1. Jack Glasscock (3) - Clear cut #1 this year. The wait should finally end.

Did any MLBers take off 1898 to fight the war with Spain?

I know it was posted somewhere else, but can we get a list of the new eligibles for 1904 here?

The adjusted WS I've calculated for Mike Griffin look like this (Joe's will probably differ slightly):

Career -- 294

This is the list RobC posted last year of the new eligibles.

1904:

As a companion to Rob's list, here's mine which shows a little basic info on the leading candidates:

Win Shares - WARP3 - Rookie Year - Position - Year Died

***1904 (July 6)

Originally posted by me on the Cap Anson/Roger Connor election thread:

Of the new guys, Billy Nash probably makes my top ten (better than Williamson) and Clements should make the bottom of my list. I have no idea where Rusie will be (or where Galvin and Spalding will be on my ballot).

I like Joyce and Griffin, but their careers are too short.

Also, note that my eligibles list includes Kilroy and Wilmot, FWIW.

With Rusie now a candidate, it brings up the issue of the pitching distance change from 1892-93. The change signaled the end of many careers, but how many pitchers successfully bridged the gap?

Here is a list of the nine pitchers who won 60 games on each side of the gap:

Wow, this is my toughest ballot yet! I've incidently switched my peak value assessment to best 8(not necessarily consecutive) years. I weight this roughly even with overall career value. For pitchers, I still use Best 5 due to the 'shooting star' nature of most pitchers at this period. I still make reasonable adjustments for guys like Start, Pike, Spalding, etc.

1)Glasscock- I think this is finally Pebbly Jack's year. He's at the top of both my career and peak lists of eligible hitters.

Pitching across the distance divide is a "toolsy" argument, it has nothing to do with value. I think the apparent inability of players to pitch across the divide is caused mainly by the fact that the stars of the '80s pitched too many innings. There is no reason to believe that Clarkson or Keefe could not have done what Stivetts or Weyhing did, but at their own higher level, but for all those innings. There are enough pitchers who pitched across the divide to suggest it did not require "special tools." So, value is value is value. And Rusie ranks highly on value, regardless of his timing.

Prelim ballot:

Prelim:

Semi-regular HoMers per year, played at least half the games (for hitters) or pitched at least 100 innings, 1871-1890.

1871 - 4

There were twice as many Major-leaguers, so why not?

Maybe I will try to participate.

To start with, what about Rusie? He's the original hot, wild kid, a member of the same tribe as Feller (and Feller wasn't done at 27).

Here's one name that's hovering around, fogging my vision: Bill Hutchison. Hutchison was a rookie in 1889, same as Rusie, but he was 11 years older than Rusie, and no, I don't know why he didn't make it to the majors until he was 29. For pitching, the 1890's start out as an extension of the 1880's, but there was that big change in pitching distance in 1893. Ace pitchers of the 1880's started 60 to 70 games a year, and eye-popping single seasons were possible (Radbourn, King, et al.). By late in the 1890's, ace pitchers were only starting 45 or so games a year.

In 1890, 1891, and 1892 (before the change), Rusie had an 1880's-style workload: 181 starts (60 per season), 1581 inninng (527 per season). Those weren't his best three seasons, but they acount for a lot of the bulk in his numbers, including 45% of his career starts. But Hutchison was pitching even more than Rusie, with about the same ERA+. Here are some 3-year totals for 1890-1892:

Hutchison 1786 IP, 120-78

Because, assumedly, the Top 10 or 20 players would have been the ones playing irrespective of how many teams there were (as long as there were at least 2 or 3).

Doubling the number of leagues merely increased the number of below average (from a single-league perspective). The number of greats should remain fairly constant over short periods of time.

OCF,

Rusie sat out 1896 due to a contract dispute. He was fined for violating team rules in 1895. He demanded that the amounts of those fines be added in to his 1896 contract. He didn't get it.

He missed everything after 1898 because he tore his shoulder muscles on a pick-off move. (I've never heard of that happening to anyone else!)

