Baseball for the Thinking Fan

Login | Register | Feedback

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

Tuesday, March 09, 2004

1922 Ballot Discussion

 WS   W3 Rookie Name-Pos (Died) 
496 176.4 1896  Nap Lajoie-2b (1959)
426 128.8 1901  Christy Mathewson-P (1925)
296  70.3 1903  Mordecai Brown-P (1948)
222  70.8 1904  Miller Huggins-2b (1929)
206  48.3 1905  Ed Reulbach-P (1961)
177  54.3 1907  Nap Rucker-P (1970)
158  30.9 1905  Solly Hofman-CF (1956)
138  34.6 1908  Chief Wilson-RF (1954)
120  31.9 1907  Otto Knabe-2b (1961)
135  28.8 1908  Doc Crandall-P (1951)
114  30.2 1906  Roy Hartzell-RF/3b (1961)
 91  14.4 1902  Red Dooin-C (1952)

Pretty strong entry class this year . . . let me know if there are any Negro League candidates, or anyone else that I may have missed. 1921 results should be up sometime tomorrow.

Joey Numbaz (Scruff) Posted: March 09, 2004 at 08:13 AM | 225 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

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

Page 1 of 3 pages  1 2 3 > 
   1. EricC Posted: March 09, 2004 at 12:28 PM (#522608)
1922/23 prelim

Honus Wagner
   2. Howie Menckel Posted: March 09, 2004 at 01:18 PM (#522609)
I think Brown vs Waddell comparisons would be very useful.
   3. Marc Posted: March 09, 2004 at 01:39 PM (#522610)
I think Brown vs Waddell vs McGinnity vs Plank comparisons would be very useful.
   4. EricC Posted: March 09, 2004 at 01:40 PM (#522611)
I think Brown vs Waddell comparisons would be very useful.

I probably discount the 1903-teens NL and the FL relative to the AL more than anybody, and Brown still comes out on top. I don't think that they'll even be close in the consensus rankings. It may be more interesting to see how Brown (239-130) does compared with McGinnity (246-142) (I assume that the Iron Man hasn't been elected).
   5. Daryn Posted: March 09, 2004 at 01:58 PM (#522612)
Prelim

I agree with EricC -- brown is close to mcginnity, not waddell. I have waddell at 19 or 20 now.

1. Nap Lajoie ? 496 win shares is a good sign to me.
   6. Marc Posted: March 09, 2004 at 02:02 PM (#522613)
I like Eric's list BTW. Since '22 is a "no..." er, well, never mind. But it's not too early to be thinking about the '23-'24 cohorts.

1. So along with the obvious pitcher comparisons--well, maybe that's not so obvious. I'm also very interested how a long career-no peak pitcher like Plank does against the shorter career-higher peak threesome. I agree with Eric, too, re. Brown. He is closer to Walsh than to McG.

2. But also, does anybody see a position player other than Wagner and Lajoie as being close to Crawford. Is Sam a "no..."? Or is there some other position player who could possibly get ahead of him?

3. And finally, if indeed about 7 or 8 of our current backlog look poised for election by 1932, I'll ask Eric, e.g., why Bresnahan, Joss, Wallace, Pike et al. Are you sure they're HoMers?
   7. EricC Posted: March 09, 2004 at 02:12 PM (#522614)
For further comparison:

Brown: 3172.3 IP, 239-130; 138 ERA+
   8. Howie Menckel Posted: March 09, 2004 at 02:18 PM (#522616)
I tend to agree that Three-Finger does rate just ahead of McGinnity (which stinks, I want McGinnity in ASAP), and well ahead of Waddell. But looking for a Waddell-backer to differ..
   9. Marc Posted: March 09, 2004 at 02:21 PM (#522617)
Hope you guys don't mine me re-posting this from the 1921 ballot thread, I got tired of scrolling. Note that I missed Pete Hill and Louis Santop, who may change things considerably. Like I said, #3-8/9/10 in the 1921 results will be the favorites to get into the HoM by 1932, so I like this post as a reminder how important those 7-8-9-10-11 spots can be.
   10. Rusty Priske Posted: March 09, 2004 at 02:22 PM (#522618)
Prelim

1. Nap Lajoie
   11. EricC Posted: March 09, 2004 at 02:28 PM (#522620)
I'll ask Eric, e.g., why Bresnahan, Joss, Wallace, Pike et al. Are you sure they're HoMers?

Plank and above are n*- b*. I don't think that anybody else on my list would be a bad addition to the HoM, though the candidates get more and more marginal as the list goes on.

On the subject of the backlog- I've tried looking ahead to see how deep on my current ballot I might have to go to construct my early-30's ballot. It's possible that a candidate who is currently as low as 28th could make my ballot by then!
   12. Howie Menckel Posted: March 09, 2004 at 02:38 PM (#522621)
Clint,
   13. karlmagnus Posted: March 09, 2004 at 03:12 PM (#522622)
According to Bill James, ERA as a stat was only invented in 1913. Therefore I think we're being grossly ahistorical in judging pitchers as if they were pitching to ERA or ERA+ when they weren't; they were pitching to get WINS, and coasting in the late innings of laughers, a role in which today the #11/12 man in the bullpen would be brought in. It is quite wriong therefore to criticise Welch for an ERA+ of only 113, and deify Walsh for his ERA+ of 146 (except, I suppose, for Walsh's last 2 years.) It was wins that mattered to both men, and Welch got 50% more WINS than Walsh, a fact that I'm sure he frequently pointed out to young whippersnappers in his long and healthy retirement.

When people say Caruthers was "Only half" a pitcher like Walsh they are in cuckoo land; he pitched only 138 less innings, and got 25 more wins, the stat by which both he and Walsh measured themselves during their careers. Of course, Caruthers also hit a bit, a fact to which I shall return in a later post!
   14. MattB Posted: March 09, 2004 at 03:21 PM (#522623)
I think Brown vs Waddell comparisons would be very useful.

My stat of choice is Wins Per Finger

Brown = 239/3 = 79.7 WPF

Waddell = 193/5 = 38.6 WPF

On a WPF basis, Brown is more than TWICE as valuable and Waddell.
   15. Chris Cobb Posted: March 09, 2004 at 03:35 PM (#522624)
No prelim ballot yet, but a few words on Mordecai Brown.

