hot topics

Edd Roush vs. Max Carey:

Any thoughts?

Roush was the better player at his peak, but Carey had the longer career. I like Roush a little better, but I'm not sure if he'll make my ballot.

Edd Roush vs. Max Carey:

Interesting one.... deserves some discussion. Roush certainly could hit quite a bit better, even with the SB/SB%'s of Carey factored in.

Roush has a shorter career in a few ways. Fewer seasons, Federal League seasons, and in-season durability issues.

As a peak voter, I'm a bit torn. Still thinking about it. Career voters might have an easier time swaying towards Carey, but I'll let them chime in.

To me, Roush performed at a very similar level to Van Haltren (conistent .310ish EQA's for many years, league-average defense in CF, WS totals well into the 300's). The problem is that VH put up his career in the very tough one-league environment of the 1890s, while Roush was a second-tier star in the second best major league in a world where there were more major league teams and whose longevity didn't stick out among his peers. VH is going to be in the top 5 on my next ballot; I currently have Roush 10th or 11th, somewhere just in front of the similarly talented but less consistent Jimmy Ryan.

Several people have said their systems love Schang. I thought mine would too, but find him coming in 16th or 17th. My big problem is that I think he was playing a less valuable 19th century catcher's role at a time when some other catchers had adapted to a more durable 20th century model.

I haven't yet gotten a handle on Redding. He seems to be a ballot contender, but I'm having a hard time convincing myself that he was a bigger achiever in black baseball than guys like Monroe, Poles, and Mendez, all of whom are sitting just a few spots off my ballot.

Several people have said their systems love Schang. I thought mine would too, but find him coming in 16th or 17th.

I'm actually having the same problem. I was surprised by this turn of events myself.

I'm going to go over him again this week. Since we have White, Bennett, Santop, McVey and Ewing, we're doing OK with catchers anyway.

Chino Smith - will not make my ballot, as he becomes part of the Jennings/Chance/Dobie Moore group.

Edd Roush - Will make the bottom part of my ballot, ahead of Duffy, behind Van Haltren, Carey and Ryan.

Dick Redding - He will make my ballot, I'm just not sure where. Based on his All-Star and GSA selections (both James and Holway) he might deserve to be #1. However, his stats don't seem to support this ranking. It seems from Chris Cobb's data that Redding pitched well in the teens and early 20's, but hung around later (kind of like Dolf Luque).

Wally Schang - My system favors catchers, but only Bresnahan, Schalk and Santop have made my ballot since I refined everything. Santop got elected right away, while Bresnahan and Schalk now linger just outside. Schang had the career to fit well in my system. With subjective catcher bonus points, he will rate #4 or #5, depending on where I place Redding. (Heilmann, Torriente and Van Haltren will beat him on my ballot).

I am all but certain that Heilmann and Torriente will be #'s 1 and 2 on my ballot this year. However, I am still deciding where to put each of them. It is very likely that both will go in, but jsut in case someone like Redding or Groh beat them out, I want to make sure I get it right in deciding who gets #1 and who gets #2.

So my question is...

Just how good was Torreinte's defense since we have no numbers to really go by? I understand him to be a gold glove CFer, but is he a gold glover the calibur of Bernie Williams, Ken Griffey Jr., and Torii Hunter (good but no where near historic) or one more like Darrin Erstad, Mike Cameron and Andruw Jones (virtuosos in their prime)? If it is the latter, he may very well get #1, if it is the former, #2 is his place.

I know that Charleston had to move to accomodate him, but that doesnt' say too much to me. This is in the dinosaur age for statistics, and in a league where there were even fewer. Could this be like ARod playing third so Jeter could play short? I mean they had even less to go by then. If there weren't stats Jeter would be condisered a decent glove man and Hunter would be considered an all-time great.

So where do people think Torriente falls?

Newbies:

Wally Schang, C - Hit like an all-star outfielder while playing Catcher, so gave his teams a tremendous advantage. Negatives are only FAIR defense back when Catcher defense was still extremely important and relatively fewer plate appearances vs. players at less demanding defensive positions. Will definitely make Top 10 on Ballot.

Dick Redding, P - Have him around top 6 All-time Negro League pitchers, so he'll be on the ballot somewhere.

Chino Smith, RF/OF - Tremendous player, but very short career. Similar to McGraw, but looks like Lip Pike comp right now among OF'ers, which puts him just off my ballot.

Edd Roush, CF - He's right there in the Duffy, Van Haltren, and Jimmy Ryan mix, and with only average defense, he probably won't be able to break out of that pack and onto the end of my ballot.

It seems from Chris Cobb's data that Redding pitched well in the teens and early 20's, but hung around later (kind of like Dolf Luque).

Not sure I understand the comment about Luque. After his rookie season at age 27, Luque was an above average starter until his age 40 season in 1931, and which point he switched to the bullpen the following year and was effective as the Giants "closer" for 3 MORE seasons at ages 41-43. That hardly seems to describe a pitcher that just 'hung around later.'

Using WARP numbers to do these calculations, I have Bresnahan as 12 runs per season better than Schang with the bat and with a stronger peak, but:

(1) Schang about 10 runs per season better with the glove when both were catching.

(2) Schang with about 18% more playing time.

(3) Schang playing 78% of his games at catcher vs. 67% for Bresnahan.

Based on those calculations, I think Schang is definitely ahead of Bresnhan, but that it is hard to justify a gap of more than 3 or 4 places between the two. Right now, Schang has squeaked onto my ballot in 14th place; Bresnhan is 16th.

Prelim Ballot. The - represent significant gaps. The players within each group could easily be rearranged.

1. Cristobal Torriente
-
2. Harry Heilmann
-
3. Max Carey
-
4. Lave Cross
5. Harry Hooper
6. Heinie Groh
-
7. Bobby Veach
8. Fielder Jones
9. Stan Coveleski
10. Ben Taylor
11. Jake Beckley
12. Cupid Childs
-
13. Wally Schang
14. Dave Bancroft
15. Rube Waddell
16. Mike Tiernan
17. George VanHaltren
18. Tommy Leach
-
19. George J. Burns
20. Jimmy Ryan
21. Del Pratt
22. Clark Griffith
23. Hughie Jennings
24. Billy Nash
25. Vic Willis
26. John McGraw
27. Joe Tinker
28. Ray Schalk
29. Ed Konetchy
30. Pete Browning

As a peak voter, I'm a bit torn. Still thinking about it. Career voters might have an easier time swaying towards Carey, but I'll let them chime in.

Similar question. When evaluating peakish candidates, how about Roush versus Sisler? Because of the league-difficulty question, this one's a little thorny, and Roush's advantage over Sisler might be even less pronounced than the slim 314/292 career WS advantage suggests.

The offensive numbers I play with involve a context-adjusted (using, in part, PythPat) number derived from RC above average. In the table that follows, the first number (in arbitrary units) comes from this RCAA, the second number is an extra bonus for having big years, the third number is derived from RC above 75% of average, and the fourth number is some sort of meaningless composite of the first three.

Cochrane    33   17   51   110
Bresnahan   29   16   44    98
Schang      25    9   43    82
Bennett     10    4   24    39
Kling        5    2   19    27
Schalk      -4   (0)  14     8

This is for offense, period, not offense while playing catcher. Of course, quite a bit of Bresnahan's credit, including most of that "big years" bonus, comes from his years as an outfielder.

Some initial thoughts on Dick Redding

Redding’s record does not raise him above the pack of pitching eligibles, he’s somewhere in the group of well-qualified pitchers who are not no-question HoMers.

Within the group of eligible pitchers from 1910-1930, I place him fourth, behind Coveleski, Mendez, and Shocker, but ahead of Cooper, Mays, Cicotte, and Adams.

His career shape is most comparable to Cooper’s, but he’s just a bit better than Cooper in terms of both career and peak, as I see it.

On peak:

He was as good as anybody for about three seasons, 1915-1917 and very good from 1915-1923 (with a couple of strong years prior to 1915), which puts him ahead of Cooper, et al. in the peak category. He doesn’t have the lights-out peak of Mendez or the sustained excellence of Coveleski or, to a lesser extent, Shocker, so he doesn’t catch them in my system, but more career-oriented voters might prefer Redding to some or all of these three, though I can’t see even a career voter putting him ahead of Coveleski.