You're assuming that all of the greatest players in the country played in the NA. It was the best available league, but that doesn't mean all the best players were there. This can be assumed nowadays, but was less likely in the 1880s and even less so in the 1870s.

Billy Nash:

Best major league third baseman: 1888, 1889, 1892, and 1893.

I'm surprised he's not receiving more consideration here.

Jack Clements:

Best major league catcher: 1891 and 1895.

14)Williamson- I've officially flip-flopped on the Williamson-Sutton debate. They have similar career value, but it took Sutton 6 more years to compile it.

They only have similar career value if you don't count Sutton's NA work and don't adjust for schedule. If you do that, Ezra is far ahead.

Browning--moving up, 164 OPS+ and his defensive numbers are not as bad as his rep

Those BP numbers are suspect, at best. They have Billy Hamilton as inferior to Sam Thompson (which flies in the face of their reps and common sense). You have to question what they are doing with DERA, too.

Thanks Matt for:

1871 - 4

John, I certainly do question DERA. And the 164 is just one number among many--I looked at a lot of numbers, I quoted one.

As to Thompson and Hamilton, I believe they are pretty close. Common sense does not help much, you gotta dig a little, as I know you have done. One was a mediocre CF, the other apparently an outstanding RF. Does common sense say which is better? Depends on the particulars. One gets on base and scores the runs, the other bangs them in. Which is more valuable? Depends. Analogous to Henderson and Mattingly (at one time, not for career). I would expect both to go in fairly easily, though there is now evidence that Thompson will in fact have to wait a while. I like Thompson better than Browning, Browning better than Stovey, Hamilton certainly better than Browning or Stovey, and in the end Hamilton and Thompson are the only ones among the four who are obvious HoMers. But don't ask me which was better until I hafta.

Nash went from 222 WS to 278 when adjusting for season length, at the same rate we'd get Griffin around 307.

I don't know where you are getting 307, Joe. I have 290 for him (not that I could be wrong myself).

Nash had a much higher peak and played a position that was tougher on the body than centerfield (so the hot corner guys played less games). I think Billy wins pretty easily.

BTW, I don't have Griffin as the best major league centerfielder for any season.

'That there were players not in the NA who were as good or better than the best players in the NA is an opinion, not a fact. Name some!?'

I haven't studied the league strength issue myself, but others here have. I was under the impression that the consensus opinion was that the NA and AA were of similar strength, the only difference being that the NA was the best league available at the time, while the AA was not. If all of the best players at the time were in the NA, why was it not a stronger league? If all that were added in the 2-league 1880s were below average and replacement level players, why was the NA not as strong as a combined 1880s NL-AA league would have been? 2 possibilities: 1)There were less great players overall in the 1870s- Not likely, but there is a strong possibility that less of the people with great baseball talent were actually playing baseball. There is also the possibility of a simple statistical anomaly. 2)A smaller percentage of the 'great' players were actually in the NA.

One was a mediocre CF, the other apparently an outstanding RF. Does common sense say which is better?

My common sense statement was only directed at the difference between the speed of the two players, not anybody's opinion. Common sense tells me that the speedster in center is going to be (though not always) better than the slower guy in right (not that Sam didn't have speed himself).

I have Thompson as an easy top fiver (without actually ranking him). Without a doubt, the best centerfielder for the second half of the nineties. He was also the better offensive player over Big Sam. The only real negative is the shortness of his career.

As for their fielding, Hamilton had a fine rep, while Thompson didn't. I'll stick with the WS numbers over BP.

There are no doubt many others (possibly Lip Pike if I must name 1) who had a few not-so-great years at the end of their career in the NA whose best years are unrecorded and might be in otherwise.

Huh? He was the best centerfielder for the NA (which is recorded). He was also terrific in 1876. I'm still not sure why he left the game in '78 when he was still a fine player, though. At any rate, it was the "increasing" competition.

I have him low on my ballot, so I'm not saying he belongs, but he deserves much more consideration than has been given so far.

Carl, since you have Start, Spalding and McVey on your ballot, it doesn't seem that you are not giving the '70s guys enough attention.