1) For anyone who cares at all about peak value, McGinnity must rank well above Brown. Brown's 3172 innings are spread over 12 full seasons, plust a couple of scraps. McGinnity, although his career was shorter in terms of seasons, pitched a full season's more innings (3458.3 for his career) and concentrated them into 9.5 years. Compared to this difference, the ERA+ distinction (even if you trust ERA+) is trivial. Walsh was valued highly, and rightly so, by the electorate for the amount of innings he threw. McGinnity, it should be remembered, was more of a workhorse than Walsh, though not quite as effective on a per-inning basis.

2) Brown's ERA+ overrates him substantially, in my opinion. In addition to extending his career by pitching in the weaker FL, he also had the advantage of pitching in front of one of the great defenses of all time, which helped his era.

3) Brown also had the privilege of pitching for outstanding teams for most of his career. This is true also for McGinnity, but it's less true of, say, Clark Griffith.

Griffith 237-146, 3385.7 IP, in 11 full seasons, two part-seasons, and some scraps. He won 20+ games 7 times, including six seasons consecutive.
   16. Marc Posted: March 09, 2004 at 03:41 PM (#522625)
>According to Bill James, ERA as a stat was only invented in 1913. Therefore I think we're being grossly
   17. Marc Posted: March 09, 2004 at 03:47 PM (#522626)
And as for the "pitching in a pinch" argument, even if it is correct, which we really don't know for sure, but even if it is, I don't see how one is supposed to actually apply that insight. E.g.

> Brown: 3172.3 IP, 239-130; 138 ERA+
   18. karlmagnus Posted: March 09, 2004 at 03:50 PM (#522627)
In trying to win a ballgame, you are automatically increasing your WS and doubtless your WARP. That's not true with ERA; the "pitching in a pinch" philosophy is designed to maximize wins, not ERA+. The general point is also true, that one should take sabermetric stats with a big pinch of salt when interpreting an era against which they weren't designed, but ERA is a particular case where a pitcher may have been (and generally was) using strategies that didn't maxinmise it.

Another example is stolen bases. Sabermetrics tells us that the value of these is grossly overrated, and that attempting stealing except at very high percentage success rates is counterproductive. BY all means, let's judge today' players, who (or whose managers) should be expected to know this, by this standard. But it would be most unfair to dismiss altogether the value of Ty Cobb or Sliding Billy Hamilton on the grounds that they should not have been stealing bases -- they didn't know that, and their game was designed to do supremely well something that they thought was valuable to the team.
   19. Marc Posted: March 09, 2004 at 03:55 PM (#522628)
karl, I understand your point about anachronisms and all. I made the same argument about Ross Barnes.

But the problem is what to do about it. SB are discretionary, and more to the point we can isolate SB and CS and evaluate Hamilton and Cobb on all their other virtues. Throwing the ball over the plate is not discretionary for a pitcher, and we have no data to suggest if and when a player was coasting and when he was pitching.

So again how do we know that "pitching in a pinch" moves Welch or McG or any other pitcher up or down the ballot?
   20. karlmagnus Posted: March 09, 2004 at 03:58 PM (#522629)
Before somebody points out that Hamilton's already in the HOM and Cobb is sure to be, let me suggest thinking instead of a borderline candidate, a Sheckard/Wallace/Ryan type. If such a player also had the all time record for stolen bases, as Ty Cobb had until Brock broke it in 1974, this would in my view be a powerful argument in favor of enshrinement, even if sabermetrics tells us that his stolen base prowess was pointless.
   21. User unknown in local recipient table (Craig B) Posted: March 09, 2004 at 04:15 PM (#522631)
let me suggest thinking instead of a borderline candidate, a Sheckard/Wallace/Ryan type. If such a player also had the all time record for stolen bases, as Ty Cobb had until Brock broke it in 1974, this would in my view be a powerful argument in favor of enshrinement, even if sabermetrics tells us that his stolen base prowess was pointless.

karlmangus, I agree with this 100%.
   22. Chris Cobb Posted: March 09, 2004 at 05:13 PM (#522633)
Marc asked:
   23. karlmagnus Posted: March 09, 2004 at 05:32 PM (#522634)
Chris, clearly Welch's "pitching in a pinch" moments will be reflected in a lower ERA, and higher ERA+, as you have ably demonstrated. However, his overall ERA+ also reflects his pitching when not "in a pinch" at which point he will tend to give away runs, increasing his EA and reducing his ERA+, while not losing many ballgames (presumably there's the occasonal 8-2 laugher that gets away when he loses it and has a 7 run inning.) This is why Welch and Caruthers' W/L records are better than their ERA+ would indicate; it is the W/L records (suitably adjusted for team quality) and not the ERA+ that are the true indicator of their ability and Merit.
   24. karlmagnus Posted: March 09, 2004 at 05:33 PM (#522635)
Chris, clearly Welch's "pitching in a pinch" moments will be reflected in a lower ERA, and higher ERA+, as you have ably demonstrated. However, his overall ERA+ also reflects his pitching when not "in a pinch" at which point he will tend to give away runs, increasing his EA and reducing his ERA+, while not losing many ballgames (presumably there's the occasonal 8-2 laugher that gets away when he loses it and has a 7 run inning.) This is why Welch and Caruthers' W/L records are better than their ERA+ would indicate; it is the W/L records (suitably adjusted for team quality) and not the ERA+ that are the true indicator of their ability and Merit.
   25. John (You Can Call Me Grandma) Murphy Posted: March 09, 2004 at 05:44 PM (#522636)
Prelim:

1) Christy Mathewson
   26. KJOK Posted: March 09, 2004 at 05:45 PM (#522637)
Looks like it's time to compare pitchers:

Innings:
   27. OCF Posted: March 09, 2004 at 06:37 PM (#522639)
Two notes on Chief Wilson. Since it's very unlikely that he'll get any votes at all, this is our only chance to look at him.

The first note: how could it have happened? I'm referring, of course, to his 36 triples in 1912. This is the most freakishly out of line of all single season offensive records. Wilson's XBH line that year was 19-36-11.

How good was Wilson's year? Evaluating it by my current favorite gimmick, it comes out as 30 RC above average scaled to a 5 R/G context. That's a good year, to be sure, but Jake Beckley did better than that 6 times in his "no-peak-at-all" career.

