write off Evans's peak for having occurred in the launching pad. Again, like I said about Boggs, that won real games for the Braves that counted in the standings. The teams' opponents didn't take advantage of the HR-friendly park the way that they did.

Other than that, I'm baffled. Would you care to clarify? "MVP votes" and "clutch runs" seem like "arguments" I'd expect to hear from Murray Chass or Joe Morgan, not a respected HoM voter.

I've got to say I tend to agree with Dan's comments, especially on Allen.

Where are the teammates (other than Frank Thomas) who say he was such a prick? Most of his teammates say they'd do anything for him, right?

Was he a distraction the same way Barry Bonds was, in that the media hated him, but his teammates didn't care?

Dan, have the WARP numbers you created changed since the version I used 10/15/2007 to create my database? When was the latest version created? Do you have post 2005 numbers yet? Pre-1893? Thanks for any guidance.

No, no update since then. I do have a bunch of new data (SFR, TZ, DRA, and EqBRR) which will dramatically improve the accuracy of the FWAA and BRWAA scores for the post-1956 period (and, by extension, the accuracy of the estimation equations for the pre-Retrosheet era), but I haven't yet incorporated it into my system. As soon as I get the time to do so, I'll post.

Can someone with more authority than me in the HoM cosmos tell me whether TomH's #18 for Allen is constitutional? Doesn't he have to at least make a minimal effort to provide some, any sort of evidence or backing or support for a vote that simply cannot be justified on quantitative Merit grounds?

An un-Constitutional finding is pretty hard to stick. This isn't one of those times.

And it's not like we've ever actually *required* ballot commentary. Oh yeah, there was one time....

Well, I would definitely say things that get you a one-year boycott should not apply at all in the position rankings. I want that to be clear.

As an example. You can dock Joe Jackson for all of 1919 if you think he tanked the World Series. But you can't rank him last if he otherwise wouldn't be after removing 1919, based on 'moral' grounds.

If Tom thinks Allen just wasn't that good compared to the rest of the list, that fine. He had a short career, well not necessarily short, but not much durability in season.

But things like personality, unless you can demonstrate clear evidence they cost the team games, should not be considered.

Does that help, in terms of clarification?

Well, the only comment TomH made was actually on Beckwith: "as with Allen below, the name of the game is to help your team win. 99% of the time, individual stats capture accurately a player's contribution." It seems to me TomH has established a pretty high standard of proof there--he must demonstrate that Dick Allen is in that, by his own estimation, 1% of the time when individual stats DON'T capture accurately a player's contribution. That eviddence has not yet been provided--in this thread or in any other--and I do not think his vote should be accepted until it is.

I am not sure, but didn't TomH once argue for the negative effects of Allen on his team by looking at how his teams performed before and after they traded him? And, for that matter, the fact that he was traded so often? I don't think that is a good argument, but doesn't it at least meet the minimal requirements for constitutionality?

Of course, if that argument is valid, then A LOT more than 1% of players are not accurately measured by their individual stats...

Yeah, I want to be clear that I'm not saying Tom's ballot is unconstitutional.

But it's definitely not out of line for DanR to ask for clarification/explanation, etc. either. That's part of the reason why we have people post comments with the ballot to begin with.

Meanwhile TomH sees George Brett as being closer to Matthews than to Boggs. Given that I've yet to see a statistical argument for Brett over Boggs (one that incorporates the knowledge that Brett played 900 games at 1B and DH, and thus played fewer games at 3B than Boggs), I'm still unconcvinced Brett should be above Boggs, let alone closer to Matthews.

Brett mostly played 3B from 1973 (cup of coffee) through 1986 (his age 33 season). Here are his games played by position for those years (according to baseball-reference.com):

3B 1677
1B 23
LF 18
DH 13
SS 10
RF 7

From 1987 through 1993, here are his games played by position.
DH 493
1B 438
3B 15 (with 11 of those games coming in 1987, only 4 more after that)
RF 7
LF 4
SS 1

Here are his career totals for games played by position:

3B 1692
DH 506
1B 461
LF 22
RF 14
SS 11

For approximately 950 games of a 2700 game career, Brett played 1B and DH. Two of his five best seasons (by OPS+) occurred during that time. As a 3B, Brett played 130+ games 7 times in 13 seasons. As a non-3B, Brett played 130+ games 5 times in 7 seasons. One can't simply look at batting runs or simply look at OPS+ (which undervalues Boggs's tremendous OBP+ advantage in the first place) because these things don't adjust for the fact that Brett's not being a 3B for the last third of his career.