On career:

I9s projects him at 3556 ip, with time lost for WWI, so he’d be at 3800-3900 IP without the war, which is better than any pitcher in this group, though not quite as good as Faber, Grimes, and Rixey.

Working with his record and his team records, I project him to a support-neutral ERA+ of 111.6 (that would be a DERA of 4.03, for those wanting to do WARP comparisons) with a support-neutral W-L record of 226-181, + wartime credit. That ERA+ a good bit lower than most of the top major-league eligibles we’re looking at, but most of them receive some benefit to their ERA+ from defense, so I think using DERA will produce more accurate comparisons than ERA+

My best guess on placing Redding is that, in WARP1 terms, he is Wilbur Cooper +10%.

My current career win-share estimate is about 270 win shares, + war credit.

I’ll be posting the full rationale for these estimates on the Redding thread later this week, when I’ve had time to write them up coherently.

Same set of numbers, same explanations as in #14 (no league strength adjustments):

Van Haltren  41   21   67   138
Duffy        38   25   62   136
Ryan         36   20   64   131
Roush        37   21   60   129
Thomas       37   23   55   127
Carey        31   13   58   109
F. Jones     31   13   54   104

That's Carey with the last couple of years of his career left off; he would look worse on this if you included them.

Harry Heilmann is about to be elected; Mike Tiernan dropped out of the voting last year. Both were historically poor RF's who were among the most feared hitters in the game for over a decade. It's clear that Heilmann was better than Tiernan, but I wanted to try and quantify how much better,

Again, using the Prospectus numbers as my guide, I offer the following two potential comparisons:

(1) Heilmann is Tiernan plus about 12-15 runs per year with the bat and plus four additional slightly better than average seasons.
or
(2) Tiernan is Heilmann minus his two monster years (1923 and 1927) and his two worst (basically league-average) years.

In the end, I can't use Heilmann as a reason to return Tiernan to my ballot, but the comparisons are close enough to make me think that Tiernan is moderately underrated by the electorate.

With Schang joining Bresnahan and Schalk on the ballot, I realize how badly I've been underrating catchers. While I've made adjustments, Bresnahan and Schang just miss my ballot. I'll take a closer look at both during the week.

Prelim ballot.

1. Torriente
2. C. Jones
3. Heilmann
4. Pike
5. Browning
6. Childs
7. Coveleski
8. Jennings
9. Mays
10. Williamson
11. Groh
12. Leach
13. Mendez
14. Duffy
15. Griffith

16-20 Roush, Monroe, Carey, Cooper, Bresnahan
21-25 Schang, F. Jones, Van Haltren, Poles, Willis
26-30 Doyle, Waddell, Sisler, Bond, Tiernan
31-35 McGraw, Welch, Griffin, Chance, Moore
36-40 Burns, Veach, Taylor, R. Thomas, Ryan

Also, I haven't looked at Dick Redding yet, so he may make my ballot also.

Re. Edd Roush, it seems that in addition to the CF glut that a comparison with his old teammate Heinie Groh would also be constructive.

Career WS (James, unadjusted, CF over 300 only)--Carey 351 Van Haltren 344 Ryan 316 Roush 314 Groh 271

3 yr non-consecutive peak WS (James, unadjusted, CF over 90 only + Carey)--Roush 96 Groh 95 Seymour 93 Duffy 90 Carey 84

5 yr consecutive peak WS (James, unadjusted, CF over 130 only)--Groh 147 Duffy 144 Roush 136 Carey 133 Thomas 133

WS/162 (James, CF over 27.5 only + Carey + Roush)--Kauff 33 Browning 20.8 Seymour 28.8 Thomas 28.6 Van Haltren 28.1 Duffy 27.5 Groh 26.2 Roush 25.9 Carey 23.0

My Peak adjWS measure = top 3 consecutive + top 5 non-consecutive (CF over 23 only + Carey)--Browning 286 Groh 242 Duffy 234 Roush 232 Seymour 230 Carey 217

Reputation Monitor = peak and career WS + TPT + Hall of Fame Monitor and Standards + OPS + defensive bonus, etc.--Browning 187 Roush 184 Carey 176 Duffy 170 Groh 168

All-Star and MVP points (CF over 30 only + Carey)--Groh 40.5 Duffy and Roush 38 Browning 37.5 Beaumont 31.5 Carey 23.5

I based my ratings to some degree on these last three measures of my own devise, and of course I am a peak and prime voter, and to summarize, among the CF glut + Groh:

1. Groh--great peak at a physically demanding position
2. Browning--great hitter
3. Carey--many of the above measures are heavily weighted to WS, which I prefer to WARP, but which discriminates against exactly guys with Carey's skill set, so I adjust accordingly
4. Roush--definite ballot contender
5. Duffy--definitely rises above his approx. contemporaries Van Haltren (6 on this list) and Ryan (7) from the perspective of a peak/prime voter, but has not been on my ballot since 1908. Could return, however, while Van and Ryan are top 25-30 now and unlikely to return.

Add in the subjective candidates and I get Pike above Groh and Poles between Duffy and Van Haltren. Lots 'o' CFers.

I am not so much of a peak voter, however, that I cannot distinguish the very short peak/short career guys like Seymour, Beaumont, Thomas and Kauff, who are not among my consideration set (top 50). As a peak voter I have to warn you that I will take Hack Wilson over Wally Berger in future years.

Oh, P.S. a first pass:

1. Jennings (3)
2. Heilmann (4)
3. Pike (5)
4. Torriente (6)
5. Bond (7)
6. Sisler (8)
7. Groh (9)
8. Redding (new)
9. C. Jones (10)
10. Waddell (12)
11. Childs (13)
12. Coveleski (15 in 1935, 16 in 1936)
13. Williamson (15)
14. Browning (last on ballot in 1933)
15. Carey (11)

Close: Bancroft, Mendez, Roush, Doyle, Duffy

Drops out: Bancroft because Carey falls behind Browning and Bancroft remains behind Carey. However, I cannot believe I was Bancroft's best friend at #14. I thought defense was a good thing!

Back in: Coveleski in part due to favorable comparison with Redding

In addition to the above head-to-heads, I have always had Williamson ahead of Browning and that pretty much dictates the last few slots. The head-to-heads I am most uncomfortable with are Redding-Coveleski-Waddell, of course. I'm somewhat comfortable with Mendez-Redding, but only somewhat.

I would really like to have all of my top 18 on my ballot.

~#32. Schang
~#40. Chino Smith
~#100. Edgar Wesley--it is good to speak and remember his name, but probably not to vote for him.

I would really like to have all of my top 18 on my ballot.

Me, too. There are players that I think are very reasonable HoM candidates who are not getting onto my ballot. I think it quite possible that some players who are not getting onto more than a handful of ballots now will have a chance at election in 30-50 years. We will see.

I would really like to have all of my top 18 on my ballot.

Me, too.

Me, three. :-)

148 36.2 1921 Bibb Falk-LF (1989)

I should look it up to be sure, but I'm pretty sure the baseball field at the University of Texas, Austin is partly named for him.

I should look it up to be sure, but I'm pretty sure the baseball field at the University of Texas, Austin is partly named for him.

Bibb Falk

Looks correct to me.

Ferrell went from the dentist's chair to the mound in Fenway park on June 4 to beat the hapless Red Sox. The final score was 10-2 in Cleveland's favor and it was the club's tenth win in a row. "Didn't throw more than three or four fast balls in the whole game and the three or four were mediocre," wrote Cobbledick. "They didn't have the zip that one looks for when Wes Ferrell cuts loose.

"One watched him through the early innings today and felt a bit sorry that a pitcher like Ferrell, gifted by nature with brilliant blistering speed should have to get by with half-speed curves and looping slow balls." The Indians scored four times in the first and six more in the ninth when Wes smashed a home run "deep into the centre field bleaches."

The next stop was New York where Wes beat the Yankees with a three-hitter on June 9. Times' writer John Kiernan, for his on-going quest to determine if it were Ferrell or Earnshaw as the leagues' top right-hander, interviewed Willie Kamm and Bibb Falk before the game. "It's hard to say," said Bibb, "Ferrell's had a sore arm for some time. Can't throw his fast ball and that's his big bet."

