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

Wednesday, September 18, 2002

Pitchers for the Hall of Merit

Let’s start discussing the pitchers here. I don’t have any adjusted numbers to post yet, but there’s no reason we can get the discussion cranking.

I take that back. I went through season by season a ways back and came up with pythagorean W-L records for each pitcher, based on his ERA vs. park adjusted league (season by season), adjusting for an average number of decisions in each season (based on the pitcher’s career IP/dec ratio for his career). Those numbers will be in the extended text.



John Clarkson 327-178 (361 Fibonacci wins)
Tim Keefe 351-215 (353)
Old Hoss Radbourn 304-200 (288)
Amos Rusie 261-158 (267)
Al Spalding 216-107 (254)
Tony Mullane 293-211 (252) .464 as a hitter too.
Pud Galvin 362-306 (252)
Jim McCormick 279-200 (241)
Mickey Welch 294-226 (234)
Will White 238-157 (225)
Sliver King 217-139 (212)
Jack Stivetts 202-134 (190)
Bob Caruthers 193-124 (186) .668 as a hitter.
Charlie Buffinton 220-178 (179)
Larry Corcoran 166-100 (170)
Guy Hecker 187-132 (164) .562 as a hitter.
Tommy Bond 220-178 (164)
Sadie McMahon 177-123 (158)
Candy Cummings 150-89 (156)
Bill Hutchinson 195-152 (153)
Ed Morris 171-122 (149)
Monte Ward 158-108 (144) .594 as a hitter.
Jim Whitney 210-185 (136) .548 as a hitter.
Dave Foutz 131-82 (130) .542 as a hitter.

Joey Numbaz (Scruff) Posted: September 18, 2002 at 05:16 PM | 571 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 6 pages  1 2 3 4 5 6 > 
   1. scruff Posted: September 18, 2002 at 05:37 PM (#510729)
My rankings, before looking at Win Shares, etc. would be:

(I'm including time as a position player for guys like Caruthers and Ward)

1. John Clarkson
2. Tim Keefe
3. Old Hoss Radbourn
4. Monte Ward
5. Bob Caruthers
6. Amos Rusie
7. Pud Galvin
8. Mickey Welch
9. Tony Mullane
10. Jim McCormick
11. Will White
12. Silver King
13. Jack Stivetts

Caruthers was a helluva hitter, even though he had his big years in the AA. I'm not sure 7 pitchers will go from this era, but I had always drawn the line somewhere around Galvin/Welch/McCormick in my head.

These rankings are very tenative, I'm open to lots of persuasion.

Note some of the major differences between the records I've posted and the pitcher's actual W-L record. Rusie should have won 16 more games, and Caruthers should have won 24 less, had he played with average teams. I put Caruthers ahead, despite Rusie being a better pitcher, because I figure a pitcher that goes 68-34 (the marginal difference) in that era is probably worth a little less than what Caruthers did with his bat (and glove). I could be wrong.
   2. John (You Can Call Me Grandma) Murphy Posted: September 18, 2002 at 08:00 PM (#510730)
Thanks, Joe. Do you still need help with the spreadsheets?

Here are my top four again:
John Clarkson
Tim Keefe
Amos Rusie (depending on when the year cutoff is)
Al Spalding (could move up higher)

Galvin, Radbourne and Welch would come next. Caruthers and Mullane, IMO, have equal value.

This is tentative still. There are few important pitchers, such as McCormick and Ward, that I haven't finished my work on yet.

BTW Joe, good job with the Deacon White sponsorship!
   3. Marc Posted: September 18, 2002 at 09:44 PM (#510732)
Clarkson is clearly number one unless (like me) you hate to denigrate NA achievements. Then Spalding could rate very very highly indeed. But right now I'd say Clarkson, Spalding, Keefe, Rusie, Radbourne, Bond, Caruthers, Ward and Mullane are worthy candidates, in that order, (maybe). Clarkson and Keefe are probably the only whose credentials cannot really be questioned at some level. I realize Bond is an eccentric choice, but kind of a 19th century Dizzy Dean. After Spalding he had the highest peak value of anybody, including Radbourne. Also agree not all will get in. Welch and Galvin are grossly overrated; each was sort of a Don Sutton type of pitcher, racking up lots of innings and lots of wins for some very good teams, but not frankly pitching much above average.

And yes, Kid Nichols beats anybody that doesn't move quickly. Not sure Clark Griffith is much of a candidate as a player, however. Also, hard not to vote for a man named Silver King; let me guess, he died broke?

Marc
   4. Brian H Posted: September 19, 2002 at 03:53 AM (#510733)
I don't think he'll make it but shouldn't Bobby Mathews at least warrant some discussion with his 297 wins (includes his NA wins)?

Also, given the way that the game was played -- with Pitching staffs of two or three and no relief -- it would seem to me that it would be wise to consider electing an otherwise disproportunate number of Pitchers.
   5. User unknown in local recipient table (Craig B) Posted: September 19, 2002 at 04:10 AM (#510734)
I'm (mildly) surprised to see these lists, because for me Rusie is #1 on this list fairly easily. Anything achieved after 1893 as a pitcher is worth, to me, far more than anything achieved before that time, and particularly before 1884 or so. I pretty much draw the line at roughly 1884 where a pitcher's defensive contributions become more than just slightly more than a shortstop.

This is all provided we don't have a Hall of Fame-like "10-year regular" rule, in which case Rusie doesn't qualify no matter what the actual Hall says. :)

I always wonder whether I misread early baseball, but I don't see Al Spalding as anything more than a borderline HoM candidate. Spalding had 142 strikeouts in 2890-2/3 innings, which just about sums up the importance of the pitcher in National Association baseball. His pitching stats are essentially the Red Stockings' defensive numbers. Spalding's job, essentially, was to throw the ball where the batter asked and handle the gate receipts. He went 55-5 in 1875 and struck out *nine* men in 575 innings. His fielding was probably as important defensively as his pitching.

I'm not trying to denigrate him; he was a very good player. A fine hitter, a good fielder, a good pitcher according to those who observed him at the time, but pitcher was not at all the position we think of it as now. Pitchers from the NA, in my mind, are best seen as another class of infielder.
   6. User unknown in local recipient table (Craig B) Posted: September 19, 2002 at 04:51 AM (#510735)
I'm (mildly) surprised to see these lists, because for me Rusie is #1 on this list fairly easily. Anything achieved after 1893 as a pitcher is worth, to me, far more than anything achieved before that time, and particularly before 1884 or so. I pretty much draw the line at roughly 1884 where a pitcher's defensive contributions become more than just slightly more than a shortstop.

This is all provided we don't have a Hall of Fame-like "10-year regular" rule, in which case Rusie doesn't qualify no matter what the actual Hall says. :)

I always wonder whether I misread early baseball, but I don't see Al Spalding as anything more than a borderline HoM candidate. Spalding had 142 strikeouts in 2890-2/3 innings, which just about sums up the importance of the pitcher in National Association baseball. His pitching stats are essentially the Red Stockings' defensive numbers. Spalding's job, essentially, was to throw the ball where the batter asked and handle the gate receipts. He went 55-5 in 1875 and struck out *nine* men in 575 innings. His fielding was probably as important defensively as his pitching.

I'm not trying to denigrate him; he was a very good player. A fine hitter, a good fielder, a good pitcher according to those who observed him at the time, but pitcher was not at all the position we think of it as now. Pitchers from the NA, in my mind, are best seen as another class of infielder.
   7. John (You Can Call Me Grandma) Murphy Posted: September 19, 2002 at 06:40 AM (#510737)
Spalding looks to be the best pitcher for each year between 1872-1875. He wasn't too shabby in 1871 and 1876 either. You have to balance the quality of his pitching (ERA) with the amount of games he pitched for a season. He was, without a doubt, the best pitcher of the NA.

In his career Spalding only won 2 ERA titles and only 1 ERA+ title, we are not talking about Koufax here.

Koufax "only" won two ERA+ titles. I think he's very comparable to Sandy in terms of dominating their times. Plus, Spalding destroys him as a hitter.
   8. John (You Can Call Me Grandma) Murphy Posted: September 19, 2002 at 07:05 AM (#510738)
BTW, why is Monte Ward listed here (rather than at short)?
   9. Rick A. Posted: September 19, 2002 at 03:19 PM (#510739)
I'm really interested in learning more about 19th century baseball and its players, especially in regards to how to best rate these players. My knowledge of 19th century stats, rule changes, reasons why certain stats should be discounted, is woefully inadequate. What books or websites (besides baseball-reference) do you guys recommend to learn about this stuff?

Thanks,
Rick
   10. jimd Posted: September 19, 2002 at 06:17 PM (#510740)
baseball-reference.com is a wonderful site, but it's criteria for pitching leaders/black-ink during this era are inadequate.

No way would anybody from that era consider Cherokee Fisher winning any kind of pitching title in 1873. He only pitched 84 innings; he was a part-timer, a back up. It's like giving out a modern batting title based on 84 AB. Spalding earned that ERA+ title.

1875 is more interesting. Spalding came in third behind both of the Hartford duo of Bond and Cummings, who split the innings 352 and 417 respectively. Spalding pitched 575 innings, with 140 going mostly to Jack Manning. Would Cummings or Bond have been eligible for any title given out at that time? The innings are impressive by today's standards, but for that era, it's somewhat analogous to a modern team with co-starting catchers. Yeah, they're both good, but these guys are only playing slightly more than half of the time. (Does anyone know if this was an injury situation with the backup performing as well as the starter, or is Hartford a few years ahead of its time? Two coequal starting pitchers didn't become popular until the schedule began lengthening in the early 1880's.)

Spalding earned two, maybe three, ERA+ titles.
   11. John (You Can Call Me Grandma) Murphy Posted: September 19, 2002 at 08:43 PM (#510741)
Spalding earned two, maybe three, ERA+ titles.

Good point.
   12. DanG Posted: September 20, 2002 at 05:02 AM (#510744)
According to Total Baseball, The leaders in WAT for this era are:

McCormick..56.3
Galvin.....54.3
Radbourn...52.2
Will White.43.5
Amos Rusie.36.8
Welch......36.8
Mullane....35.8
Buffinton..35.0
Jim Devlin.32.5
Guy Hecker.31.5
Ed Morris..30.8
Caruthers..30.4
McMahon....29.5

Players in the NA are not included.
All of these marks are among the top 50 pitchers in history. BTW, Clarkson is down some at 23.2. Keefe is at 22.5.