1912 was the peak of the lively-ball bubble in the NL, a year later than the peak in the AL. Wilson was certainly a power hitter. He was in the top 5 in the league in HR each year from 1911 through 1914, and he was among the league leaders in triples in some other years. Was he fast? Not so much evidence of that. He was only credited with 98 SB in a 9 year career, which isn't great for a deadball outfielder, and the one year that CS are recorded, he was 8-15. He was used some in CF, including 87 games in 1912, but after that, the Pirates sent him back to RF and rotated a half-dozen other players through CF. Forbes Field was notorious for requiring the services of two centerfielders - in CF and LF. And they moved Wilson back to RF. Whatever else that means, it doesn't suggest exceptional speed. Was it a year of unusual aggressiveness for Wilson? If so, he wasn't getting caught, since a failed attempt to stretch a double into a triple would still be recorded as a double.

Forbes Field was presumably a good triples park. The Pirates led the league (barely) with 106 triples in 1911. They were 3rd in league, 10 behind the leader, with 86 triples in 1913. In 1912, they had 129. If Wilson had had 14 instead of 36, the team would still have led the league by 17. So maybe it was something about the field that year. But why did Wilson benefit so much more than anyone else? The vast open spaces in Forbes Field were in center and left center, so it's not immediately obvious how Wilson, a left-handed hitter, took advantage.

So in the end, I don't understand it.

--

Wilson is listed as "Chief" Wilson. I put "Chief" into bbref to see what came up, and got 11 names. In chronogical order of birth:

James Roseman - born 1856, New York, NY
   28. Marc Posted: March 09, 2004 at 06:39 PM (#522640)
>Posted 11:49 a.m., March 9, 2004 (#31) - David Foss
   29. Chris Cobb Posted: March 09, 2004 at 06:47 PM (#522641)
KJOK, could you explain the methodology behind the "support-neutral fibonacci wins" number? I have no idea what that metric means.

Also, thanks for using "shoo-in"!

As to Mullane over McGinnity: I'm afraid I don't see it. IP prior to 1893 can't be compared on a one-to-one basis with IP after 1893, and their seasonal distribution needs to be considered as well. In his context, McGinnity was a tremendous workhorse; Mullane was not. He has a long career, but not an outstanding one. Most of Mullane's best pitching was done in the AA, and his peak years came when the AA was a decidedly inferior league and needs a 15%-20% discount.

Among eligible 1880s pitchers, I see Welch over Mullane, and McCormick over Welch. Considered only as a pitcher, Caruthers lags behind them all in career, and tops only Mullane on peak, but he has an outfield career alongside his pitching.

How does McCormick look by the measures you're presenting?

Incidentally, I think that now is an excellent time to be looking at pitchers. We will shortly be expanding our HoM pitching cohort from the aughts to match the number of pitchers inducted from the 1880s and to exceed the number inducted from the 1890s. The lean years of the late 1920s will be a time in which it will be possible to add some trailing pitchers, so we should be a sure as we can that we're getting the right ones. I still don't think we have a firm handle (does anybody?) on comparing pitchers from the different eras we're looking at, so I look forward to more discussion.
   30. KJOK Posted: March 09, 2004 at 07:01 PM (#522642)
KJOK, could you explain the methodology behind the "support-neutral fibonacci wins" number? I have no idea what that metric means.
   31. OCF Posted: March 09, 2004 at 07:02 PM (#522643)
Of Chris Cobb's several points in #17, the one I take most seriously is the issue of defensive support. Brown pitched in front of a great defense, and in a system that didn't abuse pitchers (although they got more work out of Brown than their other pitchers.)

As for peak: if you take RA+ at face value, and convert each year into equivalent FWP, McGinnity's 5 best years are 50, 42, 33, 21, 15. Brown's 5 best years are 41, 40, 39, 29, 24. McGinnity's career RA+ equivalent record is 230-152. Brown's is 232-121. If you remove the two FL seasons from that, it was 202-100 in the NL. I have Griffith at 209-138.

But of course, RA+ depends on defensive support.

Mathewson will be at #2 on my ballot (with Lajoie #1). McGinnity was #3 on my ballot last year, so he's my highest ranking returnee. That's one of this year's big issues - to place Brown and McGinnity against each other. (The other big issue is to find the right level for Grant Johnson.) Put me down as not yet decided.
   32. Howie Menckel Posted: March 09, 2004 at 07:05 PM (#522644)
Marc,
   33. KJOK Posted: March 09, 2004 at 07:13 PM (#522645)
As to Mullane over McGinnity: I'm afraid I don't see it. IP prior to 1893 can't be compared on a one-to-one basis with IP after 1893, and their seasonal distribution needs to be considered as well. In his context, McGinnity was a tremendous workhorse; Mullane was not. He has a long career, but not an outstanding one. Most of Mullane's best pitching was done in the AA, and his peak years came when the AA was a decidedly inferior league and needs a 15%-20% discount.

Just a couple of points:
   34. ronw Posted: March 09, 2004 at 07:19 PM (#522646)
1922 Prelim

1. Lajoie
   35. KJOK Posted: March 09, 2004 at 07:19 PM (#522647)
How does McCormick look by the measures you're presenting?

If you're interested, the list of Supt Neut Fib Wins has been available in the FILES section of the HOM egroup for awhile now.

McCormick pitched 4,276 innings, his RSAA is 190, his Neut Fib Wins are 229, and his ERA+ is 118. He looks better than Welch to me, but not as good as Griffith or Waddell.
   36. Devin has a deep burning passion for fuzzy socks Posted: March 09, 2004 at 07:26 PM (#522648)
KJOK, since I have my statistical information at home, could you check where Jim McCormick comes in on the lists you're referencing?
   37. Devin has a deep burning passion for fuzzy socks Posted: March 09, 2004 at 07:27 PM (#522649)
Sorry about that, I was skimming...
   38. Marc Posted: March 09, 2004 at 07:29 PM (#522650)
Howie, my point is not that guys don't coast or whatever. My question is how much? How often? To what effect? And to David's point, without a play-by-play, how do you know who to give X-credit to and who not?

You infer that McGinnity...but let me start over.