As for Pearce and Meyerle... :-)

OK Lip Pike was a bad example. My point was that there was less talent in the 1870s playing professional ball than in the 1880s. This was probably true across the board because, if it was only true in the lower levels of talent, the NA would have been a far stronger league the the NL or AA in the 1880s. It was not.

Carl, I agree with most of your analysis. The lack of '70s players in the HoM is due to a disinclination by many voters to consider subjective evidence from the '60s.

But then there's the issue of quality of play in the '70s vs. the '80s that also contributes to downgrading the '70s guys. I call that the timeline issue, since it is a different issue than quality of play of AA vs. NL at the same time. Those who use a timeline adjustment would be as opposed to the "pennant is a pennant" school of thought.

Some use a particularly steep timeline relative to the '70s. This is not unrealistic based on the assumption that players in the '70s had had less time on the field, fewer games, fewer practices, poorer coaching, etc. etc. etc., and so if plopped down in 1920, say, they might not do as well as a player from the '80s or '90s who got more time on the field, better coaching, more opportunities to hone his skills. It also reflects the demographic argument though the change in population and demographics from the '70s to the '80s was small and not as important as the improvement in training, playing opportunities, etc.

But the pennant is a pennant crowd, of which I am a member, would say, so what? The objective in 1875 was to win the 1875 pennant, not to win the 1920 pennant.Nevertheless, you are probably being generous when you say the NA was equal to the AA. In absolute terms the AA was probably better on a timeline basis. On the basis of "a pennant is a pennant," the NA would rate higher because its was the toughest pennant in the world, the AA's was not.

I think the crux of this discussion is this: the quality of play based on Davenport's numbers (and we have seen some comparisons here as well of how well players played who played in the AA and in the NL) goes across all levels of players. Well, not all levels because the very best players did NOT generally change leagues. The guys who changed leagues were journeymen or worse, as a rule. So the league quality numbers are not really all that helpful comparing superstars, they refer rather to the quality of the superstars' opposition.

I myself use a discount but based on the foregoing, I (and you) have to be very careful. Would Pete Browning have hit 164 OPS+ in the NL? Probably not, but is the appropriate discount for Pete Browning the same as the average discount across the entire leagues? Probably not. So what is appropriate? Take a guess. I don't think we know, we just take a guess (and I generally guess the average, even knowing it is probably wrong, but probably less wrong than any other guess I might make).

So put it all together and the pennant is a pennant perspective, which I share, is that Al Spalding rates very highly because he dominated in his own time, forget the quality of the opposition. The timeline crowd says, no, the quality of the opposition was everything (e.g. Bill James). OTOH, the quality of Pete Browning's opposition DOES matter even to the pennant-is-a-pennant crowd. Except that we also know that the quality of the opposition is not the same thing as the quality of the player being evaluated. So...

There are a bunch of issues intertwined here and that of course is what makes it interesting. And while we all may disagree and sometimes vociferously about the evaluation of a given player, even the hardest core voter who thinks he has all the answers must agree that there is vast legitimate room to disagree.

Did any of this make sense? On second thought, don't answer that.

'Carl, since you have Start, Spalding and McVey on your ballot, it doesn't seem that you are not giving the '70s guys enough attention. '

You're probably right, John. I'm having a bad week so far and I'm probably overreacting.

And while we all may disagree and sometimes vociferously about the evaluation of a given player, even the hardest core voter who thinks he has all the answers must agree that there is vast legitimate room to disagree.

Yup.

Been trying some different measures of league strength, and to be honest, I'm not coming to definitive conclusions. I've taken all of the NA seasons, the AA seasons, the PL season, the UA season and the NL seasons through 1891, and examined the park-adjusted runs created of every player with at least 50 at-bats in those seasons. I wanted to see what the standard deviations looked like for each season, to see if they indicated a pattern. Because a large standard deviation might be large because the mean RC is large, I also normalized everything to a standardized season in which the average team scored 750 runs.

It is still in process, but so far, inconclusive. The AA numbers look about the same as the NL numbers during the years they both existed. There's a big shift from 1881 NL to 1882 NL, indicating that the league quality of the 1882 NL changed as some players went over to the AA. There's also a blip in the rule change years.