Total batting runs (from bb-ref) for Boggs's 3B years ('83-'99): 461.0
Total for Brett's 3B years, not including cup of coffee ('74-'86): 394.5

If we're ranking third basemen, why does Brett get credit for his time at 1B and DH?

There seems to be a "But who I would want at bat with runners on and the game on the line?" reasoning that leads to Brett, but the problems with that question are (1) it assumes that Brett would be healthy, and he very often wasn't -- we get to put Brett up there at full health when he frequently wasn't; and (2) the question itself tends to favor a slugger, which isn't fair to OBP which is more valuable in the first place. As far as value goes, it also ignores positional adjustments -- we might want Joe Dimaggio (OPS+ 155) up before Mike Schmidt (OPS+ 147), but that doesn't mean JD was a better player than MS (and again, Boggs was a 3B for 97% of his career, while Brett was only one for 63% of his).

Boggs was about as good a hitter as Brett (slightly worse by OPS+, but that undervalues his OBP -- he might well be better overall). He played more games at 3B and (likely) played better defense there as well. His peak is comparable (frankly, it's better -- Brett's advantage is in the length of his career, ignoring position). We can denigrate some of this with "Boggs's baserunning" and other arguments, but several of these arguments seem designed to fit the previously-determined conclusion ("I never thought of Boggs being the best player in baseball, but I often thought of Brett that way"), while ignoring what they actually did on the field. If we put a value to Boggs's performance (and make sure to look at Brett's as well -- what was his baserunning like? has anyone ever looked at it?) and then compare, does Boggs still finish behind? So far, I'm unconvinced.

So, what's the statistical argument for Brett's ranking above Boggs on the list of best third basemen of all time?

See my post 10 Robert. Boggs 1.29, Brett 1.28.

Pretty close, I don't see how either one of the other is really an issue. They are basically even.

I lean towards Brett, because I think in a neutral environment, he'd be better. Sure Boggs took advantage of the park he was given, but to me, that's kind of a tie-breaker the other way, when evaluating the immortals.

If we're ranking third basemen, why does Brett get credit for his time at 1B and DH?

Because players get ranked based on their entire careers. They get on the list for the position where they had the most value, but they most definitely get ranked based on their entire career value.

Any ballots that don't take this into account are definitely not acceptable. I thought this was clear, but if it wasn't, it should be now.

If we're ranking third basemen, why does Brett get credit for his time at 1B and DH?

You're not really ranking third basemen, you're ranking the overall performance of a set of players that happen to be generally considered as thirdbasemen rather than belonging to any other fielding position, as I understand it.

I have Boggs and Brett about as close as it gets. Brett was indeed a better hitter (52 BWAA2 in 13.9 above-replacement seasons, against 46 BWAA2 in 15.5 above-replacement seasons), and a substantially better baserunner (about 7 wins' difference there), so that's 13 wins to Brett on offensive value. But Boggs makes them up by being a substantiaally better fielder (by about 6-7 wins), by playing all rather than merely most of his career at 3B, and by retaining above-replacement MLB value for longer (a replacement player in Boggs's playing time and positional breakdown would have been 30 standard deviation-adjusted wins below average, versus just 24 for Brett's). All told, it comes out to 82-83 WARP2 for both of them. Brett's two best seasons are better than Boggs's best two, but Boggs has a strong advantage in the next batch of seasons, so peak/prime considerations are a wash as well. You really can't go wrong either way. The only thing I would say that might tip it meaningfully to Boggs is if you give him minor league credit for not reaching the majors until halfway through his age-24 season, but his minor league numbers in 1980 weren't nearly inspiring enough to justify that, IMO.