This all started with yest's plea on behalf of Mickey Welch whose record alongside that of long-time teammate Tim Keefe looks very odd. Similar W-L records, but by every other measure Keefe was a vastly (or, at least, a consistently) more effective pitcher. Why?

There are two hypotheses on the table that I'm aware of. 1) pitching in a pinch, and 2) random chance.

My position is that both are worthy as hypotheses, both are worthy of further testing. But without play-by-play, how can you test them? As an advocate for 1, that is your job to figure it out.

In the absence of data to support one or the other, my position further is that 2 is vastly more likely. We KNOW that all pitchers (and players) are subject to random/small sample chance all the time, every inning of every game. We KNOW that some pitchers coasted some of the time, but we don't know when. So since the differences between Welch and Keefe are specifically and exactly measured one game at a time, we KNOW that chance was involved in every single decision. We don't know how often pitching in a pinch was involved but we KNOW that it was less often than chance.

So without further data, I would say the discrepancies are 90 percent chance and 10 percent residing in something the pitchers did that the numbers we've seen so far can't explain.

Ditto Brown and McGinnity. 90 percent random.
   39. Chris Cobb Posted: March 09, 2004 at 07:51 PM (#522651)
KJOK,

Thanks for the explanation of support-neutral Fibonnacci wins. Another measure to play around with . . .

1. I think it's a bit of a fallacy that it was MUCH easier to pile up innings pre-1893. Pitchers could pitch more innings per year pre-1893, but they had shorter careers.

Is there data to support this? If so, that would influence my view, but the seasonal career lengths of the top pitchers whom we're assessing don't seem significantly different when one compares pre-1893 pitchers to post-1893 pitchers.

2. McGinnity was a workhorse for a few years, but Mullane pitched more innings per year AND pitched a greater % of his team's innings, so it's not clear McGinnity should be accorded a big edge in value.

Given the state of the ballot now, McGinnity wouldn't need a _big_ edge in value to rank significantly ahead of Mullane. That said, Mullane certainly did pitch a greater percentage of his team's innings, which has significant value. But the starting pitchers of his era all did this. I've been basing my comparison of pitchers between eras on the extent to which they exceed the average value of a pitcher in their era. One part of my value assessment compares a pitcher's workload to a benchmark of "average innings pitched" in a season. This normalizes era-specific workload ranges to a uniform standard. One might argue that this measure is unfair to early pitchers, because it ignores the fact that they had, on average, greater value relative to position players than later pitchers would. But one could argue that to do it the other way round is unfair to later pitchers, whose average value relative to position players was set by a different, more physically demanding, set of playing conditions. How best to balance these two different ways of looking at value?

3. It's true Mullane pitched in AA, but your discount seems extremely high to me. The AA was considered a "major" league at the time, and their pennants were viewed as just as legitimate as the NL's.

The information that I've accepted indicates that the AA was much lower in quality than the NL in its early years, rising much closer to parity 1885-1889. Mullane's three biggest years are the first three years of the AA, and I discount those 22.5%, 15%, and 10%, respectively. My discounts for 1885-1889 are 7, 6, 7, 4.5, and 6 %, respectively. I'm not a maven on league strength, but I'm persuaded that the folks who know more than I do about this and who argue for substantial discounts for the early AA are correct. (It's been a while since I've gone round on this question, but I believe jimd has been the best explicator of discounts derived from Davenport translations. jimd, would you care to comment?)
   40. karlmagnus Posted: March 09, 2004 at 07:54 PM (#522652)
Chance of random fades to insignificance when you're considering the matter over the pitcher's entire career; it's not the same as Ryan's 8-16 in 1987.
   41. OCF Posted: March 09, 2004 at 07:55 PM (#522653)
I'm guessing from KJOK's explanation of "support neutral Fibonacci wins" in #35 that he's using ERA+ to get an expected winning percentage. I've been doing something quite similar, with three differences. (1) I use RA+ instead of ERA+. (2) I use IP/9 instead of W+L. (3) I have concentrated on post-1893 (mostly post-1900) pitchers, and have no data on the likes of Mullane and Welch. In career FWP, the eligibles and elected come out as:
   42. Jim Sp Posted: March 09, 2004 at 07:55 PM (#522654)
1) Lajoie--Best position player we?ve seen on the ballot yet.
   43. karlmagnus Posted: March 09, 2004 at 07:58 PM (#522655)
Support-neutral Fibonnaci wins doesn't work when like Caruthers you're a substantial part of the support. If Barry Bonds pitched, he'd win a lot of games, even if his ERA was mediocre.
   44. karlmagnus Posted: March 09, 2004 at 08:07 PM (#522656)
If you're using ERA+ to derive "support-neutral Fibonnaci wins" it's a circular argument, and the stat should be thrown out with the trash. SNFW are then not wins, they're another form of ERA+, a highly dubious stat before 1913.

Now let's start the real argument!

The top of my preliminary 1922 ballot:

1. Parisian Bob
   45. Howie Menckel Posted: March 09, 2004 at 08:27 PM (#522657)
Pretty good, Marc, but you're still telling McGinnity's ghost that he isn't as effective as he thought he was - because of stats he had never heard of.
   46. Al Peterson Posted: March 09, 2004 at 09:03 PM (#522658)
karl, you're going to have to help me with something here...

Hitter-Caruthers had 1,047 hits (Caruthers, normalized to 130 game years) times 3, or 3,141 hits, compared to Nap?s 3,242, with 13,133 PA compared to Nap?s 10.460. He also had 1885 BB, compared to Nap?s 516 -- a MUCH more patient hitter, albeit with less power. (Incidentally, most 1880s stars had few BB, because except in 1887 -- 4 strikes-- the rule required more balls to get one. Caruthers thus had EXTRAORDINARY strike zone judgment.) Caruthers' (TB+BB)/PA and (TB+BB)/Outs were -- .483 and .793, both significantly higher than Nap?s .477/.757.

So extrapolating takes Freedom Bob -The Hitter to 3141 hits and 1885 BBs? He's going to do that in 10 years of major league play? Maybe his name should be Superhero Bob.