Also, I think Craig B makes an important point. A pitcher's role 125 years ago was a totally different concept than we're used to. I'm not quite sure what to make of it. Was Spalding as worthless as 9 strikeouts in one season would indicate? Should we think of him as analagous to a slow-pitch softball tosser? How much do we need to devalue Spalding's achievements?

The game was evolving rapidly at that time - the percentage of baseball that was defense in 1872 was A LOT higher than in 1882. This percentage, I think, continued in a gradually lessening decline, until about 50 years ago. I think the decline in defense's importance has continued since then, but not significantly.

   13. Marc Posted: September 20, 2002 at 12:19 PM (#510745)
Can somebody help me on WAT vs. WAA? It seems to me that WAT discriminates against a guy on a great team, especially a great team from this era that is probably winning .700 or more. Looking at the list, it seems to especially knock Keefe down for that reason?

Secondly, there was talk some time back of adjusting win shares of the early pitchers to reflect about a 50-50 split of "defensive" shares between pitchers and position players. I think that is what scruff et al are working on?

But regardless of the debate about the importance of pitching, having played a bunch of slow pitch softball, if your slow pitch pitcher makes it moderately tough for the batter to get a cut the ball that he likes, and fields his position, and is one of the top one or two hitters on the team, that guy is a a star. Spalding was the equivalent of Pedro Martinez (ok, a slow pitch softball version of Pedro Martinez) and Barry Bonds rolled into one for five or six years. Taking pennants as the denominator, well, he only had five or six of them that he influenced. But you could also argue that he had ten to twelve years worth of achievement.
   14. scruff Posted: September 20, 2002 at 01:41 PM (#510746)
One thing to consider, on the 9 K's for Spalding issue . . . DIPS does not necessarily apply to 19th century baseball. It's quite possible, I'd assume probable that different pitchers would be able cause batters to not hit the ball as hard, based on a variety of factors. I don't know that this is true, but it's worth investigating.

I have a very hard time accepting DIPS before 1920. It was a completely different game back then.

Errors were a huge part of defense back then, because there were so many of them. So I'd think that ERA does a pretty good job of setting apart the pitcher and the defense.

Spalding's ERA+ numbers are pretty damn good. Sure his defense probably helped him, but his ERA+ was 142 for his career. I can't imagine that's all good D. Boston had very good defensive teams, but I don't think they were so outstanding as to make him look like a great pitcher.

My issue is more that quality of the league, combined with a short (7-years) career makes him borderline in my opinion. But I do think he was a pretty good pitcher, more than just a slo-pitch softball pitcher. And I don't think you take DIPS and apply it pre-1920. Until I see a study that proves I'm wrong at least . . .
   15. User unknown in local recipient table (Craig B) Posted: September 20, 2002 at 06:40 PM (#510747)
Scruff,

I'm not suggesting that DIPS-style research applies to 19th century baseball. I haven't run any numbers, but there are large differences in hits/balls in play numbers for pitchers on the same team at that time, and while I want to do some research before saying that there are correlations over time, it certainly looks like it (of course, it also looks like that today, but we know it isn't true). That being said, the differences are not *that* large. I might try studying that this weekend.

After 1881 or so, strikeout numbers rise substantially and on that metric alone pitchers clearly start becoming more important.

Spalding's ERA+ numbers are pretty damn good. Sure his defense probably helped him, but his ERA+ was 142 for his career. I can't imagine that's all good D. Boston had very good defensive teams, but I don't think they were so outstanding as to make him look like a great pitcher.

We can gain some insight into how good Spalding was by looking at the ERA+ of his teammates over that span. Spalding's teammates, for the six years (1871-76) that he was pitching, earned an ERA+ of 102.

Clearly, Spalding at 142 is well above that. But look at who those teammates are: they are just position players, mostly Harry Wright, Cal McVey, and Jack Manning. None of those players was a pitcher (Manning pitched fairly well - as the third pitcher for Boston - one more year after his two years with Spalding and then had just a few disastrous outings the two years after that) they were just position players pressed into service as pitchers, and they posted an ERA+ of 102 for Spalding's teams.

Remember the rules in those days. Batters get to call for a high or low pitch. All pitches are underhand and are straight-armed. Nine balls for a walk. The ball is pitched from 45 feet.

Bill James says the game resembles fast-pitch softball, but these pitchers are nowhere near as important as the pitchers that we see in modern fast-pitch, where strikeouts are extremely common.

Also important to remember is that to take 1875 as an example, an ERA of 2.46 meant you were allowing about 6.14 runs per game (those were the league averages). ERA is a bad measure of pitching performance now; it's terrible for 1875 because the errors throw it all out of whack.

It's not that Spalding has nothing to do with his teams' excellent runs allowed totals, and he may well have a much larger contribution than DIPS would indicate. But as I said before, I think his role is much closer to being another infielder than a modern pitcher. His excellence as a fielder and his fine hitting, though, make him a candidate.

It would be a mistake to ignore Spalding's fielding. Contemporary sources talk about his excellence at covering the bunt; he has very low error totals (his error rate is 40% better than average - about 50 errors over his career) and a very high range factor. Playing every inning, his fielding contribution is significant - 50 errors and 170 extra plays made is a lot.
   16. Marc Posted: September 20, 2002 at 08:43 PM (#510748)
At the risk of repeating myself, a good slow-pitch softball pitcher makes all the difference, in part by putting spin on the ball and in part by throwing anywhere except down the middle. It is especially effective in slow-pitch softball to be able to throw nice "deep" (ie. high) strikes. Anyone who says a slow-pitch softball pitcher or a 19th century professional baseball pitcher doesn't make a difference (or that he is just another infielder) clearly has not played high level slow-pitch softball.

I would say a slow pitch pitcher is more like (maybe I should say, more analogous) to a modern catcher who calls his own pitches and locations, in the sense that every single pitch and every single play begins with a ball that is easier or harder to hit in part because of what he does. Except that he obviously has even more of a role than a modern catcher. We don't have much of a clue how to measure a modern catcher's defensive contributions on those dimensions, but we know they're there.
   17. jimd Posted: September 20, 2002 at 10:25 PM (#510749)
Taking a different tack here. The estimates that I've seen for the impact of pitching in the NA (based on previous posts by Charles Saeger and scruff) is that it's about twenty-something percent of defense (contrast with James' assertion that it's around 68.5% of defense in the modern game). Doing some rough calculations here for the impact of the pitcher in the NA: if pitching is 22% of defense, and the pitcher is an average pitcher/hitter/fielder on a .500 team, and he pitches every inning, then he will contribute about 40 (adjusted) Win Shares (13 on Offense and 27 on Defense). In other words, an average pitcher is still the star/MVP of the team, though it is now possible for a position player having a GREAT season to compete with him.

My crude calculations of last July on Spalding has him worth about 20 (adjusted) WS on offense each year. Give him any extra credit for being an above average pitcher and his (adjusted) WS totals are into the 50's. If these estimates are at all close, his impact on individual NA seasons is Ruthean, but it's a short career. ("Koufax/Dean rule" can then apply.)

Note: remember also that there are only limited spots available for pitchers. If the talent search was highly efficient, then one team would have Randy, another team Schilling, another Pedro, another Lowe, another Zito, another Glavine, Maddux, Clemens, etc. You run out of pitching slots quick with only 6-12 teams. Consider "average" being in the middle of that group.
   18. Marc Posted: September 21, 2002 at 02:19 PM (#510750)
Ruthean, yes. Al Spalding was indeed the Babe Ruth of the NA, good analogy. Fitting the NA into the broader history of MLB is tough, but I think at least we can rank the impact that the stars of the NA had within their own sphere. Spalding was clearly first among those ranks.

After that you've got to figure out how to rank Barnes and Wright, whose contributions were generally limited to pre- and NA days. On the field and beginning in 1871 Barnes gets the edge but Wright's other contributions would probably give him the edge if you want to take the broader view. With all due respect, any others such as Meyerle, Start and Force, fall way short.

Then you've got the NA guys who continued to play beyond 1876, basically O'Rourke and White but perhaps including Hines, Sutton, probably not McVey. And Anson, of course, though his status is clear whether or not you even consider his NA career. In fact, all of these can be considered with or without the NA stats and you probably come to about the same ranking.

In any event, however, it seems crystal clear that for six or seven years Al Spalding was a giant among men--Ruthean, Gretzky-esque, Tigerish, Woodsian, pick your adjective. His value was way beyond what Dizzy Dean or Sandy Koufax achieved. Other than Anson and perhaps O'Rourke, I can't see how any of the above could possibly go into the HOM ahead of Al.
   19. jimd Posted: September 23, 2002 at 07:09 PM (#510752)
kind of average for his (very fine) team

The composite record of Boston for the 5 years of the NA is 225-60, a .789 winning percentage. Their best season was 1875 when they went 71-8 (.899), a record that any great NBA team would be proud to claim. Dividing the adjusted WS evenly among the 9 starters on that team will give each player 48 WS apiece. (!)


Boston is essentially the NA All-Star team taking on all comers and creaming them.

It also brings up another issue when using WS. Win Shares are designed to evaluate teams that are fairly close to .500, because of its assumption of a linear relationship between winning and run scoring/run prevention. This assumption allows it to give approximately the same number of WS for the same statistical performance independent of the team performance. Extreme teams violate that assumption and can produce distorted results; for extremely good teams, the result is that it UNDER-estimates the number of win shares the players would receive on another team, because their collective performance doesn't produce as many wins as it would if the team was broken up and the players placed distributed onto other teams. (Adding another superstar to a .750 team
doesn't add as many wins to that team as it would adding him to a .500 team; on the .750 team, only 25% of the added benefit gets applied to losses, 75% goes to more and bigger blowouts.)

   20. jimd Posted: September 23, 2002 at 10:45 PM (#510753)
The fact that some of these guys' stats fell off the table when they went to the NL (McGeary, Force) suggests that the NA league quality was relatively poor.

I'd be surprised at any numbers indicating the 1875 NA was much different from 1876 NL. 6 of the top 7 NA teams carried on into the NL; the bottom 6 had already quit at various times during the 1875 season after discovering they had no business being in that association. (Any team that wanted to could join the NA; pay your entry fee, agree to abide by their rules, and you were in.) The 5th place Philadelphia Whites weren't invited, because the NL was run by the team owners, instead of the players, and the Athletics, who had finished ahead of the Whites, got the Philly franchise.