I did an earlier study comparing the O2PS+ of hitters who played fairly regularly in both the AA and the NL during the 1880-1893 years, to see if they performed better in one league than another. I didn't do it on any kind of season by season basis, because there aren't enough samples for each year. But the aggregate of the seasons showed the O2PS+ numbers of the AA hitters to be about 4% lower. Obviously, some AA seasons are stronger than others, so I don't recommend any sort of across the board discount. I did not study the pitchers numbers to determine whether maybe the AA pitchers were 4% stronger, thus affecting the AA hitters. Not sure how to do that study.

I'm skeptical of the numbers we've seen posted about Davenport's AA discount. They are huge and we don't know how they are derived. Plus, the NL during 1882-1891 also lost some strength to the AA, but none of us applies an NL discount during this period, to my knowledge.

But your question was NA-directed. There's no serious change in the RC numbers I've produced between 1875NA and 1876NL. There is quite a bit of change from 1871 to 1875, but I haven't yet figured out what it means. I think it's fair to say that the early years of the NA pulled from a smaller geographical base than the later years and the earliest years of the NL, but whether that is significant or not, I don't know.

But your question was NA-directed. There's no serious change in the RC numbers I've produced between 1875NA and 1876NL.

I assume the biggest disparity is between 1871 and 1872.

The thing to point out is that as the competition became tougher, the same names were still on top (even in 1876).

Marc wrote: "As to Thompson and Hamilton, I believe they are pretty close."

The Hamilton vs. Thompson debate usefully anticipates the 1907 ballot discussion, when Hamilton comes on, but it also raises issues that voters should now consider when evaluating Thompson in relation to Harry Stovey. As a FOHS and an EOST, I'd make the case that Stovey rates far ahead of Thompson (as does Hamilton, who ought to be a first-ballot HOMer) because his speed and his ability to get on base enabled him to manufacture runs.

_Lots_ of runs scored in this period without being "batted in," and I think the hitters who got on base and got themselves around (with the help of defensive miscues) should get credit for that. Here's an example to put this in perspective.

Harry Stovey played for a _terrible_ team when he started his career in the National League -- the Worcester Ruby Legs. In 1882, his last season with the team, their record was 18-66. He was essentially the only good player on the team. (Small wonder he jumped to the AA.)

On this terrible team, Stovey managed to score 90 runs in 84 games. The team as a whole scored 379 runs, so you can see that Stovey was not getting a lot of help. He got even less help than that number suggests: his team was credited with 279 rbi. Stovey was third in the league in runs, ninth in the league in OBP. Given the fact that his teammates were not knocking him in, the fact that he ranked higher in the league in runs scored than he did in OBP suggests not only that OBP was very important but also that speed and savvy on the basepaths were just as important. But _none_ of that valuable baserunning skill and shows up in OPS.

The two players who scored more runs than Stovey were George Gore (99 runs) and Abner Dalrymple (96), both of whom played for Chicago, which was loaded with great players, won the pennant with a 55-29 record, and scored 604 runs overall, with 451 rbis. Gore reached base 146 times on hits and walks (plus an unknown # on errors), and scored 99 on a great team. Stovey reached base 126 times on hits and walks (plus an unknown # of errors), and scored 90 runs on a terrible team. That's a .692 runs per recorded time on base for Gore, .714 for Stovey. That's an impressive number, and it really suggests to me that Stovey was among the greatest baserunners of his time, when that meant a lot (in ways we can measure if we look for them), in addition to being an outstanding hitter.

Speed matters somewhat less in the 1890s game and unearned runs are less important, but in 1895 when Hamilton and Thompson were teammates, Philadelphia scored 1068 runs on 931 rbis, so 10% of the team's runs still come from sources other than rbis. For this team, Hamilton scored 166 runs. Thompson drove in 165, with both Hamilton and Ed Delahanty batting ahead of him (I think). I don't see how one can rate Thompson ahead of Hamilton on the basis of this kind of evidence. In the modern game, rbi leaders are almost always far ahead of run leaders, but even in the 1890s, runners got around on their own.