"I'd pick Ferrell," said Kamm.

"He's a lot younger," said Bibb. "He ought to pile up a great record before he gets through."

"That isn't the only reason I'd pick him," said Kamm. "You see, he's had a sore arm but he went in their as a relief pitcher just the same and he stopped rallies. Then he went out and beat the Red Sox, sore arm and all. That's what I claim is a good pitcher."

Preliminary 1937 ballot
1)Dick Redding—3rd best NeL pitcher ever is an easy choice for #1
2) Lip Pike—It gets harder and harder to feel confident about the cross-era comparisons as time moves on. For consistency’s sake, Lip stays on top.
3) StanCoveleski—Only 3 slots below Redding on my all-time (through this election) list.
4) Pete Browning
5) Rube Waddell
6) Cristobel Torriente
7) Ben Taylor—In last year’s ballot I wrote that he should be on more than 8 ballots. He wound up on 7. I find it funny that many voters compare him to Beckley, but nobody votes for Ben. 28 people voted for Jake. I rank Beckley very low, relatively, so I could understand if nobody voted for Taylor because they thought Beckley wasn’t that good either.
8) Hughie Jennings
9) Jose Mendez
10) Addie Joss
11) Harry Heilmann
12) Heinie Groh—Like the peak. I don’t know what I was looking at that kept him off the edge of the ballot.
13) Fred Dunlap
14) Roger Bresnahan—After looking at Schang, I think I may have been overrating him. And underrating the rest of the catchers, Schalk in particular.
15) Wally Schang—Very close to Bresnahan. Both guys were great hitters for catchers, OBP especially. Maybe I’ll give Roger the nod because Wally owes a little bit of his longevity to the tools of ignorance that Bresnahan created.

New:
Chino Smith—Near McGraw and Cravath…that puts him around 50.
Edd Roush—Below GVH and J. Ryan crowd, 40-50 range.
Edgar Wesley—Not particularly noteworthy, but still in top 100.

Edd Roush vs. Max Carey:

Any thoughts?

Okay, I've looked through this a few ways, and I'm having problems not concluding "Carey by a country mile"!

The comparison is actually very easy, since they were almost exact contemporaries playing the same position in the same league.

Career WS: Carey 351, Roush 314. Advantage Carey

Career WARP-1: Carey 134.8, Roush 111.4. Advantage Carey

Top 5 years, by WS: Carey 133, Roush 136. Advantage: Draw

WS/162: Carey 23.0, Roush 25.9. Advantage Roush

Times Best CF in the NL: Carey, 8 (1916-1918, 1921-1925)(plus 2 times best LF, in 1912 and 1915), Roush 2 (1919 and 1920). Advantage Carey

Times Best of 2nd Best CF (or LF) in NL: Carey, 11 (add a second place in 1913), Roush 5 (add 2nd place in 1917-1918, and 1925, to Carey all three times) Advantage Carey

Head to Head: Carey played one full year before Roush (1914) and Roush played two full years after Carey (1929 and 1931). Looking at their 14 overlapping years, and awarding a tie for differences betwen appx 1 WARP or 3 WS: Carey better: 8 times (1915-1917, 1921-1925), Roush better: 3 times (1919-1920, 1926), essentially equal: 3 times (1918, 1927-1928). Advantage: Carey

John Murphy wrote:

Roush was the better player at his peak, but Carey had the longer career. I like Roush a little better, but I'm not sure if he'll make my ballot.

Actually, I don't think Carey had the longer career. Their careers were almost identical in length. Carey simply played more games per season, giving him 500 more games for his career. If you match them up game by game, perhaps Roush was SLIGHTLY better on a per-game basis, but if you match them up season by season, Carey wipes the floor with Roush.

Actually, I don't think Carey had the longer career. Their careers were almost identical in length. Carey simply played more games per season, giving him 500 more games for his career.

That's what I was referring to, Matt. More career value for Carey.

If you match them up game by game, perhaps Roush was SLIGHTLY better on a per-game basis

Actually, it appears to have been significant. Roush created 25.86 Win Shares per 162 Games, while Carey created 22.96.

PhillyBooster has almost won me over. I had Roush at 11 and Carey about 20th, but I restudied everything and can't sustain that conclusion.

If WS is your metric of choice, Roush is probably still slightly ahead b/c/

(1) His peak was better: His best four seasons beat Carey's by 15, an advantage that is big enough to keep him ahead through their best 11 seasons despite Carey winning each of seasons 5-19.
(2) He put up almost as much value while playing over 20% percent fewer games and using up almost a third fewer outs: As Carey was playing more games per season than Roush, his WS are inflated in comparison with Roush b/c/ of WS's low replacement level.

On the other hand, the gap between the players for seasons 1-4 is within or just over the system's 3-WS margin of error; Carey does win EVERY season from 5-19; and Carey ends up with 37 more total WS. So, if you think that Carey deserves extra credit for showing up a lot more often than Roush than it wouldn't be wrong to say he wins the WS comparison.

If you use WARP, it is a clear victory for Carey. Roush wins their best season comparison by an insignificant margin (0.4 WARP) and then Carey beats him for years 2-20 by margins that range from very small (4 times under 0.5 WARP) to moderate (4 times under 2 WARP) but which usually fall between 1-1.5 WARP. According to WARP, the two contemporary CF had very similarly shaped careers but, on a year-in, year-out basis Carey was about 1 win per season more valuable.

I think the two systems nicely capture the range of possibilities: the best case scenario for Roush is that he put up slightly less peak-adjusted value as Carey while using up about 30% less outs. That would seem to justify a somewhat higher ballot placement, but not a huge gap. the worst case scenario for him is that both guys had very similar careers, but Careywas consistently 1-game per season worse than his contemporary. If you buy that reading, then Carey should rank substantially ahead.

This is tough.

correction: when I refere to moderate margins above the parenthetical should read "(4 times OVER 2 WARP)"

Actually, it appears to have been significant. Roush created 25.86 Win Shares per 162 Games, while Carey created 22.96.

But this is where "per 162 games" gives us troubles. Carey was more likely to actually PLAY 162 games.

Let's line 'em up by games played, over their best 14 years (picked to help Roush, you had exactly 14 seasons of over 100 games -- Carey had 3 more):

Carey: 156, 155, 155, 154, 154, 153, 150, 149, 144, 140, 140, 133, 130, 129. Avg.: 145 games

Roush: 149, 145, 144, 140, 138, 136, 134, 133, 121, 115, 113, 112, 108, 101. Avg. 127 games

Comparing Rate Stats are helpful if the players played the same length seasons, but one player just happened to play more seasons. Here, that is not the case. In the average season in their prime, Carey was playing 18 more games per season.

Saying Roush was 3 Win Shares better over 162 games, here, is not the same as saying he was 3 win shares better per year. Roush's higher rate is actually spread over 18 more team games.

Carey played 162 games every 1.12 seasons. Roush played 162 games every 1.28 seasons.

If the rate question was (and I think this is a better rate question) how many Win Shares did each player have per year, then the answer -- over their 14 year peaks -- is that Carey gave you 20.50 WS per season (145 games at a 22.96/162 rate) while Roush gave you 20.20 WS per season (127 games at a 25.86/162 rate).

So, if you use seasons as your rate metric, Carey was actually better in both career AND rate.

JM:

Actually, it appears to have been significant. Roush created 25.86 Win Shares per 162 Games, while Carey created 22.96.

AS:

As Carey was playing more games per season than Roush

This is looking like the old Ripken/Larkin argument. Roush played 140 games only 3 times (+1 FL) while Carey played 140 games 11 times. Drop in down to 130 and its 7(+1) for Roush and 13 for Carey.

Even as a peak voter I recognize that in-season durability issues have the effect of lowering peak relative to guys that play every day. That is, I'm a "pennants added" type of guy in that I think the shortest meaningful length of time for judging a player is an individual season.

There are no interpositional differences in playing time expectations here either. These guys played the same position in the same league.

I agree with Andrew, this is tough.