This has been a very valuable discussion. I've decided to give Home Run Baker some credit for his missing years and he jumps up. Sutton as a 117 OPS+ is pretty good but his glove isn't and I am struggling to believe he's really a 10000 plate appearance infielder when almost nobody puts up 10000 plate appearances in the infield. He'd be 5th highest among eligibles after Molitor, Robinson, Brett and Boggs.

Here's a revised prelim

1) Schmidt
2) Mathews
3) Brett - he's basically tied with Boggs. The answer to the question "Would you like George Brett or Wade Boggs on your team?" is "Absolutely!"
4) Boggs
5) Jud Wilson
6) Frank Baker (2 bonus years credit)
7) Paul Molitor - the "undurable" player has the most plate appearances
8) Ron Santo - 7&8;are a virtual tie also. Luckily the spreadsheet prefers the St. Paul native to the Cub
9) Darrell Evans - Evans to Hack are barely separated from each other
10) Heinie Groh
11) John Beckwith
12) Ezra Sutton
13) Brooks Robinson
14) Stan Hack
15) Dick Allen - short career, bad glove, lack of time actually spent at 3B hurt him
16) Jimmy Collins
17) Graig Nettles
(Tommy Leach)
(John McGraw)
18) Ken Boyer - and he's not a mistake, he's just last. Defines in/out line.

The HoF has whiffed on half of our list and only recently got Wilson inducted.

The only thing I would say that might tip it meaningfully to Boggs is if you give him minor league credit for not reaching the majors until halfway through his age-24 season, but his minor league numbers in 1980 weren't nearly inspiring enough to justify that, IMO.

Why wouldn't one look at post-season performance to attempt to break the tie before giving (or not giving) minor league credit?

re: Brett/Bogs, some of it is baserunning, some of it is Fenway-only value, some of it is post-seaosn perfromacne, which many people where seem to virtualy ignore. Posts 87-89 I agree mostly with.

re: Dr Evans, versus Hack/Boyer. I will relook at Evans, to see if I had incoporated the more modern def raitngs when we were discussing him. I will say if you use win shares, Hack has Evans beat on peak and conesecutive prime. And it ain't close. Now, if you say "what does Win Shares tewl lyou that system X doesnt'", I might reply "show me where Win Sahres is wrong first"; IOW, it should ot be my responsibility to prove MY ballot to your satisfaction. And "win shares disgarees with system X is not a winning argument in and of itself. Not that I use WS exclusively at all, but it's a tool.

re: Allen, should I re-post everything I said 'back in the day'? I'm not upset at strong questioning, and I heartily agree with our tradition of needing comments to make a legit ballot, but sometimes in the rush of "real life", it really is tough to take the time to go back and re-hash those areas over which many of us have already tread (as post 84 alludes to). I mean, if I need to, I will - Dan, your comment "That evidence has not yet been provided--in this thread or in any other" is either premature, incorrect, or simply highly opinionated, I don't know which.

And the line ' "MVP votes" and "clutch runs" seem like "arguments" I'd expect to hear from Murray Chass or Joe Morgan, not a respected HoM voter. ' - hey, I know where you're coming from, really. With most of my runs, it seems that I spend half of my time convincing them that they can't base a defensible opinion on Pat Burrell and Derek Jeter on these silly measures. But what has happened in some stat-heavy places is that many have pretended that Mazeroski's home run is the same as every other home run, that Yaz's September of 67 was a lucky streak that has no bearing on his overall value, etc., that a player's runs scored are 100.000% a function of his walks/hits/teammates. Excuse me if I use some Morgan/Chass-like arguments little bit, and if because I call them underrated here that it appears at times that I mention them more than a little bit. I'm not saying anyone else has to use them. But somehow in this project and others, people have put a much higher burden on the allowable use of said measures than anything else. I mean, Joe, "unless you can demonstrate clear evidence they cost the team games"; what will the jury accept as clear evidence? Team performance I tried once, and some shot it down with a barrage of "there were other reasons, so that's invalid". Well... I guess that means it's pretty tough to provide clear evidence, now, doesn't it? But aside from that, why should non-individual statistics be somehow held to a different standard? Baseball is not bowling or golf or tennis, guys. It's a team game. I work in an office, and some of the best measure of an indivudal's worth is ... team project success. Some people don't work well on teams. Some people leave teams if they have to work with some individuals. That's how life is. Thankfully, understanding of the individual parts that make up MLB can often bring us to a close understanding of why teams win and lose. To pretend we have it down to 100% is silly. And those who pretend it is are acting silly.