I can see you trying to get a handle of the combination value for Caruthers but I'm not understanding this way right now.
   47. karlmagnus Posted: March 09, 2004 at 09:14 PM (#522659)
Since we're working with new statistics here, I propose instead of "Fibonnaci wins" that we look at "Gaussean ERA+." This is defined as the ERA+ a pitcher would have had, given his W/L record, if he'd been bearing down all the time and not "pitching in a pinch." (note that for some pitchers, like Walsh and Waddell, Gaussean ERA+ is less than ERA+, indicating that they were very ineffective in determining when the "pinch" was.) It is defined as the Runs For/Runs Against Pythagorean equivalent of the pitcher's W/L.

For some of the pitcher's we've recently been discussing, Gaussean ERA+ was as follows:

W L Gaussean ERA+

Caruthers 218 99 1.54
   48. karlmagnus Posted: March 09, 2004 at 09:19 PM (#522660)
Al, Pitcher-Caruthers and Hitter-Caruthers are lay figures, to get all Caruthers' value in the form either of picthing or of hitting. However, since Caruthers was only 30 when he retired, we can certainly imagine that Hitter-Caruthers would not have had a sore arm from pitching, and so would have played till a normal age, retiring at about age 36, in 1900.
   49. karlmagnus Posted: March 09, 2004 at 09:22 PM (#522661)
Al, Pitcher-Caruthers and Hitter-Caruthers are lay figures, to get all Caruthers' value in the form either of picthing or of hitting. However, since Caruthers was only 30 when he retired, we can certainly imagine that Hitter-Caruthers would not have had a sore arm from pitching, and so would have played till a normal age, retiring at about age 36, in 1900.
   50. Al Peterson Posted: March 09, 2004 at 09:41 PM (#522662)
karl,

The problem with giving Caruthers credit hitting til age 36 is maybe then we should play the what-if game with more players. Three Finger didn't pitch in the bigs til 26, Iron Joe age 28. Fielder Jones leaves at age 36 after arguably his best year. Clark Griffith cuts back on pitching to do his managing. Same thing with Bresnahan. Do we do adjustments for all these players? Its a slippery slope.
   51. John (You Can Call Me Grandma) Murphy Posted: March 09, 2004 at 10:02 PM (#522664)
Karl:

I'm trying to understand how a .282 hitter like Caruthers would wind up with almost the same amount of hits as Lajoie in four less seasons. Am I missing something?
   52. karlmagnus Posted: March 09, 2004 at 10:22 PM (#522665)
Hitter-Caruthers had 13,000 at-bats, which means that translating Caruthers' value entirely into a hitter (AND scaling up into 130 game seasons) is somewhat unrealistic. But Nap didn't walk, Caruthers and Hitter-Caruthers did; H-C had 1800 BB. The .282BA isn't a problem, though; I give you Ripken, Yaz and Yount, all over 3000 hits with BA thereabouts or below.
   53. karlmagnus Posted: March 09, 2004 at 10:31 PM (#522666)
Sorry, 13,133 PA, which is somewhat more realistic.
   54. karlmagnus Posted: March 09, 2004 at 10:38 PM (#522667)
Further thought. Caruthers, pitching and hitting, crammed a huge and artificial amount of value into his best seasons. Regular pitchers and hitters couldn't do that -- even Old Hoss couldn't go 60-19, and they won't let you take 1,000PA in a season. Hence, to have the same total value as Caruthers, Hitter-Caruthers and Pitcher-Caruthers would have had to have longer careers, in terms of years.
   55. KJOK Posted: March 09, 2004 at 11:06 PM (#522669)
This whole "pitching in a pinch" argument seems rather spurious.

A pitcher's job is to prevent runs, whether or not ERA was an "official statistic". It's a TEAM's job to win the game, not the pitchers job (unless he's Bob Caruthers, but that's a different argument...)

At the beginning of the game, a pitcher does not know how many runs he can allow in order to win the game, SO he SHOULD be trying to pitch a shutout.

In a few selected games, a team may get out to a very large lead. In these games, it's POSSIBLE that at pitcher will "let up" and let in a few more runs that he would otherwise if the game were close.

The problem is that:

1. There are not enough of those games over the course of a year to make a significant difference in runs allowed, at least vs. all the other pitchers in the league who would be doing the same thing when they had leads.

2. There doesn't seem to be any statistical evidence to support the theory. If you look at the "support neutral wins" vs. actual wins for 19th century pitchers, some exceed actual wins, and some fall short, both in around equal numbers and magnitude. There are a few outliers. Just what you would expect from random chance. If this theory were true, then when you calculated "pythagorean" wins and losses for teams, you would expect good teams to fall short of their pythagorean projections and bad teams to exceed MORE than what happens for the 20th century, and that doesn't appear to be happening either....
   56. MattB Posted: March 09, 2004 at 11:16 PM (#522670)
Chris Cobb wrote:

<i> As to Mullane over McGinnity: I'm afraid I don't see it. IP prior to 1893 can't be compared on a one-to-one basis with IP after 1893, and their seasonal distribution needs to be considered as well.
   57. KJOK Posted: March 09, 2004 at 11:40 PM (#522672)
MattB - Thanks, you posted before I could get my post in, but here are the numbers I have for 2 18 year periods:

<pre>
   58. jimd Posted: March 10, 2004 at 02:14 AM (#522674)
I'm curious. Does anyone know what the contemporary opinion on the early AA quality was? Both inside the partisan AA cities, and in non-partisan cities (1882 New York had no team in either league.)

Anyway, here's a study comparing the regulars in the 1882 AA and the regulars in the 1882 NL, broken down by team. (I hope the formatting can be figured out.)

Columns:
   59. jimd Posted: March 10, 2004 at 02:26 AM (#522675)
Oh, well. The "curly braces" got turned into parentheses...
   60. Marc Posted: March 10, 2004 at 02:39 AM (#522676)
>ERA+, a highly dubious stat
   61. Marc Posted: March 10, 2004 at 02:42 AM (#522677)
Actually wouldn't ERA+ be dubious until, what?, about 1988?
   62. Marc Posted: March 10, 2004 at 03:00 AM (#522678)
Mark, re. HR Johnson. You're on the right track. Taken at face value HR Johnson is in (OK not quite in) Lajoie-Wagner territory. More like Gehringer-Arky Vaughn territory. At face value. Kinda like Tony Mullane and Joe McGinnity.