Oh... Philly's post came as I was typing in mine. He says exactly what I was trying to say, except he phrases it quite a bit better than I did. :-)

Just to reiterate where I am in the Roush/Carey battle, I have Roush slightly ahead of Carey. I don't think it's a slam dunk for Edd and I can see someone selecting Carey over him. I just don't feel it's a huge difference between them either way.

Do we take into account player's managerial careers too?

Re Carey & Roush in WARP & WS:

The difference in view of Carey & Roush has less to do with their evaluation of the individual players than with their evaluation of the importance of centerfield defense in general.

WS
Roush 255.7 batting + 58.3 fielding = 314 total
Carey 254 batting + 94.8 fielding = 351 total

WS sees Roush as the better hitter, substantially better by rate, and sees Carey as by far the better fielder, with 1.6X more fielding WS than Roush

WARP
Roush .295 EQA, 594 brar + 426 frar = 114.4 WARP1
Carey .283 EQA, 623 brar + 605 frar = 134.8 WARP1

WARP sees Roush as the better hitter by rate, though he's not as far superior a hitter to Carey as WS counts it. WARP sees Carey as by far the better fielder, with 1.4X more frar above replacement than Roush.

In that respect, the two systems assess the pair _very_ similarly. The difference, of course, is in the relative weight of batting and fielding in total value. According to WARP, 46% of Roush & Carey's combined value is defensive. According to WS, it's 23%.
WARP places twice the weight on defense, in this case, than WS does.

Do we take into account player's managerial careers too?

No, we don't, Buddah.

On the Roush vs. Carey issue, I need to repeat a point that we used to make all the time but has kind of gotten lost in the shuffle.

If Player A (call him Carey) plays 25% more games than Player B (call him Roush), simply comparing their WS overrates Player A/Carey.

Because WS doesn't use a true replacement value (it claims not to use a replacement level at all but that isn't really true), a player's WS include his value above replacement plus a showing-up bonus that any replacement level player would have earned given that playing time. If two guys play the same amount of games, their unearned showing-up bonuses cancel out. If, however, one guy plays a lot more than the other guy then the WS for his additional playing time will include both earned WS above true replacement and unearned showing-up WS. To compare the two guys, you need to discount the guy who played more's WS by some difficult to calculate percentage so as not to credit him with WS that any Tom, Dick, or Harry would have earned given the opportunity.

To state that more succinctly, comparing Carey's WS earned in 150 games to Roush's WS earned in 130 games overrates Carey b/c/ it does not take into account the portion of his WS for the last 20 games that his team would have earned simply by rolling out a replacement level stiff.

I think Carey, or Player A in general, should get credit "for the portion of his WS...his team would have earned simply by rolling out a replacement level stiff". I may be in the minority, but I think playing at replacement level has value.

That said, neither Carey nor Roush make my ballot.

Daryn--

You might be right that playing at replacement level has value, but there has to be some level at which playing no longer has value. Unless you think that level is at or below about a .280 winning percentage, you need to deduct some of the WS from the player who played more games to get an accurate comparison.

I may be in the minority, but I think playing at replacement level has value.

Let me take this issue from another angle. What "replacement level" means in theory is the quality of player that any team can immediately find to fill a position. That player will contribute to team wins, because bad teams beat good teams in baseball with some frequency. Therefore, that player should, theoretically, earn win shares. But, if you want to calculate the true cost to Edd Roush's team of his fragility relative to Carey, you need to include the value of the replacement level player who takes Roush's place -- Roush's injury (or his hold-out), doesn't deprive the team of replacement level work. If he adds enough value in the games he does play, it may benefit a team more to use him when he's healthy and fill in with Joe Replacement when he's not than to trade Roush straight up for Max Carey.

Here's a model calculation: Let's set replacement level at 9 ws/154 g, which is about what players on a .250 team would earn. Carey's rate of ws per 154 g = 21.82; Roush's 24.58. I'm not saying that the results of this calculation _will determine_ whether Roush or Carey was better, but it will show how replacement level figures in to the equation.

Carey's team earns from its centerfielders

21.82*145/154 + 9 * 9/154 = 20.54 + .53 = 21.07

Roush's team earns from its centerfielders

24.58*127/154 + 9 * 27/154 = 20.27 + 1.58 = 22.85

So, in this model, the team with Roush gets more production from centerfield, even though Roush's personal contribution is actual slightly lower than the more durable Carey's contribution.

The issue, then, is less whether or not playing at replacement level has value, but how much value it has and how you incorporate the value of replacement level into your assessment of a player's merit.

Great analysis, Chris C. That's the type of peak-rate I am interested as a peak voter...

DavidFoss,

If you're interested in true peak rate, I don't think the numbers I've given provide that. There've been a lot of (highly informative) posts on Carey and Roush, so I grabbed data of different sorts to throw together this example, but the data I used doesn't give a truly accurate peak rate according to win shares.

The avg. games played comes from Phillybooster's analysis of their 14 best seasons in post 32.

The rate comes from John Murphy's post 29, and I _think_ John was giving us career rates. Their actual rates for their 14 best seasons as listed by Phillybooster would be higher, and their rates for their shorter periods of truly peak performance would be higher yet.

Those rates could be calculated fairly easily and turned into team seasonal values according to this method, but I haven't done so.

Carey's team earns from its centerfielders

21.82*145/154 + 9 * 9/154 = 20.54 + .53 = 21.07

Roush's team earns from its centerfielders

24.58*127/154 + 9 * 27/154 = 20.27 + 1.58 = 22.85

While this analysis is probably correct in addressing the value provided by centerfielders in the WS system, the difficulty is that we are essentially adding two players together.

If I reasonable conclude that Max Carey was worth more than Edd Roush, but that (Edd Roush + Charlie See) was worth more than (Max Carey + Fred Nicholson), that may help the Pirates against the Reds in the standings, but it doesn't put Edd over Max on my ballot.

And, if fact, there is a real value to "just showing up." Baltimore didn't have to waste a roster spot of "backup shortstop" for 20 years, giving them more roster flexibility than its competition. Similarly, if I'm an owner looking at going with Carey and Roush, I'm thinking to myself "If I go with Roush, I've got to make sure I've also got another guy on my bench who can handle center for 30 games per year. If I go with Carey, I can expend that roster spot on an immovable post with power."

While this analysis is probably correct in addressing the value provided by centerfielders in the WS system, the difficulty is that we are essentially adding two players together.

That's why many people think it's a good idea to set replacement level at zero.

If you're interested in true peak rate, I don't think the numbers I've given provide that.

OK... it was more the algorithm I was impressed with and not the exact numbers. Some of the inputs are career, some are averaged, and the replacement level of 9 WS is probably not etched in stone either.

One number that could come out of this is Win Shares Above Replacement (WSAR):

Carey: (21.82 - 9)*145/154 = 12.07 WSAR
Roush: (24.58 - 9)*127/154 = 12.85 WSAR

...with the caveat that I'm using rate numbers and replacement levels that might not be quite right, but using them to illustrate the formula.

Anyhow, this would take into effect in-season durability issues and not count multiple players.

The rate comes from John Murphy's post 29, and I _think_ John was giving us career rates.

Correct.

Approximate replacement levels with Win Shares:

Assume we have a .240 team (a little better than the 1916 A's). It goes 37-117, with 111 Win Shares to distribute.

Assume it is equally bad offensively and defensively, scoring at 3/4's of the league rate and giving up runs at 4/3's of the league rate. (a .240 team by Pythagorus).

The distribution then is 67 Batting WS, 14 Fielding WS, 30 Pitching WS.

This works out to 10 Win Shares apiece for the 8 position players, and 6 Win Shares for each of 5 starting pitchers (about 260 IP each).

This also shows how pitchers do not receive the same bonus as position players for "just showing up".

The distribution then is 67 Batting WS, 14 Fielding WS, 30 Pitching WS.

Aarghh! Forgot the 48-52.

The distribution then is 65 Batting WS, 15 Fielding WS, 31 Pitching WS.

Doesn't change the final conclusion, it just increases the likelihood that a pitcher gets that last unallocated 111th Win Share.

6 Win Shares for each of 5 starting pitchers (about 260 IP each).

How many IP per WS-Rep?