Two new teams were added. Louisville is composed of practically all NA veterans, an "expansion team" built from free agents (there is no reserve clause until 1879). Cincinnati was apparently another independent team willing to try its luck in the new league. It also had no business being in the league from a competitive viewpoint, but it worked hard on improving and was in the pennant race two years later.

   21. jimd Posted: September 23, 2002 at 10:54 PM (#510754)
Actually, calling them "owners" is a misnomer, because there was nothing to "own". Backers is a better term; they agreed to invest in the team, put up a pool of money to cover start-up expenses, initial travel and salaries, etc. in return for a substantial share of the profits, if any. I suppose it was like getting a Broadway play off the ground. It wouldn't be until years later that there'd be some real estate and a brand name that would be worth "owning".
   22. scruff Posted: September 24, 2002 at 12:33 PM (#510755)
Jim -- You are right. Robert's research (not sure if he ever posted it) and some other things I've seen agree that the 1875 NA was significantly stronger than the 1876 NL.

Spalding will be a very, very interesting one to watch.
   23. MattB Posted: September 30, 2002 at 01:37 PM (#510756)
"I don't think he'll make it but shouldn't Bobby Mathews at least warrant some discussion with his 297 wins (includes his NA wins)?"

I definitely don't think you should right off 297 wins, and I'm terribly surprised the Mathews doesn't crack anyone's Top 10/ Top 5 pitcher lists.

Have we determined that ERA+ is the only stat worth considering?

Check out the All Time Gray Ink Leaderboard:

http://www.baseball-reference.com/leaders/gray_ink.shtml

Bobby Mathews leads ALL 19th century pitchers in gray ink. When you look at his full career up against his peers, you will be hard pressed to find an all-around better pitcher.

He was consistently first or second in K/9. 14th in career innings pitched (behind only Pud Galvin and Tim Keefe for the era). COnsistently among the fewest hits allowed.

Now, gray ink then obviously doesn't mean as much as gray ink now, but ranked against his peers, it certainly has some meaning.
   24. MattB Posted: September 30, 2002 at 01:38 PM (#510757)
Note also, on the link above, the Mathews has the most gray ink of any eligible player not in the Hall of Fame. (Bert Blyleven is second).
   25. User unknown in local recipient table (Craig B) Posted: September 30, 2002 at 01:52 PM (#510758)
gray ink then obviously doesn't mean as much as gray ink now

I should certainly think not. Mathews spent most of his career in leagues where around 14 or 16 guys qualified for the ERA title. Coming in the top 10 in those leagues is meaningless.
   26. MattB Posted: September 30, 2002 at 02:36 PM (#510759)
Craig,

But we are comparing him to his peers. If coming up in the top ten in meaningless, why didn't other pitchers who had longer careers do it more?

If we are electing 19th century guys without any thought to how they'd rank in a timeline comparison to 20th century guys, then we should look at each portion of the 19th century.

It is only an artifact of the voting structure that all 19th century pitchers are thrown together in the first election.

Imagine the first vote was only for players who had retired by 1890. It would be a contest only between Bobby Mathews and Jim McCormick, and it's a contest I think Mathews wins.

No one will be saying in the 1990 election, "Well, Jim Palmer -- the best pitcher of the '70's and '80's -- is eligible, but I know that Nolan Ryan and Steve Carlton are coming up in the next ten years, so I'll vote for them first."

20 years is more than a career length. If you were the best player at your position over any 20-year period, I'd say you are one of the best.
   27. John (You Can Call Me Grandma) Murphy Posted: September 30, 2002 at 02:52 PM (#510760)
Imagine the first vote was only for players who had retired by 1890. It would be a contest only between Bobby Mathews and Jim McCormick, and it's a contest I think Mathews wins.

I would take McCormick over Mathews (because of better competition). Spalding would win the competition hands down over both of them.

Mathews was durable (for his time), but except for one or two seasons, he was a journeyman pitcher (especially in the NL). Maybe when his NA Win Shares are compiled, I might change my mind (but I doubt it).
   28. DanG Posted: September 30, 2002 at 02:54 PM (#510761)
I have to agree with Craig that Mathews isn't quite one of the top five pitching candidates. The few years that he pitched well (1872-74 and 1883-85) were in leagues of questionable quality. There was a five year stretch (1877-81) where he couldn't hold a full time job in the big league. Was a tearing up the minors somewhere in that stretch?

Yes, we are comparing Mathews to his peers. But in his day there were only 10 to 20 regular pitchers in the league. Many of Mathews top ten finishes were due to simply showing up, unrelated to any excellence.
   29. John (You Can Call Me Grandma) Murphy Posted: September 30, 2002 at 03:18 PM (#510762)
The only season that one could point to as a great season would be Mathew's 1874 campaign. I probably would pick Spalding over him, though. 1873 wasn't too shabby for him either.
   30. MattB Posted: September 30, 2002 at 03:23 PM (#510763)
"Maybe when his NA Win Shares are compiled, I might change my mind (but I doubt it)."

Speaking of which, are we still considering Win Shares a valid way of looking at 19th century players, after it so clearly failed to correlate with observed reality in 2001?

Giambi and Ichiro ended up with nearly identical WS totals (38 to 36), after almost everyone on this board though Suzuki was one of the worst choices ever. I can't think of anyone who would have traded Giambi for Ichiro and a 1 game head start.

The best answer I heard was that Seattle won so many games, that the extra Win Shares sloshing around fall on players who individually didn't deserve them. The 19th Century is full of team with obscene winning percentages that would warp the numbers even more. That goes triple for pitchers who were either (a) the most valuable players on the team since the pitched every day, or (b) the least, because they threw underhand and didn't strike anyone out.

For a stat that seems least effective with extreme players or extreme teams, it strikes me as a particularly bad way to judge an extreme position in an extreme era.
   31. John (You Can Call Me Grandma) Murphy Posted: September 30, 2002 at 03:44 PM (#510764)
The post 1987 Win Shares are different because of the "clutch performance" factors that James included (which Ichiro did very well with the formula). I am not as sold on them as the pre-1987 Win Shares.

The best answer I heard was that Seattle won so many games, that the extra Win Shares sloshing around fall on players who individually didn't deserve them.

That wouldn't explain why Boone didn't have more WS than Ichiro still.
   32. MattB Posted: September 30, 2002 at 03:57 PM (#510765)
Yeah. It's hard to figure out. The formula itself doesn't look particularly problematic. Each piece seems just as reasonable as not. It just doesn't really add up very well. And it doesn't always fail to add up in the correct way.

I have nothing against revolutionary new techniques, but when one of them contradicts the older techniques, I need some convincing as to why the new one is better. This one gives me numbers that are intuitively wrong (Ichiro too high, Pedro too low), but then fails to explain why (except to say that the pitching numbers all seemed low, so he fudged them by a few percent.)

If WS is lightyears ahead in defensive measures, but trailing well behind in offense and pitching, I don't see how it's any net gain.
   33. John (You Can Call Me Grandma) Murphy Posted: September 30, 2002 at 05:05 PM (#510766)
When I do my analysis with WS, I compare each player to other players at that position. This tends to overcome most problems with the stat. Besides, it's only one measure. You still have to use your own judgment.
   34. MattB Posted: September 30, 2002 at 05:58 PM (#510767)
Gee, what good are stats if I still have to use my own judgment? :-)
   35. MattB Posted: September 30, 2002 at 07:04 PM (#510768)
Gee, what good are stats if I still have to use my own judgment? :-)
   36. John (You Can Call Me Grandma) Murphy Posted: September 30, 2002 at 07:10 PM (#510769)
Matt:
Did you have a seizure in between your double posts? :-)
   37. MattB Posted: September 30, 2002 at 07:13 PM (#510770)
Man. That's got to be some kind of record for time between accidental double posts.
   38. MattB Posted: September 30, 2002 at 07:15 PM (#510771)
I hit the "refresh" button after being away from the computer, and the screen said, "Unable to refresh, do you want to re-load?" and I said yes, and it re-posted. Weird.
   39. Marc Posted: October 01, 2002 at 01:02 AM (#510772)
>If WS is lightyears ahead in defensive measures, but trailing well behind in offense and pitching, I
don't see how it's any net gain.

You gotta admit, however, that at least it's a lot better than TPR, and if we got to choose between WS and BBWAA to pick the MVP, I'd take my chances with WS every time.

John, tell me how you do your analysis within the position. You mean with WS? Or just your own ranking or rating or what?

   40. John (You Can Call Me Grandma) Murphy Posted: October 04, 2002 at 05:18 AM (#510773)
Marc:
Sorry about ignoring your post. Before I explain what my method for analysis is, I need to refine one aspect of it. I'll lay it out (for better or for worse) within a few days.
   41. John (You Can Call Me Grandma) Murphy Posted: October 05, 2002 at 09:18 PM (#510775)
I have Keefe as the best pitcher in the majors for 1883.
   42. Marc Posted: October 06, 2002 at 02:02 AM (#510776)
Andrew, by all means, I'm sure we're all interested in seeing your analysis. I undoubtedly have not looked at it in as much detail, but I so far have arrived at exactly opposite conclusion--Keefe in, Welch vastly over-rated. I agree that Clarkson is the obvious choice, but I like Spalding's peak value.
   43. John (You Can Call Me Grandma) Murphy Posted: October 06, 2002 at 04:08 PM (#510778)
For the seasons that they pitched together, I have Keefe as the best pitcher on the staff for 81, 82, 83 (best pitcher in baseball), 84, 86, 87, and 88. I have Welch as the best in 80, 85 (best pitcher in baseball), 89 and 91 (Keefe was traded to the Phillies during the season). Keefe's WS per Game was higher than Welch.
   44. jimd Posted: October 07, 2002 at 05:31 PM (#510779)
1883-- Troy folds; the Giants acquire the best players on the team including Wlech but not Keefe; Keefe forced to move to the lesser league (unable to get a job in the NL?),

The story I've read is that the New York group bought the rights to the Troy contracts. And they were backing both of the New York teams, NL and AA, hedging their bets on which league would win out. The Giants got Welch, Connor, Ewing, and OF Gillespie. The Metropolitans got Keefe, OF Roseman, and C Holbert. Young SS Pfeffer wound up with Chicago. 2B-Manager "Death-to-Flying-Things" Ferguson moved on to attempt building the new Phillies team from scratch. After a couple of years, the owners decided the NL was the league to back (particularly after the AA added a team in Brooklyn) and began consolidating the talent into the NL. Question: batteries were a big deal back then: was it Welch/Ewing and Keefe/Holbert, or did the catchers and pitchers mix it up?
   45. jimd Posted: October 07, 2002 at 05:52 PM (#510780)
In case my point wasn't obvious: Keefe was not a free agent; he played where he was told, though I don't know whether he (or any of the others) had any input into the decisons.
   46. Marc Posted: October 07, 2002 at 07:36 PM (#510781)
Certainly on a career basis it is hard to make a case for Welch over Keefe. Tim only spent two years in the AA and you'd have to discount his achievements there a whole lot to even them up. In fact, discount Keefe's AA achievements by 50 percent (cut both his wins and losses in half) and he would be at 305-203 vs. Welch's 307-210, which is probably nothing more than an interesting coincidence. I don't think you should discount Keefe's AA career by half, 10-25% would be more accurate.