But I think that Mullane's league, the AA, was weaker than the NL, not in '86 but otherwise, by anywhere from 3-5% to 30-35% depending on the year. Take the appropriate discounts year by year and it averages out to about 8%. Enough that Mullane drops down pretty decisively compared to the McGs, Waddells, Browns et al.

So with HR Johnson, the question is what discount he gets. Surely he gets one. 3-5%? 30-35%? A UA-like 65%? Well, I don't know. I would bet that replacement level was 30-60% below the AL-NL, exspecially 1900-20, but the "greats" are all outliers, so who says Jim Thorpe couldn't be just as good as Red Grange, maybe better. So I don't know the answer, but I know the question. Everybody's got to decide if and (if yes, then) how much.
   63. jimd Posted: March 10, 2004 at 03:06 AM (#522679)
Is ERA also dubious? Did pitchers stop "pitching in a pinch" in 1913?

Measuring something will often affect what's measured. (Check out how the Save rule transformed relief pitching.) So ERA may be more valid before 1913.

Except that, IIRC, 1913 is when ERA becomes an official statistic. It probably was used before that, informally. The counts of earned-runs that are used to calculate ERA from say the 1880's have to come from somewhere, like scoresheets from that time. If they hadn't been compiled at game-time, it would require a Retro-sheet like effort to recompile them years later from play-by-play data. Maybe MacMillan/HOF did that and aren't sharing; I don't know.

I think ERA/ERA+ is dubious, but for other reasons (defensive effects).

The rise of the home run made "pinching in a pinch" obsolescent. Unless you had a big lead, you had to bear down against the home-run hitters, and when that became practically everybody, there was nobody left to coast against.
   64. Howie Menckel Posted: March 10, 2004 at 03:44 AM (#522680)
Good post, Mark (#66)..
   65. Jim Sp Posted: March 10, 2004 at 05:07 AM (#522681)
Was "pinching in a pinch" about coasting with a lead, or pacing yourself to save your best stuff when there are runners on base?

I suspect that pitchers have always been in the business of preventing runs, the home run era just meant a run could happen at any time.

Otherwise wouldn't "pinching in a pinch" have been out of vogue in all high-offense leagues such as the 1894 NL? Or wouldn't good teams today still ease up with a decent size lead to the same degree as back then?

Something doesn't add up here.

To just say that pitchers back then only cared about wins, not runs...consider Bob Caruthers at his peak, he's presumable a cocky young guy on top of the world in St. Louis. Is he going to let people score runs off him voluntarilly in the 5th inning of a 3-0 ballgame?

I doubt it...but perhaps someone can provide evidence that these voluntary runs actually existed.

And that the pitchers we're comparing him to didn't do this to the the same extent.

And that the effect is large enough to bump him way up on the ballot.

Then I'll put Caruthers ahead of Tony Mullane, but probably not Christy Mathewson. Who after all did write the book on "pinching in a pinch".
   66. OCF Posted: March 10, 2004 at 05:47 AM (#522682)
I tried an exercise with the RA+-equivalent record system. The idea is this: pitching in front of a great defense has effects reasonably similar to pitching in a pitcher's park. The runs will come out just a little bit lower than in a neutral environment. There might be a way to do this systematically, with a great deal of effort, but I didn't try. Instead, I did something incredibly simple-minded: a lowered the park factors for each of Brown's 1903-1913 years by 5 points. And while I was at it lowered the Federal League park factors by 10 points. Was the value of the Cubs' defense worth 5% of a park factor? I have no idea if that's even the right order of magnitude.

Without that adjustment, I have Brown as the equivalent of 232-121, for 262 FWP. If you back-compute that winning percentage into the equivalent of a RA+, you get 138.

With the adjustment, Brown's record comes down to 223-132, for 237 FWP, and a back-computed RA+ of 131.

All this is compared to McGinnity's 230-152, 217, 123.

If you push the defense-caused "park factor" for Brown down by 10 points instead of 5, he comes down to 216-138, 210, 125, which is just slightly below McGinnity.

How much of that is enough? I'm starting to lean toward the view that Brown may have had the advantages, but he was also a great pitcher - and deserving of the highest non-Matty slot among pitchers on my ballot.

If you want to make a bigger downward adjustment to Brown, then that value can't just disappear. It must be added to someone else's value, and the most prominent names under "someone else" are Tinker, Evers, and Chance.
   67. Howie Menckel Posted: March 10, 2004 at 01:39 PM (#522683)
Jim Sp,
   68. Al Peterson Posted: March 10, 2004 at 02:48 PM (#522684)
Since he won't get a vote I'll play trivia guy for Mr. Nap Rucker. Three things of note:

1. His ERA+ of 119 bought him a career mark of 134-134.
   69. MattB Posted: March 10, 2004 at 04:01 PM (#522685)
The Native American baseball player who will likely get the most support for the HoM will likely be Bob Johnson, a Cherokee who played for the Philadelphia A's in the 1930s and 40s.

Chief Meyers and Allie Reynolds might get some votes also.
   70. Howie Menckel Posted: March 10, 2004 at 04:35 PM (#522686)
some Three-Finger Brown/Mathewson info:

"The peak years of Brown's career coincided with those of Christy Mathewson, and they were often matched when the Giants and Cubs met. One game he lost to Mathewson was Matty's no-hitter in 1905. After that, Brown rolled off nine consecutive victories over Mathewson, the ninth coming in the playoff that decided the famous 1908 pennant race after the "Merkle Boner." In 1916, they faced each other for the final time, each with 12 wins. Mathewson beat Brown, in what turned out to be the last game for each."

baseballlibrary.com
   71. Jim Sp Posted: March 10, 2004 at 06:49 PM (#522689)
Howie,
   72. Chris Cobb Posted: March 10, 2004 at 07:03 PM (#522690)
OCF wrote:
   73. OCF Posted: March 10, 2004 at 07:22 PM (#522691)
If you accept the "orthodox" DIPS position that hits on balls in play are not significantly influenced by the pitcher, you could pro-rate these numbers for the two pitchers' innings, apply them to the pitchers' RA totals, and pythag from there.