1937 Preliminary Ballot

Time to give this a try.

Well, the ballot is suddenly much less top-heavy than it’s been, with the backlog of all-time greats finally elected. Still, there are a lot of great players on this ballot, and several serious candidates arriving this year. Edd Roush, Dick Redding, and Wally Schang all break into my top 40.

1. Cristobal Torriente. (3) Reaches the top. A shoo-in for 1937.
2. Harry Heilmann. (4) Right behind Torriente.
3. Stan Coveleski. (5). Unlike the other eligible pitchers with similar records, Coveleski didn’t benefit from above-average run support or fielding support.
4. Clark Griffith. (6) Best remaining player from the still-underrepresented 1890s. Statistically similar to Coveleski, but at a time when his innings pitched didn’t mean as much. I hope he will get a very serious look for election in 1939-40.
5. Hughie Jennings (7) During his 1894-1898 peak, he was the best position player in baseball, and better than a pair of contemporary first-ballot HoMers, Billy Hamilton and Ed Delahanty, who were also at their peaks during these years.
6. Mickey Welch. (8) 8th-10th best player of the 1880s. We have “enough” 1880s pitchers, but not necessarily the right ones. Welch belongs in.
7. Heinie Groh (9). One of the best peaks of the teens, setting aside the all-time greats.
8. Jose Mendez (10) Career Cuban winter-league record of 59-18, if achieved with avg. support, would translate into a 150 ERA+ in major-league competition. Even assuming he was pitching for good teams, they would have to have been powerhouses indeed for Mendez to get that W-L record without being a solid HoMer. Negro League record of 40-14 tells the same story, and much of that record came while pitching for the Cuban Stars, who didn’t do much more than break even against the American black teams of that era.
9. Max Carey (11). Despite not being as valuable on a per-game basis as Roush, his durability pushes him slightly ahead. My weighing of peak, career, and rate stats has Carey and Roush dead even, so if WS is undervaluing defense _at all_, then Carey would rank ahead. That breaks the tie.
10. Edd Roush (n/e). Less career than Carey or Van Haltren, but has the peak that they lack. A great contributor to a World Series winner.
11. George Van Haltren (12) All-around, consistent talent.
12. Tommy Leach (13) Another player similar to Carey. Last star of the aughts who is a serious candidate for election in my view. He slipped through the cracks in the teens elections, but I think he’ll continue to rise now.
13. Lip Pike. (14) Still around (35th consecutive appearance on my ballot), treading water until the next gap in great new arrivals gives an early player another shot at election. He had a great peak, however one adjusts for era.
14. Urban Shocker (14) A very underrated player; he might well be a HoMer. He had a couple of great seasons in 1920 and 1921, and he was above average every single year he pitched. In comparing Shocker to a pitcher like Waddell, the electorate should keep in mind that average innings pitched for a starting pitcher dropped from an average of 277 for 1900-1909 to 230 for 1917-1926 as conditions for pitchers became increasingly difficult. In that context, Shocker’s innings-pitched totals are as good as Waddell’s, and he was a more consistently effective pitcher.
15. George Sisler (16). Rises up onto my ballot this year.

Top Ten Returning Players from 1935 Who Don’t Make my 1936 Ballot

Rube Waddell; see #22 below
Jake Beckley: see #39 below

1936 Off-Ballot

16. Cannonball Dick Redding (n/e). Just misses the ballot in his first year of eligibility. Best Negro-League pitcher of the late teens and early twenties after Smokey Joe Williams. His record is not all that impressive, but he often pitched for bad teams, especially in the latter part of his career with the Brooklyn Royal Giants. Negro League pitchers are the hardest group of players to rank, period, so I’m not yet confident in my placement of him.
17. Larry Doyle (17).
18. Spotswood Poles (18)
19. Harry Hooper (19).
20. Hugh Duffy (20).
21. Wilbur Cooper (21)

-------in-out line lands about here. The ballot is very flat from 14-28; I think about half of the players in that group of 15 will eventually be elected; I’d pick the players from here up. ----------

22. Rube Waddell (22) See Shocker comment above for more on how I compare Waddell to later pitchers. Waddell was a great talent, and he was one of the greatest characters in the history of major-league baseball. He’s thus deserving of his place in the Hall of Fame, but I think his value is just below the threshold for Hall of Merit induction. In the context of his time, just the eighth-best pitcher of the aughts.
23. Carl Mays (23).
24. Ben Taylor (24)
25. Bobby Veach (25)
26. Roger Bresnahan (26)
27. Jimmy Ryan (27)
28. Cupid Childs (28).
29. Fielder Jones (29)
30. Dobie Moore (30)
31. Gavvy Cravath (31)
32. Herman Long (32)
33. Tommy Bond (33)
34. George J. Burns (34)
35. Charley Jones (35)
36. Bruce Petway (36)
37. Bill Monroe (37)
38. Babe Adams (38)
39. Wally Schang (n/e). Quite a few favorable things have been said about Schang, and he was a very good player for quite a long time, but nothing about his career makes him look like a HoMer to me. Among eligible catchers, I prefer both Petway and Bresnahan. For those worried about the underrepresentation of catcher, I’ll note that after the 1936 season, there are five active catchers who look much better than Bresnahan, Petway, or Schang. Hartnett, Cochrane, and Mackey already look like future HoMers. Bill Dickey has put in eight strong years for the Yankees so far, and Josh Gibson has cut quite a swath in his first seasons in the Negro Leagues. If he keeps hitting like this, he could rank among the all-time greats.
40. Jake Beckley (38) Like Childs, Beckley just doesn’t appear outstanding in comparison to his contemporaries. Lack of better first-basemen could give him a positional boost, but right now I don’t see the justification for a positional bonus for first base.

Dropping out of the Top 40

Dave Bancroft (40). He’ll be back next year.

How many IP per WS-Rep?

It looks like about 45 IP of replacement level pitching will earn 1 Win Share, as will about 15 games of replacement level play at a position.

This hypothesis can be tested by examining a number of 1 and 2 Win Share seasons and seeing how much playing time was actually given to the player.

So it's the 1936/37 offseason. You know, the Babe has retired. He's been gone for a year and half, so I think we can safely assume he's not going to make a comeback. Our rules say we have to wait a few more years to vote for him, but we already know what his record is and can soak up the meaning of that. His career OPS+ is 207. Did you know that Ty Cobb had only one season above that number, and only barely above, at 209?

But Lou Gehrig just turned 33 in the middle of last season, and he hasn't shown the slightest sign of decline. If Ruth, with all his bad habits, could be such an offensive force in his mid and late 30's and play until he was 40, then why shouldn't Gehrig, who doesn't share those bad habits, also play well until he's 40 or more?

(And don't tell me about some guy playing high school ball in San Diego. That's high school; anyone can dominate at that level, right?)

with the Hall of Fame only electing the legends of the game it looks like they will equal or surpass the 40 year old Hall of Merit in the quality of their selections.

we interrupt this broadcast with breaking news
the Centennial Commission just announced their picks for the hall of fame
John McGraw, Ban Johnson, Connie Mack, George Wright and Morgan Buckley

their goes that Idea

If Ruth, with all his bad habits, could be such an offensive force in his mid and late 30's and play until he was 40

Whatever it was that he was doing, it must have worked. The Babe aged better than anyone else before him and better than only a choice few after(Williams, Bonds, maybe Aaron). Who else can boast an OPS+ decline of 176, 161 at the ages of 38 & 39?

(And don't tell me about some guy playing high school ball in San Diego. That's high school; anyone can dominate at that level, right?)

With the folding of the Hollywood Stars, the San Diego Padres started a new team in the PCL as a Red Sox affiliate! Just in time for The Kid to graduate from high school.

Ted did alright... he joined the team in late June. He was hitting only .229 in spot duty at the end of August when he turned 18. They played him every day after that and he batted .308 in September to bring his yearly total up to .271 (29-107, 8-2b, 2-3b, 0-hr). Not bad for a kid his age in the PCL, but its still early. Teammate Bobby Doerr appears on a faster track at the moment.

we interrupt this broadcast with breaking news
the Centennial Commission just announced their picks for the hall of fame
John McGraw, Ban Johnson, Connie Mack, George Wright and Morgan Buckley

their goes that Idea

I agree with you in spirit, yest, but I agree with all of the choices except for the Governor of Connecticut. McGraw, Johnson and Mack are no-brainers for the Management wing of the HoM, while Wright was a terrific shortstop (I'm assuming he wouldn't have been on your ballot though :-).