ERA+ is perhaps most telling, Keefe's career number being 125 and Welch's 113. Keefe's AA ERA+ was 138, ie. 10% over his career average for 1/5 of his innings. Apparently his NL ERA+ is still about 122.5, almost ten points better than Mickey's. Keefe's NL only WHIP is 1.17, Welch's is 1.225.

On a peak basis, if indeed you cut his 70 WS in 1883 by half and cut his WS in '84 in half as well, which again seems a bit too much, you've now got their WS almost exactly equal as well (354.5 to 354, again probably just an interesting coincidence). And most importantly for Mickey's side, his NL peak WS of 57-42-42 are higher than Keefe's NL peak of 42-39-38.

But even in '85, by far Welch's best year, Keefe had a better ERA, gave up fewer WHIP amd lower OAV and OOB. That is not to say Keefe would have done so if he had pitched the extra 100 innings that Mickey pitched that year but.... From '86 on, Keefe was far more productive despite being 2.5 years older. They had each pitched about the same number of career innings by that time, and perhaps Tim benefited from not having pitched those innings until he was in fact a couple years older than Welch was with the same workload. (Of course, who knows how much pitching they had done before joining the NL? I don't.)

Keefe had 12 NL seasons of which 9 could be said to have been productive, while Welch had 13 NL seasons of which 11 could be said to be productive, so they each had 11 productive seasons. Welch was washed up at 32 and Keefe pitched til age 35. Keefe pitched about 300 more innings career with fewer hits, home runs and walks and more strikeouts. Keefe's peak NL ERA+ of 168, 156, 138 in fact exceeds Welch's NL ERA+ peaks of 160, 141, 131.

In short they are more equal than just a cursory look at their career totals would suggest, and certainly they are more comp than I had thought before this discussion. But Keefe seems to have a larger edge on career than you can give Welch (if at all) on peak. Keefe by a schnipple.
   47. scruff Posted: October 07, 2002 at 09:22 PM (#510782)
"The post 1987 Win Shares are different because of the "clutch performance" factors that James included (which Ichiro did very well with the formula). I am not as sold on them as the pre-1987 Win Shares."

John, the WS adjustments aren't that outragous for clutch performance. Hitters that hit better than expected w/RISP or hit more HR w/men on base than expected are rewarded, others are not. But the only adjustment is to the player's RC, and usually this isn't more than +/- 5 runs, although I'm not sure what it was for Ichiro! last year.

But this is also offset by the fact that RC totals are then adjusted to get the sum of the players equal to the team's run total.

Personally, I really don't mind this all that much. Ichiro! does deserve extra credit for hitting .450 or whatever it was w/RISP last year. It's one of the reasons the Mariners won 116 games.
   48. Marc Posted: October 08, 2002 at 04:42 AM (#510784)
1880s pitchers

Funny, Andrew, I had always thought of Radbourn coming before Keefe and Welch, and Clarkson coming later, and yes, I also think of Clarkson as the best of the four. But to my surprise they were all but Old Hoss just two years apart--or two and two, total four, born in '57, '59 and '61, and retired after '92, '93 and '94. Radbourn was older, born in '54 but started in '81, a year after Keefe and Welch, and retired after '91.

Two comments in response to yours: How much do you discount Keefe's two years in the AA? 50 percent doesn't even make Welch clearly better, it makes him even. You must be discounting it altogether. I don't think that's accurate.

And second, about innings: Clarkson threw 4500 innings in 12 years, 600 twice and 500 once. Keefe threw 5000 innings in 14 years, 600 once, 500 once. Radbourn 4500 in 11 years, 600 twice, 500 once. Welch 4800 in 13 years, never 600 but 500 three times. And games, starts, complete games--Keefe leads on all three. Who cares how old Keefe was when he got his innings. In the end he got more than any of them.

Clarkson's edge has always seemed to derive from a better winning percentage and higher peak--his win total is lower than Keefe's. Clarkson's top 3 seasons total 173 WS, Radbourn's 199, Keefe 159 (and only 119 in NL) and Welch 145. Yet Keefe ends up with the most career WS. And add up the top 3 ERA+ numbers, you get Radbourn at 491, Clarkson 466, Welch 432 and Keefe 465 (and 433 in NL seasons only). Edges for everybody except Welch.

On career ERA+, again it is Clarkson on top at 134, but Keefe beats Radbourn 126-120 with Welch down at 114. And Keefe gave up an OAV of just .231, Clarkson .244 and Radbourn and Welch .246. And Keefe has the lowest WHIP at 1.14, with Radbourn next at 1.18, Clarkson 1.23 and Welch 1.24.

Clarkson also played for better teams but played the furthest above the team. His winning percentage of .648 was 57 points above the team (.591). Radbourn's numbers are .614 over .559 (+55), Keefe's (NL only) .603 and .578 (+25) and Welch's .594 and .555 (+39). This is clearly Keefe's worst category.

In sum, for peak value it looks to me like Radbourn, Clarkson, Welch and Keefe (unless you include his AA seasons at full value, then he moves ahead only of Welch). For career value, I'd have a hard time between Keefe and Clarkson, with Old Hoss third and Welch clearly fourth. Add them together, and HOF voters liked Old Hoss way back in '39, Clarkson in '63, Keefe in '64 and Welch in '74. I think that vastly overrates Radbourn at Clarkson and Keefe's expense.

I'd have to rank them:

Clarkson
Keefe and/or Radbourn depending on whether you like the higher peak or the career totals
and then Welch

And while I haven't looked as closely at the others, I doubt that there's another pitcher except Al Spalding whose peak was before 1893 that rates with these guys. (I'd put Spalding ahead of Welch.)

   49. Marc Posted: October 08, 2002 at 05:13 AM (#510785)
PS. Where did the discussion go concerning the relative strength of the 19th century leagues? I cannot find it at the moment. Thx.
   50. Marc Posted: October 08, 2002 at 04:39 PM (#510787)
Looking more closely at these pitchers than ever before, I've certainly learned one thing. Most of the "name" pitchers of the 19th century peaked in the '80s, which (like "slugging" in the 1930s) simply tells you that the structure and strategy of the game favored those pitchers. My previous posts were based on the assumption that the Big 4 of 19th century pitchers was Clarkson, Radbourn, Keefe and Welch. The only significant outlier on the earlier side is Spalding, though I want to consider Tommy Bond and maybe Will White a little more. And the major outliers who came later are of course Rusie and Nichols, depending on whether you think of the Kid as a 19th or 20th century pitcher. Stivetts perhaps also deserves a closer look.

Yet, even knowing that those 1880s pitchers had substantial advantages (if pitching 600 innings can be called an advantage), I can't help but also want to consider some others--Caruthers, Mullane, Galvin and McCormick, at least.

So far, Andrew, I'm still looking at Rusie, like you I don't know what to do with Ward, and I think your differential between Radbourn and Keefe is too big.

But I guess this particular post is really just to say I've really learned a lot about 19th century pitchers and pitching already. Thanks everybody.
   51. John (You Can Call Me Grandma) Murphy Posted: October 08, 2002 at 05:57 PM (#510788)
Scruff:
I'm not against giving a player credit for clutch performance. That to me makes sense. I'm just not sure James' formula is correct. Wouldn't slugging percentage be a better choice for the stat instead of BA or home runs?
   52. scruff Posted: October 08, 2002 at 08:33 PM (#510789)
Yeah, it probably would be. But . . .

AVG w/RISP is a pretty simple adjustment, because if you get a hit with someone there a run is going to score. If you get a double or a triple it doesn't mean 3 guys are going to score. Maybe SLG w/MoB would be a better (additional?) adjustment.

HR w/MoB is also a decent a adjustment in my opinion. Barry's HR are not as valuable as Sammy's because he doesn't get to hit too many with MoB (I'm guessing, I haven't looked it up).

I do think there could be other adjustments, not sure why these two adjustments are the only ones made. I'd add the SLG w/MoB for sure. Leadoff OBP might be another one, although that's not a guaranteed run, whereas a hit with RISP almost always is.

Again, all of this comes out in the wash, because the totals are then adjusted to what the team scored. So if Barry gets a few extra RC, JT Snow and the others are losing them. I really don't think it's such a big adjustment as to make the findings all that different. I really think it's just a minor tweak.

One of the reason Ichiro! had so many WS, relative to Giambi was that he had 738 PA to Giambi's 671. Give Ichiro! Giambi's PA, and he has about 33 WS to Giambi's 38 which is more reasonable.

Ichiro was really good last year. He hit .350 and add 56-70 base stealing. He only hit into 3 DP (Giambi hit into 17). He was a pretty good defensive RF. Throw in the BA w/RISP as well. He was probably a little closer to Giambi than we thought at the time, WS isn't saying he was better than Giambi, just close. I think that's a reasonable assessment.

WS had Ichiro! as the 4th best player in the AL last year. Considering the team won 116 games (meaning they were quite efficient so the individuals on average were a little better than their individual stats would otherwise indicate), I don't find that all that unreasonable.
   53. Marc Posted: October 08, 2002 at 08:42 PM (#510790)
Not to beat a dead horse--OK, to beat a dead horse--are there any second tier pitchers from the 1880s who ought to be elevated to the first tier?

Caruthers only pitched 9 years despite the fact he was not overworked by the standards of the day--never even 500 innings. So obviously his career counting totals are not competitive. So the only real question is "How high the peak." And of course his peak came in the AA. Well, from his peak his ERA+ went like this--158, 148, 138, 125, 119, 111, 106, 55 and done. When did he jump to the NL? Doesn't matter. His decline was so steady that it suggests he was the same pitcher in the AA that he would have been in the NL. He would have been a star in the NL, too.