As a final note, I should add that 1895 Philadelphia wasn't particularly good at manufacturing unearned runs. Hamilton reached base 304 times and scored 166 runs with Delahanty and Thompson behind him. The run manufacturers that year were the Baltimore Orioles, who scored 1009 runs on the strength of 828 rbis. Willie Keeler reached base 264 times and scored 162 runs for that team. Even in the 1890s, a considerable amount of offensive value shows up in places where we don't typically look for it.

One thing that is very different between the 1870's and the 1880's is the age demographic of the players. The NA has very few older players. This isn't that surprising given the economics of the game pre-NA; it wasn't easy to make a steady living at it, so players often held down real jobs also. As they got into their late 20's, they often were forced to make a decision between playing and devoting more time to advancing their other career, which probably won in most cases. There were 3 regulars who were over 30 in 1872 (Harry Wright, Dickey Pearce, and Wes Fisler, about 5%); there were 27 in 1883 (18%) and 48 in 1899 (30%). (MLB today is about 45%.) As the game got more stable, the careers got longer for players other than the super-stars.

I don't know whether the lack of an older generation accounts completely for the difference in quality between the NA and the NL of 10 years later, but I think it goes a long way towards explaining it. Keeping the best of the "elders" playing deepens the talent pool considerably.

Preliminary Ballot:

1. Al Spalding

Trying to summarize where we are before attempting a preliminary ballot:

To oversimplify, how about organizing the candidates into four groupings:

1. Done and gone by about 1880.

Well, Jason, I wouldn't have used my argument that way, but I do agree with much of what you say about the 1860's.

What I would say my demographic argument implies is as follows.

Given that 45% of current major-league regulars are over 30, we could probably get an MLB of roughly equivalent quality with very few over 30 players if it only had 16 teams. By analogy, the 1870's could be on a rough parity with the 1880's because of the doubling of the number of teams. The 1880's may conceivably be weaker even because they are not at 45-50% veterans, but only about 20-25%, spreading thinner the younger portion of the talent pool. To convince me that the 1880's were significantly better (without the stats to show it) requires demonstrating that the talent procurement process was significantly more efficient. I've seen no evidence presented for that; in fact, if the NL was respecting minor league reserve clauses, the process may have been less efficient than the completely open unrestricted talent market of the 1870's.

Joe-

Respectfully, even where James readily acknowlegdes the limitations of Win Shares in his NHBA, he still rates Radborne 45th all-time while Galvin doesn't even crack the top 100 -- i.e. he thinks that there are at least 65 Pitchers worse than Hoss who are better than Pud.

As I wrote above, I really like Galvin. I think he's much better than James seems to. Galvin's a HOMer in my book; easily among the 200 greatest players ever. Radborne, however, is at least one level better -- he's among the 100 best ever.

I also think that Hoss is clearly superior to Galvin... but I certainly don't knock Galvin. I haven't done my prelim ballot yet, but I imagine that they will be my number one and number two, with Glasscock coming it at 3.

(I picture Rusie around 6. We'll see how it comes out.)

ed - warp3 says Galvin has the better career. He had 8 more wins above replacement. Period. That what the number means. Now, there are reasons to rank Hoss above Pud (not that I do) but that aint one of them. There are plenty of measures in Radbourn's favor without misrepresenting WARP3. If Galvin's extra 1400 innings had been at replacement level, he wouldnt have been able to move past Hoss. Plus, considering the 5 year peak by WARP3 has Pud better, Im not sure the extra 1400 innings need to be discussed to get Pud ahead of Hoss. On my ballot, it is Galvin's peak that moves him away from a virtual tie with Radbourn into the group with Rusie and Hardy and Thompson.

ed wrote: "? Joe wrote:

I'll be traveling until early next week, so I thought I'd toss out my preliminary ballot before I go. Mostly everyone moves up a couple of slots, as there are no top-ballot stars newly eligible this year.

1904 Preliminary Ballot

1. Jack Glasscock (3). With Anson and Connor gone, Pebbly Jack floats to the top. His defensive work is phenomenal, and he's a fine hitter to boot.