Assume we have a .240 team (a little better than the 1916 A's). It goes 37-117, with 111 Win Shares to distribute.
. . .
This works out to 10 Win Shares apiece for the 8 position players, and 6 Win Shares for each of 5 starting pitchers (about 260 IP each).

Not to be overly picky, but is the proper definition of a "replacement player" really "an average player on a replacement level team"?

If the 1916 A's are our "replacement level team", then when we look at the team, we would see that about half performed above replacement level and half performed below replacement level. Isn't the replacement level player really the worst regular on that team? Or is it credible that the team actually played 4-5 sub-replacement players for the whole season?

One thing I hadn't noticed before was that the A's, with players like Jack "1-20" Nabors, actually led the league in complete games that year! And looking at their bullpen, it makes complete sense why.

Okay, I'm having the same problem others are having with Schang. I was all prepared to put him on my ballot, and then I did a season-by-season analysis.

Here's what I got:

Career: 1913-1929

Best catcher in the AL: Once (1919)
Second best catcher in AL: Twice (1914 and 1917)

Head to head with Ray Schalk, who had exactly same career, except 2 years shorter (AL catcher, 1912-1926).

Here's how they did, in their 14 overlapping years:

Schalk better 8 times: 1913-1917, 1920, 1922, 1925
Shange better 5 times: 1918-1919, 1921, 1924, 1926
Tied one time: 1923
The post-1922 leads are pyrrhic, however, as neither player were among the best catchers in the league (Ruel and Cochrane were).

Conclusion: Despite my initial assumptions, Schang will not make my ballot. More interesting question, though, is whether this comparison upgrades Schalk enough for HIM to make my ballot. Schalk was the best catcher in the AL five times, and second best another 4 times. That should get him up into the Bresnahan 10-15 range.

The Babe aged better than anyone else before him and better than only a choice few after(Williams, Bonds, maybe Aaron).

Well, there was Wagner, who had his single best season at 34 and was a full-time shortstop with OPS+ 126 at 41. One question about the Babe: is it known whether he did any weightlifting or other strength training?

Re "What is replacement level"?

This is a question that perhaps belongs over on the WARP vs. WS thread, but since we've been discussing here I'll ask here.

Looking at the team record for the 36-117 Philadelphia A's, our example replacement level team, over on BP, I notice that this team, as a team, earned 51.5 WARP1. What could replacement level possibly mean in this case?

Harry Heilmann is about to be elected; Mike Tiernan dropped out of the voting last year.

See my posts #14 and 16 above for explanation of what this is:

Jackson      60   63   75   230
Heilmann     63   52   88   229
Clarke       60   39   88   206
Magee        54   40   78   192
Goslin       49   30   78   173
Tiernan      41   29   61   146
Thompson     39   24   58   132
Burns        34   25   57   128
Cravath      36   27   49   126
Browning     37   22   52   122

Again, no league strength adjustments or even season-length adjustments. Conclusion: Heilmann is by far too much hitter to ignore, a level above Tiernan/Thompson/Browning.

Since he's recently retired, I might as mention the guy who blows this scale to pieces at 157-241-183-701.

Another guy I'm struggling to compare Roush to is Hugh Duffy. I've been revisiting Duffy this week, and if you

a) adjust their WS to 162
b) then knock 15% off of Roush's time in the Feds and
c) and discount 91AA for Duffy,

their ten best seasons look like this:

DUFFY
41 34 31 30 30 28 28 26 21 20

ROUSCH
36 35 32 29 26 24 22 21 20 19

TOP 3
HD 106
ER 102

TOP 5
HD 166
ER 158

TOP 10
HD 289
ER 264

CAREER
HD 337
ER 330

Their rate of rawWS/G is
HD 27.5
ER 25.9

In the past in these threads, we've said the NL was a weak league during much, if not all, of Roush's career, while the NL was tough during much of Duffy's career. It looks like Duffy's got small advantages across the board unless you timeline him.

Can a FOER or an EOHD (or anyone else for that matter!) tell me why I should think that Roush is better than Duffy or visa versa?

Phillybooster #28
The comparison is actually very easy, since they were almost exact contemporaries playing the same position in the same league.
. . .
Head to Head: Carey played one full year before Roush (1914) and Roush played two full years after Carey (1929 and 1931). Looking at their 14 overlapping years,
. . .

Carey played four full seasons, 1911-1914, before Roush played his first, 1915. Roush was his team's regular CF from 1915, Carey only from 1916; he succeeded manager Fred Clarke as the regular LF, 1912-1915.

(Carey played some LF in 1916, under new manager Jimmy Callahan. 20-game CF Ed Barney played with Pittsburgh in 1915, his second of two teams that season, and he played his last MLB game in 1916, July 2. A quick search shows me that Barney played {1,CF} and Carey {2,LF}, May 7. I suppose that Barney was the end-1915 incumbent and that Callahan retained him in the role at the beginning of 1916.)

Can a FOER or an EOHD (or anyone else for that matter!) tell me why I should think that Roush is better than Duffy or visa versa?

I can't. Roush, Carey, Duffy & Van Haltren pretty close, in my opinion. Sure, we can break out the magnifying glasses to rank them, but almost any ranking of the above four could be argued.

The failure of one of them to stand out amongst the others is one of the reasons why none of them are likely to make my ballot this year.

Can a FOER or an EOHD (or anyone else for that matter!) tell me why I should think that Roush is better than Duffy or visa versa?

League quality is a big question here. WARP seems to treat the quality of competition as slightly higher in the one-league 1890s than in the NL of the 1910s and into the 1920s. I'm skeptical of that conclusion, mostly because there's pretty clear evidence that fielding and pitching improved substantially from the late 1890s through the aughts. I give Roush and Carey credit for playing against better competition than Duffy and Van Haltren.

I also place Duffy a bit lower than the other three (who I rank consecutively, 9-11, on my prelim ballot) because I think Duffy's win share numbers are inflated by his team's consistent beating of their Pythagorean projections. I don't say that it's random, but I credit the brilliance of Frank Selee more than the players' knowing how to win.

Still I have Duffy at #20 and see him as potentially electable.

Carey played four full seasons, 1911-1914, before Roush played his first, 1915.

That is right. I can't figure out exactly what I was looking at wrong that made me lop 3 years off of Carey's career. Since that part of the analysis looked at their overlapping years, however, it doesn't affect the results.

League quality is a big question here. WARP seems to treat the quality of competition as slightly higher in the one-league 1890s than in the NL of the 1910s and into the 1920s. I'm skeptical of that conclusion, mostly because there's pretty clear evidence that fielding and pitching improved substantially from the late 1890s through the aughts.

I'm with you all the way here, Chris. There are too many year in between those two eras for me to believe that the players from the 1890s would still be stronger than the NL teen and "Roaring Twenties" years.

there's pretty clear evidence that fielding and pitching improved substantially from the late 1890s through the aughts.

WARP does not contradict this. It considers the overall level of competition in the teens and twenties as higher than in the nineties. (Or at least it did when I did my study of this a couple of WARP revisions ago.) The overall MLB average had improved.

There are two factors involved in the WARP-1 to WARP-3 conversions. One is the "quality of competition", the other is the schedule length. 1890's players will benefit from the latter because they played 132 G most years, not 154 G. Is this getting mixed in with the "quality of competition" comparisons?

testing

Looking at the team record for the 36-117 Philadelphia A's, our example replacement level team, over on BP, I notice that this team, as a team, earned 51.5 WARP1. What could replacement level possibly mean in this case?


This is just a guess, but I would assume that WARP is based upon individual component stats, divorced from actual runs or wins.

The A's underperformed their pythagorean W/L by five wins, and we would all understand (if not agree with) a system that used 41 wins as the baseline, rather than 37.