Mullane was very nearly the same pitcher as Caruthers except he pitched 5 years more. His ERA+ peaks in the AA were (Mullane) 159, 135, 134 vs. (Caruthers) 158, 148, 138. Caruthers' WS edge over 3 best years is just 162-159. Mullane pitched on weaker teams so his WL is not as good but he pitched more innings (over 500 twice). Both moved to the NL in 1890 and Bob had three more years, two productive, Mullane, though older by a full 5 years, pitched another 5 years, 4 of them at least above average.

Then come the harder choices, Galvin and McCormick. Galvin pitched 14 years and at 5900 almost 900 more innings than any of the other leading pitchers of the '80s, though like most he pitched more than 500 innings three times (600 twice). He just had more of those 400 inning years. So at 361 he had more wins than anybody, and more WS than anyone but Keefe, but also more losses and a weak ERA+ of 109 with peaks of 158-127-118, but the 127 came in the AA. His 3 yer WS peak is 140, lowest of them all if you include the AA. So he really only had one great season where everbody else we have considered had 3-4-5.

Finally, McCormick has been overlooked but had a lot of black ink, a 3 year peak for WS of 149, an ERA+ of 117, a 3 year peak for ERA+ of 171-128-127 (excluding 166 in the UA). His WL is not great but when he finally played for a good team, Chicago in '85 and '86, though past his prime, he went 51-18 with ERA+ of 124 and 128.

There are of course pitchers who had higher peaks or longer careers but not both. For the record, taken strictly as a pitcher I don't see Ward in either top 10. So finally adding these guys in I would have to rank the pre-1993 pitchers as follows:

Peak Value
1. Spalding, 2. Radbourn, 3. Clarkson, 4. Caruthers, 5. Bond, 6. Keefe, 7. Mullane, 8. Corcoran, 9. Welch, 10. McCormick

Career
1. Clarkson, 2. Keefe, 3. Welch, 4. Radbourn, 5. Galvin, 6. Mullane, 7. Spalding, 8. McCormick, 9. Caruthers, 10. White

The considering peak and career as of equal weight in Total:
1. Clarkson, 2. Keefe, 3. Spalding, 4. Radbourn, 5. Welch, 6. Mullane, 7. Caruthers, 8. Galvin, 9. McCormick, 10. Bond, and of course only about half of them make the HOM.

   54. John (You Can Call Me Grandma) Murphy Posted: October 10, 2002 at 07:20 PM (#510793)
after reading all the posts, I guess it's safe to say that Candy Cummings won't make it into the HOM...

Yes and no. I don't think he deserves induction for what he did in the professional leagues. However, I'm not sure about his contributions before the NA.
   55. jimd Posted: October 12, 2002 at 12:07 AM (#510794)
Here are my rough calculations of the Adjusted-Bill-James-Win-Shares for the pitchers, derived from the numbers in his book. We can compare them with the Charlie-Saeger-based numbers whenever we get them. The WS numbers do not include any NA numbers. The Career ERA+ and OPS+ do, and are from baseball-reference.com (unadjusted). Data is in a similar format as the position players. Unfortunately, I don't have the batting/fielding/pitching breakdowns. Any errors are probably my fault.

Career WS - 3 best seasons - Total 5 best consecutive - Name Birthday
- Rate WS/162G - Pct G as P (Career ERA+, Career OPS+)

Some modern reference points: (ERA+ 141 Randy Johnson, 122 Glavine)
(OPS+ 135 Brett, 117 Fisk, 87 Ozzie, 64 Hampton, 25 Glavine)

56 WS - 51, 5, x - 56 - Candy Cummings 10/18/48 (not including 4 NA)
- Rate 85.4 WS/162G - 100% P (Career 120 ERA+, 52 OPS+)
285 WS - 162, 123, x - 285 - Jim Devlin ?/?/49 (not including 3 NA)
- Rate 143.2 WS/162G - 100% P (Career 143 ERA+, 103 OPS+)
145 WS - 132, 13, 0 - 145 - Al Spalding 9/2/50 (not including 5 NA)
- Rate 74.2 WS/162G - 48% P (Career 142 ERA+, 116 OPS+)
268 WS - 50, 46, 42 - 184 - Bobby Mathews 11/21/51 (not including 5 NA)
- Rate 59.8 WS/162G - 84% P (Career 107 ERA+, 45 OPS+)
355 WS - 132, 51, 46 - 256 - George Bradley 7/13/52 (not including 1 NA)
- Rate 58.5 WS/162G - 61% P (Career 107 ERA+, 76 OPS+)

452 WS - 109, 84, 81 - 290 - Will White 10/11/54
- Rate 95.8 WS/162G - 100% P (Career 120 ERA+, 40 OPS+)
584 WS - 129, 99, 96 - 426 - Old Hoss Radbourn 12/11/54
- Rate 96.3 WS/162G - 79% P (Career 120 ERA+, 72 OPS+)
561 WS - 162, 127, 109 - 534 - Tommy Bond 4/2/56 (not including 2 NA)
- Rate 115.8 WS/162G - 91% P (Career 111 ERA+, 72 OPS+)
368 WS - 109, 61, 60 - 310 - Guy Hecker 4/3/56
- Rate 61.5 WS/162G - 48% P (Career 114 ERA+, 117 OPS+)
354 WS - 72, 54, 50 - 242 - Dave Foutz 9/7/56
- Rate 42.5 WS/162G - 23% P (Career 124 ERA+, 102 OPS+)

569 WS - 104, 81, 66 - 391 - Jim McCormick 11/3/56
- Rate 99.3 WS/162G - 92% P (Career 118 ERA+, 70 OPS+)
621 WS - 118, 82, 78 - 348 - Pud Galvin 12/25/56 (not including 1 NA)
- Rate 89.4 WS/162G - 96% P (Career 108 ERA+, 46 OPS+)
592 WS - 116, 69, 61 - 345 - Tim Keefe 1/1/57
- Rate 106.2 WS/162G - 96% P (Career 125 ERA+, 59 OPS+)
434 WS - 94, 81, 77 - 341 - Jim Whitney 11/10/57
- Rate 83.3 WS/162G - 75% P (Career 105 ERA+, 112 OPS+)
538 WS - 91, 85, 73 - 288 - Tony Mullane 1/20/59
- Rate 76.0 WS/162G - 73% P (Career 118 ERA+, 87 OPS+)

112 WS - 68, 21, 18 - 111 - Lady Baldwin 4/8/59
- Rate 102.7 WS/162G - 87% P (Career 118 ERA+, 78 OPS+)
523 WS - 82, 77, 67 - 271 - Mickey Welch 7/4/59
- Rate 93.8 WS/162G - 92% P (Career 113 ERA+, 68 OPS+)
340 WS - 100, 63, 58 - 329 - Larry Corcoran 8/10/59
- Rate 96.2 WS/162G - 87% P (Career 122 ERA+, 67 OPS+)
255 WS - 62, 57, 47 - 211 - Bill Hutchison 12/17/59
- Rate 92.9 WS/162G - 98% P (Career 112 ERA+, 51 OPS+)
661 WS - 98, 98, 60 - 373 - John Montgomery Ward 3/3/60
- Rate 41.6 WS/162G - 22% P (Career 118 ERA+, 93 OPS+)

352 WS - 90, 51, 46 - 221 - Charlie Buffinton 6/14/61
- Rate 71.8 WS/162G - 69% P (Career 114 ERA+, 71 OPS+)
491 WS - 90, 69, 66 - 316 - John Clarkson 7/1/61
- Rate 109.9 WS/162G - 97% P (Career 134 ERA+, 60 OPS+)
273 WS - 81, 65, 51 - 257 - Ed Morris 9/29/62
- Rate 108.3 WS/162G - 98% P (Career 116 ERA+, 26 OPS+)
192 WS - 63, 59, 46 - 192 - Charlie Ferguson 4/17/63
- Rate 89.1 WS/162G - 72% P (Career 121 ERA+, 122 OPS+)
406 WS - 74, 66, 62 - 308 - Bob Caruthers 1/5/64
- Rate 79.6 WS/162G - 50% P (Career 123 ERA+, 135 OPS+)

194 WS - 69, 53, 41 - 191 - Jumbo McGinnis 2/22/64
- Rate 90.4 WS/162G - 97% P (Career 111 ERA+, 60 OPS+)
330 WS - 47, 43, 35 - 167 - Adonis Terry 8/7/64
- Rate 60.0 WS/162G - 67% P (Career 103 ERA+, 85 OPS+)
160 WS - 54, 53, 32 - 151 - Toad Ramsey 8/8/64
- Rate 81.5 WS/162G - 99% P (Career 117 ERA+, 42 OPS+)
192 WS - 59, 51, 37 - 179 - Matt Kilroy 6/21/66
- Rate 81.4 WS/162G - 92% P (Career 109 ERA+, 71 OPS+)
206 WS - 45, 38, 32 - 156 - Sadie McMahon 9/19/67
- Rate 74.0 WS/162G - 99% P (Career 118 ERA+, 34 OPS+)

303 WS - 82, 51, 51 - 250 - Silver King 1/11/68
- Rate 95.3 WS/162G - 95% P (Career 123 ERA+, 56 OPS+)
333 WS - 53, 52, 47 - 223 - Jack Stivetts 3/31/68
- Rate 76.4 WS/162G - 64% P (Career 120 ERA+, 105 OPS+)
154 WS - 59, 31, 20 - 130 - Scott Stratton 10/2/69
- Rate 50.3 WS/162G - 59% P (Career 99 ERA+, 102 OPS+)
347 WS - 69, 52, 50 - 251 - Amos Rusie 5/30/71
- Rate 98.4 WS/162G - 95% P (Career 130 ERA+, 61 OPS+)

Career WS - 3 best seasons - Total 5 best consecutive - Name Birthday
- Rate WS/162G - Pct G as P (Career ERA+, Career OPS+)

Pitchers are in chronological order by birthdate.