John -- I was just guessing for Griffith (I thought I made that clear

You did, Joe. I was trying to understand how you got the 307 number as an approximate. Seemed like a funny number for a seat-of your pants estimate. That's all. :-)

We have two centerfielders who were contemporaries late in the 19th Century. Outfielder A will likely get into the HOM within the first five elections he is eligible, while Outfielder B will linger in the bottom five spots of our ballots.

Outfielder A: 7827 PA, 122 OPS+, A+ defense

Duffy and Griffin. Hmm.

I'm not sure Duffy will get in quite so fast as all that, but he's significantly better than Griffin.

Reasons:

1. 1000 AB at 122 OPS+ is about two seasons of above-average play, accounting for the 50 WS difference between the two. In my cts on Griffin, I said that he needed two more good seasons to be a serious candidate. Well, that's exactly what Duffy has, and he will be.

2. Peak and decline. Duffy's OPS is brought down by having a decline phase, which Griffin doesn't have, because he quit while he was still a good player. So his career rate stats look as good as Duffy's, but Duffy was a better player in his prime, esp. in his two peak years, when he carried a 177 and a 147 OPS+. Griffin's best OPS+ rates are 140, 137, 133. Duffy was a very good player who had a couple of dominant years; Griffin was a very good player. Duffy has more black and gray ink than Griffin.

3. AA discount. One of Duffy's dominant years was AA 1891, which is a weaker league, but Griffin has three AA years at the beginning of his career.

Very preliminary.

1904

1. Jack Glasscock (4)--Moved ahead of Sutton this year. Ezra?s the better hitter, but I was underrating Pebbly Jack?s defense.

2. Ezra Sutton (3)--Ahead of Richardson on career and peak value.

3. Hardy Richardson (5)?Ahead of Start on defense and maybe peak.

4. Joe Start (6)?Might deserve to go higher. More career than McVey.

5. Cal McVey (7)--I like the Ross Barnes comparison a lot.

6. Lip Pike (8)--Not as good in the NA as McVey, but better before. Some credit for McVey's post-NL career moves him ahead.

7. Harry Stovey (9)--He's got career value on Thompson, Browning, et al even after correcting for the clerical error, the advantage is just a little bit smaller though. And that?s not even counting the new baserunning info.

8. Amos Rusie (-)--Looks very similar to Radbourn to me right now. Rusie?s ahead on the BP numbers though, but is that a timeline illusion?

9. Charley Radbourn (10)--Got career and peak edge on Caruthers and competition (and defense) edge on Spalding. Still think he's Dwight Gooden though. Could go either way between him and Galvin.

10. Pud Galvin (-)?-Back on the ballot. Starting to be convinced that he?s more than average.

11. Charlie Bennett (11)--Great defense at catcher keeps him in the middle of the Outfielder/Pitcher Glut.

12. Pete Browning (12)--Browning and Thompson and Tiernan and Jones (and now Griffin) all look identical to me. Browning's got a peak and a slight defensive (but probably only because he played center, he did play center right?) edge on Thompson.

13. Sam Thompson (13)--He's got competition advantages on Stovey and Browning, but a significantly lower peak than both, a big career value gap between him and Stovey and slightly less defensive value than Browning.

14. Al Spalding (14)--Here for his hitting and the adulation of his peers. This low because of the defense behind him, the hitters on his team compared to the competition and the amount of credit I give pitching vs. fielding in the pre-93 era.

15. Bob Caruthers (15)--Hitting and peak keeps him on the ballot.

Andrew,

5) Amos Rusie (new) -- I always thought I'd have him first or second on this ballot, but his career was very short and he didn't stand out as much at his peak as I'd thought. Plus those walk rates are astronomical. More likely to drop than rise.

I apologize for picking apart your comment, but Rusie "didn't stand out as much at his peak as I'd thought"? He's 23rd in career ERA+ among pitchers with at least 2,000 innings pitched! His career defense adjusted ERA+ was better than the six best seasons for every pitcher under consideration, including Radbourn and Galvin.