I see something I'm going to try - but it probably won't happen this week. Look for my reply, perhaps on the 1923 ballot discussion.
   74. Howie Menckel Posted: March 10, 2004 at 07:36 PM (#522692)
Jim Sp,
   75. Howie Menckel Posted: March 10, 2004 at 07:44 PM (#522694)
Career votes-points leaders thru 1921, based on OCF final numbers.

Congrats to Charlie, who sets a standard for all others to shoot for! Thompson is 7-8 years away, at his current pace.

still 'active' in CAPS, number of elections in parentheses
   76. Howie Menckel Posted: March 10, 2004 at 07:46 PM (#522695)
Why, yes, I DID call myself "Bennett" in the first listing. sigh.
   77. jimd Posted: March 10, 2004 at 07:48 PM (#522696)
Was the value of the Cubs' defense worth 5% of a park factor? I have no idea if that's even the right order of magnitude.

BP's estimate (using DERA/NRA (which are "Defense-adjusted ERA" and "Normalized Runs Allowed" (Runs Allowed adjusted to a 4.50 league; DERA then adjusts that for defense))) for Miner is 10%. OTOH, it's estimate for Iron-Man is 4%. So using 5% for the comparison of the two should be pretty close.
   78. Marc Posted: March 10, 2004 at 07:56 PM (#522697)
Thx, Howie. You and OCF have given us plenty of navel-gazing fun.

If there are Primeys for the HoM threads, I'd suggest two. The players on Howie's list are surely among the more controversial selections. Among all of those players, I remember two pieces of research that broke the logjam, so to speak.

One was the research on Stovey's unusual ability to score runs. Was that OCF??? We're still debating Harry if not for that information.

And for Bennett, it was jimd's work clarifying and documenting Charlie's workload in terms that were comparable over time, despite the vastly different season-lengths.

Primeys!

And also for Howie's ongoing scorekeeping and OCF's most-similar ballot work.
   79. OCF Posted: March 10, 2004 at 08:00 PM (#522698)
Musing on "Howie Bennett's" post:

I first cast a ballot in 1904. Charlie Bennett is the only player who appeared on all 18 of my ballots. His ballot ranks for me, starting with 1904:
   80. Al Peterson Posted: March 10, 2004 at 08:48 PM (#522699)
Quick and dirty metrics comparison for Three Finger Brown and Bob Caruthers:

In their prime...
   81. karlmagnus Posted: March 10, 2004 at 09:28 PM (#522700)
Thanks Al, interesting comparison. Mordecai Peter Centennial will also be high on my ballot, to make my personal HOM in '24. In reality, WARP3 in my view has the comparison about right (even Enron's accounting must have been right sometimes!)

By the way, on the Ballot thread; my listing for 3 off/3 on is wrong. Caruthers isn't off my ballot, he's #1. Waddell is the third top-ranker off my ballot (and quite a long way off, unlike Bennett, who would have made it back on by '25.)
   82. KJOK Posted: March 11, 2004 at 01:01 AM (#522701)
<i>Congrats to Charlie, who sets a standard for all others to shoot for! Thompson is 7-8 years away, at his current pace.

still 'active' in CAPS, number of elections in parentheses
   83. KJOK Posted: March 11, 2004 at 01:06 AM (#522702)
OK, nevermind, I checked the previous year's ballot post. For some reason, I thought I had seen McGinnity's name as a winner somewhere.....
   84. Howie Menckel Posted: March 11, 2004 at 01:30 AM (#522703)
McGinnity actually was ahead in the HOM-voting equivalent of the 7th inning, but Grady Little elected to stick with him.
   85. Marc Posted: March 11, 2004 at 02:31 AM (#522704)
Top primes--I calc. primes for adjWARP1 and adjWS (adj for season length), with different methods for length of prime based on the quirks of each system. This is prime only. My ballot factors peak and career as well.

1. Lajoie--WARP 15 years-199.8-13.3/year; WS 15-440-29

2. Mathewson--WARP 13-143.1-11; WS 14-417-30

3. HR Johnson--subjective

4. Pearce--subjective

5. H. Wright--ditto

6. Wallace--WARP 14-134.4-9.7; WS 14-311-22

7. Thompson--WARP 10-105-10.5; WS 10-279-28

8. Caruthers--WARP 5-73.6-14.7; WS 8-382-41. Short but sweet.

9. Sheckard--WARP 14-129.1-9.2; WS 14-337-24

10. Williamson--WARP 10-103-10.3; WS 10-331-33

11. C. Jones--WARP 8-84.5-10.6; WS 9-282-31

12. Pike--this is partially subjective. I don't quite take his '71-'75 and double it, more like about 1.5.

13. Dunlap--WARP 6-84.2-14.0; WS 7-195-28. This is WITH a 65% UA discount.

14. Van Haltren--WARP 14-125-8.9; WS 15-414-28

15. Brown--WARP 9-67.6-7.5; WS 9-228-25

Close--Ryan, Grant (subjective). Nobody else is that close. Beckley has the longest but at a pretty low rate.
   86. Chris Cobb Posted: March 11, 2004 at 06:02 AM (#522705)
1922 Preliminary Ballot

Notes on new candidates, candidates whom I'm shifting significantly this year, and candidates whose ranking is harder for me to figure.

Shoo-Ins. Will be elected this year, virtually without dissent.

1. Nap Lajoie (n/e). 3rd-best player yet eligible. 560 CWS. Total peak 161. Peak rate 01-10 = 41.95 ws/162. Just a note on this peak rate. What Hughie Jennings was for 5 years at his peak, Nap Lajoie was for 10 years at his. It's easy to see why Cleveland named the team for him.
   87. John (You Can Call Me Grandma) Murphy Posted: March 11, 2004 at 06:50 AM (#522706)
Boy, you're really tough on the pitchers, Chris. Who do you think you are? The old John Murphy? :-)
   88. Chris Cobb Posted: March 11, 2004 at 03:24 PM (#522708)
John, I'm just a couple months behind you, I guess.

Seriously, I have three pitchers on my ballot and one just off my ballot, so I don't think I'm being terribly hard on them. If Walsh hadn't been elected overhastily, he'd be in the middle of my ballot now. In 1923 I expect to be adding two pitchers to my ballot while losing one to election. I expect to have supported for election six pitchers from the decade of 1900-1910 (Young, Mathewson, Plank, McGinnity, Walsh, R. Foster), which seems to me a quite fair representation. I don't agree with voters who would have us on track for electing eight or more pitchers from this decade. I think that the cluster of high ERA+ scores from the era is at least as much a product of conditions favorable to pitchers as it is as product of having a cluster of extraordinarily great pitchers. Just as I want to see the seventh and eighth best outfielders of the 1890s in a longer view before I strongly support them for election, I want to see the seventh and eighth best pitchers of the 1900s in a longer view.