   56. Marc Posted: October 12, 2002 at 04:46 PM (#510795)
I know you're all dying to know how I rank Rusie and Stivetts among 19th century pitchers. You'll recall my ratings of pitchers who peaked prior to 1893:

>Peak Value 1. Spalding, 2. Radbourn, 3. Clarkson, 4. Caruthers, 5. Bond, 6. Keefe, 7. Mullane, 8. Corcoran, 9. Welch, 10. McCormick

Whoever said pitching from 60 feet was a different game than 50, I agree. The only guys in Rusie's ballpark, even aside from that, would be the top 3, and given the circumstances, I think Rusie's peak would be #1. In case anybody doesn't know, it has in fact been said that the rubber was pushed back to 60 feet because of Rusie's dominance at 50 feet, and yet his ERA and ERA+ from 60 feet in 1893-94 and 97 was better than before, though his SO totals certainly declined. His 3 year peak WS of 137 is not that spectacular but his 3 year peak ERA+ total of almost 500 is tops, only Radbourn is close. His career ERA+ of 130 tops everybody except Clarkson.

>Career 1. Clarkson, 2. Keefe, 3. Welch, 4. Radbourn, 5. Galvin, 6. Mullane, 7. Spalding, 8. McCormick, 9.Caruthers, 10. White

Rusie's career value is of course considerably lower. He did get up to 3700 innings, about 20 percent less than Clarkson. He got his ten years in only by pitching 3 games in 1901. He was a hold out all year in 1896 due to the nearly criminal duplicitousness of Giant owner Andrew Freedman, pitched two more years then sat out two full years with an arm injury, and finally aborted his 1901 comeback at the age of 30 after just those 3 games. His career counting stats (eg. 246 wins, the 3700 innings) are very short of the 1880s guys and of Nichols and Young, his near contemporaries. His career percentages are fantastic, but I can't put him any better than 7th after Mullane for career. Actually that might be a bit harsh considering the 60 foot distance, so maybe 6th behind Galvin.

>Then considering peak and career as of equal weight in Total:1. Clarkson, 2. Keefe, 3. Spalding, 4. Radbourn, 5. Welch, 6. Mullane, 7. Caruthers, 8. Galvin, 9. McCormick, 10. Bond

Overall, he beats Spalding for both peak and career so how can he not be at least #3? I guess in the end I'd have to put him #2 ahead of Keefe.

As to Stivetts, I was not surprised I guess that his peak numbers are not that good, especially since they came in the weak late (1890-91) AA. When he jumped to the NL in '92 his ERA+ dropped from 147 to 116 and never again exceeded 120 in more than 129 innings. With just 203 career wins, I don't think he makes top ten on either list.

Weyhring is the other 1890s pitcher worth looking at, but his peak was also at 50 feet, mostly in a declining AA, and he is vastly overshadowed by the other 1880s guys discussed above. Hutchison had a similar peak in 1890-92 but in the NL. You could argue him for the top ten peak list, but with a career record of 183-163 and ERA+ of 111, it's not worth the bother.

Then the next round is Nichols and Young, and they will both go to the top of the list ahead of Clarkson and all the rest. So the final "19th century" list considering both career and peak:

1. Young, 2. Nichols, 3. Clarkson, 4. Rusie, 5. Keefe, 6. Spalding, 7. Radbourn, 8. Welch, 9. Mullane, 10. Caruthers, 11. Galvin, 12. McCormick, 13. Buffinton, whom I have not discussed at all, probably should have, though he is just another in a long list of guys who peaked in the pitching dominated '80s, 14. Bond, 15. Will White

Note that HOF electors picked among these guys as follows--Young in '37, Old Hoss way back in '39 and spalding the same year but as a Pioneer, Nichols '49, Clarkson in '63, Keefe in '64, Galvin in '65, Welch in '74, Rusie not until '77.
   57. DanG Posted: November 04, 2002 at 07:51 PM (#510797)
Another approach to take in figuring out who to vote for is to consider who you think should be in the HoM when we're done. That view helps you to narrow down who you are seriously considering from those who don't quite cut it.

Consider pitchers. The HoF includes a little over 30% pitchers. (Bill James is closer to 25% in the NHBA.) So if we're going to end up with 216 players, 30% would be 65 pitchers.

Looking at our early elections, we'll have six elections (1906-11) before another good pitcher enters the ballot (Nichols). Per scruff's schedule, we'll elect 15 players in those first five years. If we assume our 30% quota, then about 5 pitchers should be among those early elections. The consensus here points to Clarkson, Keefe, Radbourn, Rusie, Spalding as being those five. (Along with Ward, who is more properly classified with the shortstops.)

Soon after this, the ballot starts welcoming some of the deadball stars like Griffith (1913), McGinnity (1914), Chesbro (1915), Joss, Waddell and Willis (1916), Young (1917). The election schedule also drops to one HoMer per year for awhile. It will be interesting to see how the 19th century leftovers (Welch, Caruthers, Galvin, Mullane, Bond, McCormick) fare against these.

I think advocates of these candidates will have a tough job selling them to the electorate. Especially after our 1932 election, it will be almost impossible to elevate one of these guys ahead of newer candidates. If you can show that they rank among the top 65 pitchers all-time, they got my vote.
   58. jimd Posted: November 07, 2002 at 02:34 AM (#510798)
On Clarkson vs. Galvin:

First, there is a tremendous dilution of pitching quality going on during the 1880's. In 1878 there are 7 NL pitchers with 140 innings or more. In 1888, there are 53 (NL+AA). This has to do with major league expansion from 6 to 16 teams and with schedule expansion from 60 to 140 games. It's much easier for the best pitchers to post a good ERA+ when the talent pool is being dredged that deep. (If the procurement process was efficient - I'm not saying it was - the difference could be between only today's Cy Young candidates having jobs, and the top two starters from most of the 30 teams.)

I'd suggest re-calculating Clarkson's ERA+ based on only the top 16 pitchers (norm for 1879-81) in the league and see if his numbers are still as impressive.

Now, why isn't Galvin dominating this diluted league?

Second, the rules are constantly changing. We've focused primarily on the pitching distance change that created "modern" pitching (not really, but recognizable). There are at least three generations before that. The first generation is the Spalding era, underhand pitching "arm parallel to the body", though wrist and elbow action is permitted to put spin on the ball. For 1878, the rule is changed, "hand pass below the waist", allowing submariners; the Radbourn era begins, the underhand guys disappear quickly. For 1883, another rule change, "hand stay below the shoulder", allowing sidearm. It's not an era, because the next year, 1884 (mid-season 1885 in the AA), they give up regulating it and allow overhand; the Clarkson era begins. Apparently, the best of the submariners like Radbourn and Galvin are able to hang around and adapt due to the dearth of pitchers because of the expansions; however, they're not as dominant because of the competition with the new unfamiliar style. After a few years, the overhand ball is deemed too over-powering so the mound is pushed back, and, in turn, the short-distance guys have their troubles adapting.

IMHO, this has a lot to do with why pitcher's careers are so short during this era. None of the pitchers have long careers until they stop messing around with the pitching mechanics every few years, and then Nichols and Young appear immediately. If they changed the pitching mechanics again in 1900, would it be the Walter Johnson award?
   59. scruff Posted: November 07, 2002 at 03:19 AM (#510799)
Outstanding post jimd.

I think the points you make are very important for everyone to consider. Because of this, I'll definitely reconsider my ballot, and I'll probably lobby for the best pitcher from each 'generation' to get elected eventually. This would be Spalding, Radbourn, Clarkson and Galvin.

Again outstanding post Jim. If someone has information that Jim is missing that should make me reconsider this, please share . . .
   60. Rob Wood Posted: November 07, 2002 at 04:03 AM (#510800)
If you take the view that the pitcher in one or more of those generations merely "started the action" (i.e., let the batter hit the darned ball), then it is questionable if you want any of those pitchers in the Hall of Merit. I may be in the minority on this, but I do not dismiss the above view.
   61. scruff Posted: November 07, 2002 at 04:33 AM (#510801)
Rob, I don't think that the earlier pitchers necessarily just started the action. For one, if all they did was start the action, you'd think they could have continued to do that as the generations passed. I think what happened here is that as the skill set needed to be effective changed, so did the pitchers, which makes sense, and is the best explanation I've heard yet for the short careers and sudden disappearance of star pitchers. These guys were throwing underhand, that's why they could throw 500 innings a year. Most high school and college softball teams only have one or two pitchers also, again, because they are throwing underhand.

Also, you'd see little team to team variation in K, BB, HRA if no skill was needed. But in 1876, staffs (pitchers) ranged from 22 to 125 K. They ranged from 29 to 104 walks (although that's just one team, 2nd most was 41). HRA ranged from 2 to 9.

So you may have a point, although the K numbers ranging must mean there was some individual skill involved. ERA pretty directly correlates with H/IP, so defense was definitely the major determinate in RP. Except that we don't KNOW that DIPS applies back then. It's quite possible different pitchers were hard to hit. I suppose you'd have to study pitcher movement from 1871-78 to know for sure. If pitchers stats didn't fluctuate when they switched teams, that'd be a good sign that skill was involved.

So you definitely might have a case that Spalding doesn't belong based on his pitching alone.

By 1883 (Radbourn's heyday), K range from 253-538 per team, BB from 90-217, HR from 7-22. I'd say pitcher's definitely have skill by now.

I think there's enough evidence of there being skill to give one pitcher from each 'generation', except maybe Spalding a spot. We're talking 3 or 4 of 215 slots dedicated to pitchers from the first 23 years of the game, I think that's reasonable.
   62. Marc Posted: November 07, 2002 at 04:48 AM (#510802)
At the risk of repeating myself, the idea that pitchers simply "started the action," that they didn't "try" to get batters out, and that there wasn't a differential in their ability to get batters out, is a fantasy. The hand and wrist action to put spin on the ball is all you need to know.

Some months ago somebody said essentially the same thing, then added, you know, "like a slow pitch softball pitcher." Having played a bit of slow pitch, I can tell you there is a big difference even among slow pitch pitchers, many of whom are among the best athletes on their teams.
   63. scruff Posted: November 07, 2002 at 03:03 PM (#510803)
Actually Marc, in the first HBA, James described it as more like fast-pitch softball, and the pitcher definitely matters in that game.
   64. DanG Posted: November 07, 2002 at 03:23 PM (#510804)
But what is the "it" that James is describing? Is it early 1870's, late 1870's, early 1880's, or late 1880's? As jimd decribed, the rapid evolution of pitching conditions made it hard for pitchers who had mastered one style to adapt to the changes.
   65. MattB Posted: November 07, 2002 at 07:02 PM (#510805)
Another point about ERA+ in the 19th century:

Just from eyeballing some of the numbers, it looks to me that about a third of all of the runs scored weren't "earned."