But beyond that, the A's actually scored very few runs based on their hits and walks. As just a rough approximation, I looked at (TB+BB)/R. The A's scored 1 run for every 4.42 TB+BB. The 7th place Senators scored 1 run every 3.91. The first place teams was in the 3.50s. I would guess that the Phila A's "scattered" their hits a lot in 1916, so that not only did they underperform Pythagorus, but they underperformed the number of "runs scored" that would have gone into the formula by a pretty large amount.

It seems to me that the new WARP is misnamed; it no longer is Wins Above Replacement, but a kind of WARP-Shares. The team numbers appear to add up to approximately the number of wins in the league. However, as PhillyBooster points out, the players appear to share, not their actual number of wins, but the number of wins they should have had (based on their statistics). The shares are apparently handed out based on total runs-above-replacement.

Personally, I think I'd prefer to have a metric that divvied up the actual wins, but based on the runs-above-replacement measurements (once they get the fielding replacment level adjusted; it's better than it was, but still a little too high).

It considers the overall level of competition in the teens and twenties as higher than in the nineties. (Or at least it did when I did my study of this a couple of WARP revisions ago.) The overall MLB average had improved.

jimd, I didn't do an extensive study; I just compared Roush's WARP1 to his WARP2 and Duffy's WARP1 to his WARP2 (didn't mess with WARP3 at all). The reduction from W2 to W1 for Duffy about .78; the ratio for Roush is typically between .70 and .75.

Maybe I'm making too much of this small bit of data, but it's my impression that it's representative of what I've seen when looking at other comparisons between 1890s stars and NL stars of the teens and twenties.

It seems to me that the new WARP is misnamed; it no longer is Wins Above Replacement, but a kind of WARP-Shares. The team numbers appear to add up to approximately the number of wins in the league.

This makes sense, and you're right, the name is now misleading. I assume that wins above replacement means wins above the number that a replacement player would earn.

I nevertheless can't quite wrap my mind around the idea that a team made up entirely of replacement level players would have 0 WARP, but some small number of wins (let's imagine it's 20 out of 162, to go outrageously low), so some team with 65 wins would be 45 games better than the replacement level team, but have 65 WARP. I guess the zero point in win shares creates the same problem, but it's not that important because the zero point is low enough that the difference between the zero point and zero wins really doesn't matter much most of the time . . .

I agree with you in spirit, yest, but I agree with all of the choices except for the Governor of Connecticut.
He was the source of the comment

while Wright was a terrific shortstop (I'm assuming he wouldn't have been on your ballot though
He's in my pHoM

He's in my pHoM

Good to hear it, yest!

Here's my provisional ballot for 1937:
1. Cristobal Torriente
2. Jake Beckley
3. Mickey Welch
4. Harry Heilmann
5. Max Carey
6. Ben Taylor
7. Tommy Leach
8. Edd Roush--solid career, good offensive value, good defense, just a notch below Carey & Taylor.
9. Carl Mays
10. Jose Mendez
11. Jim McCormick
12. Vic Willis
13. Dick Redding--It's quite possible he was better than Mendez, but I'm not sure yet. This is where he goes for now.
14. Stan Coveleski
15. Dave Bancroft

74. Wally Schang--close to both Breshnahan and Schalk, but well below the serious ballot contenders. In my book, Bruce Petway is still far and away the best catcher eligible.

No other new eligible is in my top 80.

Need to leave the explanations for later (probably the ballot thrread) but here is my current thinking on the ballot:

(1) Torriente (3rd)
(2) Heilmann (4th)
(3) Van Haltren (6th)
(4) Coveleski (8th)
(5) Childs (5th)
(6) Jennings (10th)
(7) Groh (7th)
(8) Duffy (9th)
(9) Pike (12th)
(10) Roush (new)
(11) Ryan (14th)
(12) Chance (11th)
(13) Willis (15th)
(14) Carey (new to me)
(15) Sisler (new to me)

16-20: McGraw, Redding, C. Jones, Beckley (13th), Doyle.

21-30: Bresnahan, Mendez, Schang, Veach, Griffin, Browning, Monroe, Welch, Leach, Tiernan.

31-40: Cicotte, Dunlap, Cooper, Cravath, Poles, Griffith, Shocker, Fournier, Williamson, Taylor.

This is pretty close to impossible. I've got Coveleski 4th and Carl Mays 41st and I'm only 80% sure that Coveleski was better.

Prelim Top 20 (With a New, But Old, Entrant at 20)

1. Harry Heilman – I like the batting average and the OPS+.

2. Mickey Welch – 300 wins, lots of grey ink. RSI data is helping Welch – those wins are real. Compares fairly well to Keefe.

3. Cristobal Torriente – nice ball player – maybe a little worse than Clemente will be.

4. Jake Beckley -- ~3000 hits but no black ink at all. Baseballreality.com has him as the best first baseman in baseball for a long time. Crawford (HOMer) and Wheat (HOMer) are two of his three most similars.

5. George Sisler – how can you keep the guy who has the second best single season hit total out of the Hall of Merit? Hits impress me and he had a lot of them, plus a better peak than Beckley. I’ve put Beckley ahead of him because I’m a career voter.

6. Rube Waddell -- I like the three times ERA+ lead, the career 134 ERA+ and, of course, all those strikeouts (plus the 1905 Triple Crown).

7. Lip Pike – 4 monster seasons, 4 more not too bad, plus 4 undocumented.

8. Roger Bresnahan – Great OBP, arguably the best catcher in baseball for a six year period. Counting stats, like all catchers of this time and earlier, are really poor. I like him better than Schang because he compared better to his contemporaries, if you count him as a catcher.

9. Clark Griffith – 921 similarity score with mcginnity, who was 1st on my ballot when elected. That being said, he is barely better than what is now a 10 person pitching glut.

10. Redding – probably the 6th or 7th best blackball pitcher of all-time (behind, at least, Williams, Paige, Foster, Foster and Rogan), and that is good enough for me.

11. Cupid Childs – nice obp.

12. Pete Browning – Joe Jackson’s most similar player, and they are pretty close – I have him as about 4/5ths of Jackson, who was 2nd on my ballot when elected.

13. Tommy Leach – 300+ WS has to mean something.

14. Heinie Groh – hard for me to analyze – I know he is not better than Leach in my mind and he is definitely better than the rest of the thirdbasemen. I’ll be glad when you (and by saying you he excludes himself from the community – please don’t set my teeth on edge with a scathing rebuke) elect him.

15. Stan Coveleski – I can’t really see a huge difference between Waddell (9) and Shocker (mid-30s). So Coveleski is somewhere in between. I really wish someone could do an analysis of the differences between the eligible pitchers who have between 190 and 215 wins. I can’t tell them apart without a program.

16. Bill Munroe – I think he was pretty good. Any blackball player that is even talked about as among the best 70 years later is pretty good. I’ll take McGraw’s word for it.

17. Jose Mendez – somewhere between here and Waddell seems about right.

18. Addie Joss – I don’t like short careers much, but I cannot ignore the second best all-time ERA, the 12th best ERA+ and the nice winning percentage. Could be below Duffy (at 26).

19. Schang – I’d like more catchers in the HoM, but this isn’t a cocktail party.

20. Jennings – Bump up from me (see the 1936 ballot thread about the Orioles’ pitchers not being very good) -- he may get more consideration from me a few decades from now, when Koufax starts to exist. But he was no Koufax.

It has been a long lime, but I think I am back. The move has been hectic. We can finally get into our house next Wednesday, but we won't move in until Friday. We have spent the last 5 weeks in a hotel because our sellers were not able to move out - the house they bought had to have a lot of work done. I have gotten settled in the Tax program so I think I have time to participate again. Now, we just have to wait for our stuff that is in storage to be shipped to us - sometime by Nov 5...Unfortunately, all my baseball books are in storage!!! I didn't want to vote without having good information, but I think I have the time now.
If someone has the time, could you post or email the following info so I could make an informed vote? For Max Carey, Harry Heilmann, George Sisler, Edd Roush, and Wally Schang: Best total of Win Shares in 3 consecutive years, Best total of Win Shares in any 7 years, Years with 20+ Win Shares and 30+ Win Shares. Or if there is an on-line source for this. I checked the Seattle Library system, but the two copies of Win Shares aren't do back until the end of the month.
Thank you in advance, and its good to be back.