Here's something that could influence my support of pitchers: we've talked a lot about comparing pitchers to each other, but there hasn't been much discussion of comparing them to position players of late. How are people making those comparisons, these days? Do you weight peak vs. career differently for pitchers? Do you rely on WS, WARP, quota, gut? I have Waddell and Brown ranked where I do not only due to their relative value compared to other pitchers (a quota element) but also due to my sense that Hugh Duffy, George Van Haltren, Cupid Childs, Ned Williamson, et al. are more valuable based on my conversion of my "value above average" measures into a win-share scale.
   89. Al Peterson Posted: March 11, 2004 at 04:19 PM (#522709)
Chris,

I don't have a problem with supporting your six pitchers (Young, Mathewson, Plank, McGinnity, Walsh, R. Foster) since I've backed them also. But is there anything wrong with supporting a couple others, albeit after thorough inspection? Say Waddell and Brown make 8 for the decade. One is Cy Young who is both a 1890s and 1900s so now you're talking 7 1/2. And what about the 1910s? If that decade is coming up short numberwise for pitchers I'm not adverse to taking 12-13 over a two decade span.

How many pitchers will be getting support from the 1910s? Maybe just a down decade like SS in the 1970s?
   90. AAAAA Posted: March 11, 2004 at 04:23 PM (#522710)
Have followed the HOM discussions for a while now, but never really felt compelled to vote or weigh in on anything until now.

It seems like there is an interesting dilemma shaping up with regard to the '06-'08 Cubs. There is a not insignificant chance that no one from these teams will make the HOM, which in itself is not necessarily a problem. Of course, Sheckard and Brown may well make it, but that's not the point. It seems like much of the problem in evaluating this team (as James points out in the New Abstract) centers around the role of defense - was the infield so good that they should be rated vastly higher (and the pitching downgraded) or vice-versa. To this end, maybe the voting body should look to create a special category for this team, to acknowedge their special place in sabermetric analysis, while admitting that there isn't really a good way to separate the individual contributions. This would also serve to refute the ever so popular notion that the election of Messrs Tinker, Evers and Chance to the actual HOF was among the biggest mistakes ever made by the electorate.

Comments?
   91. Chris Cobb Posted: March 11, 2004 at 05:10 PM (#522711)
And what about the 1910s? If that decade is coming up short numberwise for pitchers I'm not adverse to taking 12-13 over a two decade span.

How many pitchers will be getting support from the 1910s? Maybe just a down decade like SS in the 1970s?


Good point about the 1910s. I haven't started looking closely at that decade yet, but it does look like a thin decade for pitchers. If that turns out to be the case, I'll be more supportive of the 1900s pitchers when we reach the late 1920s and their superiority to the next group becomes evident. Looking at the WS records for the 1910s, it looks like a pretty weak decade overall: a few mega-stars, but not much below them. I guess that's why we'll have the 1927-1932 gap . . .
   92. Marc Posted: March 11, 2004 at 05:17 PM (#522712)
1. The shortage of good pitchers in the '20s will be even more severe than the '10s.

2. Rather than any special category for the Cubs teams, just consider that one major study had Joe Tinker as the #6 SS of all-time.
   93. KJOK Posted: March 11, 2004 at 05:59 PM (#522714)
PRELIMINARY BALLOT:

Using OWP, playing time, and defense (Win Shares/BP) for position players, applied to .500 baseline. Using Runs Saved Above Average and Support Neutral Fibonacci Wins for Pitchers.
   94. John (You Can Call Me Grandma) Murphy Posted: March 11, 2004 at 05:59 PM (#522715)
Chris:

I was only busting your chops (and busting mine for my lack of pitchers on my ballots prior to the last election). I honestly don't have a problem with your ballot.

Besides, anybody with Dickey Pearce high on his ballot is okay with me. :-)
   95. KJOK Posted: March 11, 2004 at 06:11 PM (#522716)
Shortage of Pitcher?
   96. OCF Posted: March 11, 2004 at 06:14 PM (#522717)
How many pitchers will be getting support from the 1910s? Maybe just a down decade like SS in the 1970s?

Looking ahead, I see Rixey, Covaleski, and Faber as all being behind McGinnity but ahead of Waddell - which makes for a pretty tight group. Behind them, I see Wilbur Cooper and Albert Bender as reasonably combarable to Joss, Leever, and Doc White. And towering over all of them are Johnson and Alexander. Many of these pitchers have considerable value in the 1920's, but most of the ones I named, including Johnson and Alexander, had peaked before 1920.
   97. KJOK Posted: March 11, 2004 at 06:34 PM (#522718)
Corrections - "Shortage of pitchers?"
   98. Chris Cobb Posted: March 11, 2004 at 06:48 PM (#522719)
Fine list, KJOK!

Now that you've reminded me of Smokey Joe Williams, I'd suggest that he be listed for 1910s as well as 1920s. He was the top black pitcher by 1913, and appears to have been excellent for 2-3 years before that.
   99. John (You Can Call Me Grandma) Murphy Posted: March 11, 2004 at 06:54 PM (#522720)
Since the attrition rate was higher due to the offensive explosion of the twenties and thirties, that has to be taken into account when analyzing length of career for pitchers between the wars. How many more from that era would have reached 250-300 wins if they had pitched during an easier era (Deadball Era, 1960s).
   100. Al Peterson Posted: March 11, 2004 at 07:51 PM (#522721)
Good point John. We could be dismissing pitching candidates of the 1920s in the same way we did with another high offense era, the 1890s. Context is everything or a ton of hitters from the 20's and 30's will be knocking at the HOM door when looking at the gaudy stats.

Back to this "year": Any Miller Huggins fans out there? Career seems a little short and were heavy at middle infield so I don't see him making a dent.
Page 1 of 3 pages  1 2 3 > 

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Chicago Joe
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.8721 seconds
49 querie(s) executed