For example, 1889 Boston (Clarkson and Radbourne are the two main pitchers). The team's staff allows 435 earned runs, but the team as a whole allowed 626 runs. (a little under 70% earned)

In 1886, Clarkson Chicago team only allowed the pitchers to "earn" 62% of the runs. And in 1885, Clarkson's best year for ERA+, only 53.4% of the team's runs were earned! (There were 497 team errors, most in the league. 2nd place New York had the fewest at 331.)

Today, the number of unearned runs is closer to 90% earned.

It would be interesting to see how closely ERA+ in the 19th century matched the number of unearned runs allowed by a team. For example, Clarkson's ERA+ was much better in Chicago than in Boston.

Certainly, bad defense can hurt a pitcher on balls in play, but could it also (or instead) have the opposite effect in extreme cases of bad defense (like Chicago in 1885), where errors drop the number of earned runs well below league averages?
   66. User unknown in local recipient table (Craig B) Posted: November 07, 2002 at 09:23 PM (#510806)
I can tell you there is a big difference even among slow pitch pitchers, many of whom are among the best athletes on their teams.

Marc, we've been through this one above... but that's right. Slow-pitch pitchers are probably responsible for 20-30% of their team's run prevention. It's a very important position. I'd argue about the same for pitchers before 1880. The introduction of gloves, sidearm and then overhand pitching, moving the pitcher's plate back, the proliferation of the sinker and the curveball (and a little later, Griffith's screwball), all aid the transition from that game to the 20th century game where the pitcher is responsible for more than half of a team's run prevention.

I hate to be doctrinaire about this... I wish I had time to pick up my study of pitcher contributions to run prevention again.

I just find the point I made above about Al Spalding's teammates to be incredibly instructive. He had an ERA+ of 142, his teammates an ERA+ of 102, and those teammates were just position players giving Spalding a day off. They weren't really pitchers. If you had a pitcher today that was 40% better by ERA+ than your position players, what would you think of him?

That doesn't happen after 1900.

Pitchers were, of course, trying to get batters out. But they didn't do it on their own, they had to depend on their defense... because there was about one strikeout per team per game (in 1876). As time goes on, that number increases... it's an indocator that pitchers are gradually taking on a greater role in the defense.
   67. dan b Posted: November 08, 2002 at 12:04 AM (#510807)
Craig B writes "Slow-pitch pitchers are probably responsible for 20-30% of their team's run prevention."

Maybe in a Sunday morning beer league where the talent level reflects the full spectrum of a normal distribution, but not in a professional league where presumably the talent is all coming off the right hand tip of the bell shaped curve. On second thought, I have enough doubts about the quality of play in the 1870's, maybe ....
   68. jimd Posted: November 08, 2002 at 01:31 AM (#510808)
[Spalding] had an ERA+ of 142, his teammates an ERA+ of 102, and those teammates were just position players giving Spalding a day off. They weren't really pitchers.

Not true that these teammates weren't really pitchers. Harry Wright had been a pitcher in his younger days, pre-NA; he served as backup 1871-74; 99.3 IP total. Jack Manning was the backup pitcher for the team in 1875; he had been a starter with Baltimore the previous year; 572 IP Career. I'm not sure what to make of Cal McVey in 1876 (was he trying to add back-up pitcher to his resume?); 176.3 Career IP. The real walk-ons (Barnes, G.Wright, etc.) had a 48 ERA+ in a tiny sample of 10.7 IP.

I don't think anybody here is arguing that 1870's pitchers are just as effective as 1890's pitchers, never mind 1990's pitchers. We're arguing the degree to which they are less effective. And I don't have a good quantitative answer to that.

But if you spend time with the rosters of the teams of that era, it's the same guys called in to do the backup pitching; it's not spread all around the team. The managers also seem to try to have a backup pitcher playing in most games. Substitution, except for injury, is not allowed until 1889, so if you think you might want to make a pitching change, the reliever has got to be starting the game at another position. And pitching changes are being made; most of Harry Wright's appearances are not starts (7 starts in 35 appearances) and Spalding is in the outfield for each appearance (1871-74), starting or relief, so he's not being lifted for injury, and he's not being rested (in the sense of given the game off). If pitching involves no skill, then why swap pitchers? If almost any player can pitch, then why sacrifice some offense to have a relief pitcher playing?

   69. jimd Posted: November 08, 2002 at 01:43 AM (#510809)
That should really say "Some managers seem to try...". Not all managers seem to employ this strategy, perhaps because they lacked a good hitting reliever, perhaps because their starter was more reliable (less variance start-to-start), perhaps because they had a big bat they were trying to hide in right field (the usual place to hide the reliever, unless he could field a little, too).
   70. User unknown in local recipient table (Craig B) Posted: November 08, 2002 at 03:51 AM (#510810)
If pitching involves no skill, then why swap pitchers?

Probably so his arm doesn't fall off. :) Even pitching 1870s-style is hard work.

I agree we're only arguing as to the degree which pitchers are less effective, though. It's an interesting question... my mind isn;t closed by any means. I need to do more work on it.
   71. scruff Posted: November 08, 2002 at 05:25 AM (#510811)
Craig, I don't know how hard work it really was. 1870s was fast-pitch softball. Most colleges/high schools have 1 or 2 girls doing all the pitching, one girl does it all come playoff time. I don't think underhand really strains the arm at all. But if you've seen fast-pitch softball, you know that there is definitely skill involved.
   72. dan b Posted: November 08, 2002 at 09:13 PM (#510812)
If it was fast pitch softball, where are the strikeouts?
   73. jimd Posted: November 08, 2002 at 11:52 PM (#510813)
I've never played or watched much softball so I really can't say much about it.

That said, a few things that might reduce strikeout rates. I don't think there is a mound yet (not sure when that was introduced). Foul balls are not strikes (not until 1901). Foul balls caught on the first bounce are outs (eliminated during 1870's, I think). No distinction between foul balls and foul tips (not until 1889), also catcher stands about 10 feet behind batter during this era. The strike zone is small (batter's choice, either knees to waist, or waist to shoulders, combined in 1887). The first pitch is not called either a ball or a strike (swing at your own risk, not sure when that was eliminated). The first called strike 3 is just a warning (1876-1880).

The catcher probably catches a few more foul outs because he's in better position for foul pops, not being in a crouch or wearing a mask. OTOH, there's lots more passed balls.
   74. OCF Posted: December 02, 2003 at 06:11 PM (#510814)
Thread bump.

At the moment, I only have questions, not answers. It might be nice to reactivate this thread as a place to collect comments. That, or we could create a new thread. What are my questions?

1. The circumstances for pitchers in the 1900's were very different than those for pitchers in the 1880's. How should we fairly compare McCormick, Welch, Caruthers and the other leftover pitchers from the 80's to Griffith, McGinnity, Willis, Waddell, and the flood of others who will come behind them?

2. How does one rank the short-to-medium career pitchers with medium-to-high peaks of the 1900's against each other? Snippets of systematic comparison have been coming out on ballot discussion threads - the more of that the better. And where will Eddie Plank fit into this?
   75. OCF Posted: December 02, 2003 at 06:19 PM (#510815)
Selected references from the 1915 Ballot Discussion thread:

Paul Wendt #95
Chris Cobb #106, #130, #148
Marc, #131
   76. Howie Menckel Posted: December 02, 2003 at 06:42 PM (#510816)
OCF,
I think it partly comes down to a proportionality issue.
Do we have enough pitchers overall? Do we have enough selections from a particular era overall? Do we have enough pitchers from a particular era?
Then you look at the level of play. Sometimes extra players at a slot are deserving (see SS with Davis and Dahlen this year), sometimes no one stands out for a number of years (very few of these 1900s pitchers may deserve enshrinement).
I think we did well with the 1870-1890s pitchers, and we have enough overall at the moment. In a couple of 'years' we likely will feel as if another pitcher belongs (aside from no-brainers like Young obviously).
If Stovey and Bennett get in, then whether a Caruthers or a McCormick have any shot depends on whether voters tink that the 1880s deserve ANOTHER player. Those who think the decade has bagged its limit may look elsewhere.
   77. jimd Posted: February 28, 2004 at 04:05 AM (#510817)
Long post warning!

Keefe vs Welch. Both players were born on holidays, Keefe on New Year's Day and Welch on the 4th of July, 2.5 years later. Despite being older, Keefe would also last a little longer, pitching in 1893 after Welch was done. Neither could hit when compared to real hitters, though both were better than the average pitcher of the time, Welch being a little better than Keefe (.226 to .217 EQA). Keefe was an average fielding pitcher with a fluke "Gold Glove" candidate season in the 1883 AA; Welch was not good (BP rating 89 vs 96 for Keefe).

The "delta" numbers at BP. Keefe may have had a small ability to reduce hits on BIP relative to his teammates (mostly Welch, though he was also somewhat negative); I don't think it's statistically significant, though it may be close. OTOH, Welch seems to exemplify "pitching in a pinch" when it came to run prevention, giving up 10% less runs than the components would indicate. His Delta-R (runs) is phenomenal, better than Mathewson's (though not as good as Spalding's rate). It didn't translate to Wins though, where he was essentially neutral for his career. Keefe seems to have struggled with delta-W early; but he seems no different than Welch after their reunion on the Giants.

The following table is an experiment. I've taken the lesser IP for the pair each season they were on the same club, and I've scaled back the W-L record of the other pitcher to be proportional to that IP. The 'K' or 'W' next to it indicates who actually pitched that amount of IP. The idea was to minimize any influence from unequal opportunities with varying quality ball-clubs (Welch pitched more with mediocre Troy, fairly even after that though it varied from year to year.) The rates did not need to be modified, naturally. (NRA is like ERA but also counts unearned runs.) ExtraIP indicates the magnitude of any imbalance in IP for that year, and who pitched the extra innings.

----- ---- --IP- * ! KWL WWL ! K-ERA W-ERA ! K-NRA W-NRA ! ExtraIP
   78. jimd Posted: February 28, 2004 at 04:29 AM (#510818)
It didn't translate to Wins though

I didn't say that very well. Delta-W is based on runs. So all the runs he saved relative to his component stats translated into Wins typical for the runs he actually gave up. Unlike Griffith, he did not get more Wins than the runs would tend to indicate. (Apparently, he's getting out of jams, stranding baserunners all over the place and giving Jim Mutrie grey hairs, but that's not the same as consistently squeezing out close wins, or losing them either.)
   79. Chris Cobb Posted: February 28, 2004 at 03:48 PM (#510819)
jimd, could you confirm that my sense of the meaning of WARP's deltas (hits, runs, wins) is right?