I'll try to resurrect and summarize Dick Allen sometime this weekend, but I can't promise anything. Life is busy :) But, Allen's career has been the subject of a million words across baseball fandom for decades. The 'evidence' is not hiding under some rock.

Because players get ranked based on their entire careers. They get on the list for the position where they had the most value, but they most definitely get ranked based on their entire career value.

You're not really ranking third basemen, you're ranking the overall performance of a set of players that happen to be generally considered as thirdbasemen rather than belonging to any other fielding position, as I understand it.

Ah, thanks for the clarification.

Any ballots that don't take this into account are definitely not acceptable. I thought this was clear, but if it wasn't, it should be now.

*nods*

Somebody just has to explain the Hack thing to me. He is popping up way higher than I would have thought on the ballots of a lot of intelligent voters. Here's Hack vs. Allen, after stdev adjustments:

Hack (same chart as above)



3-year peak: 19.2
7-year prime: 39.4
Career: 57.1


Allen



3-year peak: 24.7
7-year prime: 47.1
Career: 60.9

OK, so career value between Hack and Allen is fairly close--if you use a smaller war deduction than I do, you could probably get the two tied on that basis. But after that, it's allll Allen--his peak/prime are downright massive. Hack's best season would be Allen's fourth-best. I could understand if Hack were, you know, the Jimmy Collins of the 1930's, but if anything, I think my numbers are being generous to his defense--DRA has him at -55. Yes, 3B was a different position when Hack played it, and Allen's only half a 3B--you can see that easily in the Replc column--but we're talking about the difference between a war-inflated 119 OPS+ and a 156.

DL from MN, you don't say you're docking Allen for any "extracurricular activities," you just think Hack was better. What do you see there that I don't?

I lean towards Brett, because I think in a neutral environment, he'd be better. Sure Boggs took advantage of the park he was given, but to me, that's kind of a tie-breaker the other way, when evaluating the immortals.

I hear ya on that, but I'm not sure I agree with it. It seems to me that Boggs didn't have the chance to adjust his game to other home parks(*). He adjusted his game to the one he had. Why assume that he wouldn't have done so elsewhere?

(*) Of course, he did have the chance when he was 35 and when he was 40 to do just that. Boggs went from being fairly neutral at home to considerably better at home within a couple of years of joining the Yankees, then went back to being a little better at home. With the Devil Rays, he was better on the road his first year, then better at home the next year. I don't know that an especially strong argument can be made, but I think one could argue that Boggs tended to adjust to his home park.

You're shaving 9 WARP off Hack's war years, I'm not.

It seems to me that Boggs didn't have the chance to adjust his game to other home parks(*). He adjusted his game to the one he had. Why assume that he wouldn't have done so elsewhere?

I don't think players 'adjust' so much, so that was bad wording. I should have said I felt that Boggs skills suited his park perfectly. So he's going to look better there than he would anywhere else. Same for Ozzie Smith, and Mel Ott.

Rizzuto would be the exact opposite. He played in a park that just killed RHB (and not just on HR, the 1B and 2B factors for the 1940s Yankee Stadium in Diamond Mind, for example, are just killers), and a park that you absolutely have to have power in, and was a speedy guy (71% SB, great for his era) in an era where SB didn't matter.

But him in 1985 Busch Stadium and he's a helluva lot more valuable than in 1950 Yankee Stadium. Not that he wasn't still a star, but he couldn't have played in a worse environment for his skill set.

I'm not saying they are huge impacts, but it's definitely a tie breaker for me, to go with the guy whose skills were less suited to his environment, or the guys whose skills translate to other environments better, all other things equal.