Still haven't gotten around to that OF reconsideration. The list of those to consider gets one deeper this week with Roush. I have him toward the bottom of the huge group, one slot ahead of George Burns.

Schang is close to the ballot for me, but just off of it. Top eligible catcher, and has a chance to eke onto my ballot someday.

Redding and Smith I discussed in their threads.

-------------------------

1. Torriente
2. Heilman
3. Jennings
4. Groh
5. Coveleski
6. Poles
7. Mendez
8. Waddell
9. Browning
10. Veach
11. Redding
12. Duffy
13. F. Jones
14. Griffin
15. Moore

Preliminary ballot

A lot of flux over the past couple weeks, and this week's no different as I reconsider my reconsiderations and try to get a better handle on the pitching and OF gluts. Sigh...still need to figure out a best practice for working with catchers....

1. Torriente
2. Heilmann
3. Groh: Gets the nod over "G-Van" because he played third at a time when third was further along the spectrum than CF and its attrition rates were much higher.
4. Van H.: Back where he belongs.
5. Duffy: First time I've voted for him since 1933, but I've been underrating him.
6. Poles: Still one of his best friends.
7. Monroe: Ditto.
8. Coveleski: Best hurler on the ballot.
9. Griffith: The Coveleski of his generation.
10. Redding: Very tight between Stan, Clark, and Cannonball ("D-Red"?).
11. Roush: This feels about right.
12. Willis: I've never voted for him before. As Chris J.'s RSI notes point out, there's little to explain those 10 missing wins. A little better than Cooper who ends up roughly 16th. It's tight at the bottom of the ballot.
13. Childs: Back on my ballot after a year away. Third-best eligible infielder.
14. Leach: First time I've voted for him, but this is why we keep tinkering and reconsidering and trying to get things right. Not quite as good as Roush, just a little better than Ryan thanks to his play at 3B.
15. Mendez: His big peak plus a couple more shoulder years gets him into the caboose slot on my ballot over Cooper. Could go either way.

Wally Berger may be the best centerfielder in baseball but casual fans don't know it because he plays for the Braves. He slipped a little last year but only a little and he's only thirty.

Dizzy Dean is something else. You can't count out the Cardinals while he's around. Hubbell carried the Giants last year but he's getting older (aren't we all?) and I don't see him doing it again.

The first division is so strong. All four of them would beat anyone but the Yankees, I think.

How do you pronounce Heilmann? Is it HILE-man, HELL-man or something else?

Since it's a German-looking name, I've always pronounced it HILE-man myself, though, to be honest, I've never heard it spoken.

Since it's a German-looking name, I've always pronounced it HILE-man myself, though, to be honest, I've never heard it spoken.

That's the way I have always pronounced it, too.

BTW, does anyone know of a record that Heilmann owned at the time of his retirement? I can't even find a club record of his to list on his probable plaque.

His 1921/3/5/7 batting averages were truly remarkable, even if by a few decimals he failed to hit .400 in 4 alternate years.

BTW, does anyone know of a record that Heilmann owned at the time of his retirement? I can't even find a club record of his to list on his probable plaque.

Club records at the time of retirement include best Slugging percentage (season & career) and most RBI (season), most doubles (season).

Maybe a few more here

His second-to-Cobb list appears quite numerous.
But as of even now (1937), Greenberg as broken (often shattered) almost all the club records Harry did have his hold on.

His 1921/3/5/7 batting averages were truly remarkable, even if by a few decimals he failed to hit .400 in 4 alternate years.

While I will mention his alternate BA on his plaque., that's more trivial than anything.

Club records at the time of retirement include best Slugging percentage (season & career) and most RBI (season), most doubles (season).

Forgot about the season records. Thanks, David! Heilmann's plaque needs all the padding it can take.

John,

I think 4 batting titles, 12th best all-time batting average and seven consecutive years in the top 5 in the league in OBP, OPS and OPS+ is all pretty impressive.

Forgot about the season records. Thanks, David! Heilmann's plaque needs all the padding it can take.

If all else fails, there's always "Worst Defensive Right Fielder in History".

If all else fails, there's always "Worst Defensive Right Fielder in History".

I think that would be worse than including his nickname on his plaque, wouldn't you? :-)

I think 4 batting titles, 12th best all-time batting average and seven consecutive years in the top 5 in the league in OBP, OPS and OPS+ is all pretty impressive.

...and where did I say it wasn't? My point was that he had zero records (other than club records) at the time of his retirement, no postseason appearances and hardly any league leading performances. Compared to other HoMers, he's deficient in those areas and makes it more difficult to give him a comparable sized plaque. Am I wrong?

(Several of the early HOM plaques are short.)

25 Years Ago: standings in 1911 and 1936

1911 rank (of 40)      1936 rank (of 52)
 1 to 7                PREVIOUSLY ELECTED
( >25% gap in 1911)
 8  Hugh Duffy         18               
 9  Frank Grant        PREVIOUSLY ELECTED  
10  Jimmy Ryan         20               
11  Sam Thompson       PREVIOUSLY ELECTED  
12  George Van Haltren 10               
13  Lip Pike            9                
14  Hughie Jennings    12               
15  Cupid Childs       17               
16  Mike Tiernan       -- (56t last yr)
17  Pete Browning      16               
18  Dickey Pearce      PREVIOUSLY ELECTED  
19  Jim McCormick      51               
20  Bob Caruthers      PREVIOUSLY ELECTED   
( >25% gap in 1911)
21  Mike Griffin       48               
22  Mickey Welch*      15               
23  John McGraw        28               
24  Ed Williamson      39               
25  Tony Mullane       52               
26  Harry Wright       --               
27  Jim Whitney        --               
28  Charley Jones      23               
29  Herman Long        --               
30  Duke Farrell       -- (51t last yr)
31  Jack Clements      --               
32  Tip O'Neill        --               
33  Billy Nash         --               
34  Fred Dunlap        44t              
35  Levi Meyerle       --               
36  Tom York           49t              
37  Tommy Bond         46               
38  Chief Zimmer       --               
38  Bill Hutchison     --               
40  Perry Werden       --              

1911 rank              1936 rank
13  Lip Pike            9                
12  George Van Haltren 10               
14  Hughie Jennings    12               
22  Mickey Welch*      15               
17  Pete Browning      16               
15  Cupid Childs       17               
 8  Hugh Duffy         18               
10  Jimmy Ryan         20               
28  Charley Jones      23               
23  John McGraw        28               
24  Ed Williamson      39               
34  Fred Dunlap        44t              
37  Tommy Bond         46               
21  Mike Griffin       48               
36  Tom York           49t              
19  Jim McCormick      51               
25  Tony Mullane       52

Who ranks higher now than he did 25 years ago?
As I call it: Grant, Thompson, Van Haltren, Pike, Jennings, Browning, Pearce, Caruthers, Welch and Jones.

Pearce, Caruthers and Welch are the big gainers while Tiernan, McCormick and Griffin are the big losers.

(Several of the early HOM plaques are short.)

They had the same problem as Heilmann.

They had the same problem as Heilmann.

Actually, some of the early ones could use a little work. I'll try to "bulk" them up this week.

George Sisler – how can you keep the guy who has the second best single season hit total out of the Hall of Merit?
By considering what he did to help his team win throughout his career, instead of evaluating him based on one not-very-significant statistic he compiled in one year. (The other, other) George Burns had the second best single season double total, and we've managed to keep him out of the HOM. Owen Wilson had the best single season triple total, and I don't see his plaque anywhere.

Club records at the time of retirement include best Slugging percentage (season & career) and most RBI (season), most doubles (season).

OK... I have my Lee Sinin Encyclopedia handy. Tigers career records through 1932:

HOMERUNS HR
1 Harry Heilmann 164

HR% HR%
1 Harry Heilmann 2.25

ISOLATED POWER ISO
1 Harry Heilmann .175

SLG SLG
1 Harry Heilmann .518

Of course, it doesn't take very long for Greenberg to shatter these, but as of 1932 those Tiger records belonged to Heilmann.