Delta h -- did the pitcher give up more or fewer hits than his team defensive support would lead us to expect?

Delta r -- did the pitcher give up more or fewer runs than his number of hits allowed, walks allowed, etc. would lead us to expect?

Delta w -- did the pitcher win more or fewer games than his runs allowed would lead us to expect?

So these are really three independent places where pitching skill or pitching luck would show up. A pitcher with only a great delta h saves a great number of hits on balls in play, relative to his team defense, but then has normal run totals and win totals following from those lower hit totals. A pitcher with only a great delta r total is average on allowing hits in general, but by luck or clutch pitching, he allows fewer than expected runs, and his a win total consistent with those runs allowed. A pitcher with only a great delta w is average on hits, average on runs, but by luck or clutch pitching avoid giving up runs when the game is on the line and so win more games than expected.
   80. jimd Posted: March 01, 2004 at 06:42 PM (#510820)
Great summary, Chris. That pretty much sums up my interpretation of those numbers. (With one quibble...)

A pitcher with only a great delta w is average on hits, average on runs, but by luck or clutch pitching avoid giving up runs when the game is on the line and so win more games than expected.

Wins is a small enough sample that other factors such as Run Support may also play a role.
   81. Paul Wendt Posted: March 13, 2004 at 08:57 PM (#510821)
For what it's worth, Griffith and Walsh have sponsors at baseball-reference. McGinnity, Waddell, Joss, and Brown do not, and the Iron Man is for sale cheapest.
   82. Dag Nabbit is part of the zombie horde Posted: April 04, 2004 at 03:55 PM (#510822)
Retrosheet just listed all starters for every 19th century game ever. So here's the RSIs & adjusted W/L for Bob Caruthers & Mickey Welch:

Bob Caruthers - since he's BC, I'll include his OPS+ for the year in parathesis (figure out how much of his run support is attributable to himself on your own - your guess is as good as & probably better than mine). FWIW, in 1888, even though his stats list him as having 43 GS, I find him 44 times in the game log. Here he is:

1884..118 (113)..6-3
   83. Dag Nabbit is part of the zombie horde Posted: April 04, 2004 at 04:00 PM (#510823)
Oh yeah - forgot one - some guy named "Denton"

1890..101
   84. Dag Nabbit is part of the zombie horde Posted: April 04, 2004 at 04:11 PM (#510824)
Oh, a few things I feel obligated to mention:

Occastionally in these early days, the number of times a pitcher is listed starting in the gamelogs is at odds with the number of starts they have listed in traditional stat logs. For example, Cy Young's only got 15 starts in 1890 & 45 starts in 1896 (one fewer than his official total both times). Normally, I assume that retrosheet's right, but that 1890 thing gives me pause because years ago I looked through microfilm of 1890 for the Chicago Tribune & Cleveland Plain-Dealer & found 16 starts for Young in them. Maybe they made an error then that retrosheet caught, but I don't like the looks of it.

Also, the game log gives Tony Mullane 58 starts in 1882, not 55. And 23 in 1889, not 22. Jim McCormick loses a start in both 1882 & 1883.
   85. John (You Can Call Me Grandma) Murphy Posted: April 04, 2004 at 06:34 PM (#510825)
Good stuff, Chris!

The one caveat that I would mention is that there is still big difference between 300 during the 1880s than 300 during the Deadball ERA. IOW, I'm still not loading up my ballot with the 19th century guys still out in the cold. The only one that I might consider is Welch, but I'm not 100% sold on him as of right now.

Come to think of it, "300" as a number means zip to me.
   86. Dag Nabbit is part of the zombie horde Posted: April 05, 2004 at 03:33 AM (#510826)
The one caveat that I would mention is that there is still big difference between 300 during the 1880s than 300 during the Deadball ERA. IOW, I'm still not loading up my ballot with the 19th century guys still out in the cold. The only one that I might consider is Welch, but I'm not 100% sold on him as of right now.

Uh-huh. That's been going through my mind as well. Five pitchers - Galvin, Welch, Keefe, Radbourn, & Clarkson - had their careers centered in the 1880s who went on to win 300 & a sixth - Mullane - would've done it had he not held out a year. And beyond them you get Jim McCormick over 250 wins. That means who either had an unusual amount of pitching talent get centered in that one decade or there was something else afoot. Looking at how it's structured - these were the pitchers who blossomed as/after the scehedule got legthened to something approaching modern lengths but before the box was replaced with the mound. I think that may have something to do with it. Still, this will help some guys on my ballot, most notably Welch. Right now I'm thinking of him as something akin to a pitching version of Van Haltren - solid for about a decade, though never the best at what he did.

More players:

Pud Galvin:
   87. Dag Nabbit is part of the zombie horde Posted: April 05, 2004 at 06:45 PM (#510829)
Chris J, again, many thanks for your work on RSI and Adj W-L. Let me make sure I understand: are the data in the files section

using runs allowed instead of earned runs


Right.

adjusted for league avg offensive support (excluding pitcher's hitting)

Right.

adjusted for park effects

Right. Specifically, batter park effect (since we're adjusting his team's hitting)

not using actual W-L records

Well, it starts with actual W-L records. Columns F & G in the yahoo club file are actual W-L records. Columns A & B are adjusted. J & K are the W/L records derived from pythaging their actual RA/9IP & actual run support. Columns M & N are pythags of actual R/9IP & league avg run support. The difference between J & K and M & N are how I get adjusted records from actual records.

not adjusted for any 'league quality'

Right.

not adjusted for defensive support ?

Right.

RSI's only adjusting for run support. Not defense, league quality, the pitcher's own hitting ability, etc. You pretty much got it.
   88. Dag Nabbit is part of the zombie horde Posted: April 05, 2004 at 06:48 PM (#510830)
Chris J, again, many thanks for your work on RSI and Adj W-L. Let me make sure I understand: are the data in the files section

using runs allowed instead of earned runs


Right.

adjusted for league avg offensive support (excluding pitcher's hitting)

Right.

adjusted for park effects

Right. Specifically, batter park effect (since we're adjusting his team's hitting)

not using actual W-L records

Well, it starts with actual W-L records. Columns F & G in the yahoo club file are actual W-L records. Columns A & B are adjusted. J & K are the W/L records derived from pythaging their actual RA/9IP & actual run support. Columns M & N are pythags of actual R/9IP & league avg run support. The difference between J & K and M & N are how I get adjusted records from actual records.

not adjusted for any 'league quality'

Right.

not adjusted for defensive support ?

Right.

RSI's only adjusting for run support. Not defense, league quality, the pitcher's own hitting ability, etc. You pretty much got it.
   89. Dag Nabbit is part of the zombie horde Posted: April 05, 2004 at 07:08 PM (#510831)
Oops.

At any rate - Old Hoss time. He loses a start in 1882, gets it back in 1885, then loses another one in 1887, so it's out of 502 GSs:

1881..119
   90. Dag Nabbit is part of the zombie horde Posted: April 05, 2004 at 07:12 PM (#510832)
Not that anyone's paying much attention to him - put Jack Chesbro's RSI moves from 105.14 to 105.55. He had RSIs of 89 & 121 in his first two seasons.
   91. Dag Nabbit is part of the zombie horde Posted: April 05, 2004 at 07:34 PM (#510834)
TomH - right. The difference between columns J & M is equal to the difference between columns A & F.

John Clarkson:
   92. Dag Nabbit is part of the zombie horde Posted: April 05, 2004 at 07:55 PM (#510835)
Kid Nichols - gains a start in 1895 & another in 1897, so it's out of 563 GSs:

1890..86
   93. Dag Nabbit is part of the zombie horde Posted: April 05, 2004 at 10:03 PM (#510836)
Amos Rusie - the last HoF for me to do. He loses a start in 1890 but gains it back in 1892. His RSIs:

1889..115
   94. jimd Posted: April 06, 2004 at 12:50 AM (#510837)
Wow. More fascinating stuff from Chris J.

In a vaguely related vein, I spent some time this afternoon going through the new 1880's retrosheet files looking at head-to-head matchups. The results follow:

Most common matchups and approximate W-L records:
   95. Jeff M Posted: April 06, 2004 at 01:56 AM (#510838)
Appreciate your hard work ChrisJ and jimd
   96. Marc Posted: April 06, 2004 at 02:21 AM (#510839)
>Mickey Welch HTH
   97. Dag Nabbit is part of the zombie horde Posted: April 06, 2004 at 02:43 AM (#510840)
OK - ADJUSTED career W/L records for (takes deep breath) - Griffith, Waddell, Willis, McGinnity, McCormick, Mullane, Keefe, Galvin, Radbourn, Clarkson, Nichols, & Rusie:

Clark Griffith:
   98. Dag Nabbit is part of the zombie horde Posted: April 06, 2004 at 02:52 AM (#510841)
<i>Note the use of individual matchups between certain teams.
   99. Chris Cobb Posted: April 06, 2004 at 03:01 AM (#510842)
Based on a cursory glance at retrosheet, I'd say that Welch was usually getting the tougher ones. Great stuff, jimd.

That would be consistent with their RSIs from the New York years. Great stuff, both Chris J. and jimd!
   100. Dag Nabbit is part of the zombie horde Posted: April 06, 2004 at 06:13 PM (#510844)
TomH - You're using the numbers from the group file, right? That's now a little outdated. Rube's one of those guys that spanned the centuries - he pitched a little before 1901 & a lot after. So I didn't adjust his W/L record for the pre-1901 years because I had no RSI data for those years (until this weekend). When I say he's underachieved a few posts ago, I'm using the new, fully-adjusted data. Info for all pitchers whose careers began before 1901 are a little outdated on the group spreadsheet.

If you want, you can figure all 4 W-L records from post #112. For example, at the end of Waddell's bit I wrote:

Total: 205-131. Gains 12 wins. 199 FWP - good for 50th on my list. Underachieved by 6 games.

Well, his adjusted record's 205-131. Got that. That's a gain of 12 wins so his real life W/L should be 193-143. And it says he underachieved in real life by 6 games, so he should've gone 199-137. And that means a pythag of his real RA/9IP & league average run support end up with a record of 201-125 because the difference between the 2 pythag'd W/L records is the same as the difference between real & adjusted. (In fact, the difference between the two pythags is what determines the difference between real & adjusted).
Page 1 of 6 pages  1 2 3 4 5 6 > 

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Adam S
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 1.0885 seconds
68 querie(s) executed