And that bit about Rusie's high walk rates would warrant concern if he was some rookie pitching today. But he was one of the best pitchers ever, certainly the best pitcher we've voted on, and he managed to do it even with astronomically high walk rates.

If you take into consideration the possibility that Rusie's blazing fastball played a role in moving back the pitcher's mound (a claim which merits investigation), then you're looking at a first-ballot HOMer.

I don't mean to bust your chops, Andrew, but there are a lot of reasons to be bullish about Rusie.

A preliminary outlook:

1. Joe Start - if we had credible stats on his early years, he'd be a shoo-in at this point. I'm willing to accept what we've been able to document as sufficient.

2. Ezra Sutton - best all-around position player on the ballot? Best ("known") combination of offensive/defensive value.

3. Jack Glasscock - right behind Sutton in this respect. Dropped him last election because I thought I was placing too much emphasis on his defensive contributions.

4. Hardy Richardson - another all-around great player.

5. Amos Rusie - a no-brainer for the HoM (despite the walks).

6. Charlie Bennett - best available catcher. Probably shouldn't rank this high, but I just like him and I don't particularly know that much more about some of the player below him to coerce myself otherwise.

7. Cal McVey - in a similar position as Sutton, Richardson, et. al.

8. Sam Thompson - unconsciously kept him low on the ballot last time; no doubt he's a HoMer either.

9. Old Hoss Radbourne - finally come around to believe he's better than Caruthers.

10. Harry Stovey - still makes my top ten.

11. Ned Williamson - after Sutton, best 3Bman available.

12. Al Spalding - willing to give credit for pre-NL days (at last) and move him ahead of Caruthers.

13. Pete Browning - I'll take the Gladiator over Mike Griffin any day.

14. Bill Joyce/Billy Nash - not sure which I'd rather have just yet.

15. Bob Carthers - took quite a tumble, but really couldn't continue to justify his higher ranking.

Prelim:

1. Old Hoss Radbourne (3) - He has been in the top four for the last couple of years, and nobody new jumps to the top. I think he is in this year.

2. Pud Galvin (4) - Probably not this year.

3. Joe Start (5)

4. Jack Glasscock (6) - I expect he will be going in this year. I don't have him in the top 2, but he is certainly HoM worthy.

5. Amos Rusie (new) - I don't rank him as highly as Hoss or Galvin, and I worry that I am overrating him even here.

6. Bob Caruthers (8) - Slips ahead of Richardson. Could still slip back.

7. Hardy Richardson (7)

8. Harry Stovey (9)

9. Mickey Welch (11)

10. Sam Thompson (10)

11. Tony Mullane (12) - He keeps hanging around on my ballot.

12. Jim McCormick (14)

13. Al Spalding (-) - He is rising in my estimation. I came around on Start. We'll see if I come around on Spalding.

14. Mike Griffin (new) - He is worthy of a mention, but not an induction.

15. Pete Browning (-) - Off for a week, and then sneaks back on.

Dropping off: Silver King, Ezra Sutton

IN PRAISE OF AL SPALDING:

Here are some numbers to look at if you are a person who is not voting for Al Spalding because "pitchers didn't matter much in the 1870s because there were so few strikeouts and Spalding's good numbers was mostly due to his solid defense."

I took Al Spalding's numbers when he was the primary pitcher in the NA or NL (1871-1876), and compared him only against the other pitchers on his teams over the years. That group consistent primarily of Harry Wright (1871-1874), Jack Manning (1875), and Cal McVey (1875-1876), (together covering about 97% of the innings,) with a sprinkling of Ross Barnes, Deacon White, George Wright, John Peters and Frank Heifer. This group combined for 324.3 innings pitched on teams where Spalding was the primary starter. [Note: I used Prospectus' numbers, which are slightly different from b-r, e.g., Spalding has 2890.7 IP at b-r, but only 2887.3 at BPro. This should not effect the conclusion.]

The first thing to notice is the DIPS numbers. The "non-Spalding group" struck out batters at about twice the rate (1.33/9IP compared to 0.71 for Spalding), but also walked slightly more batters (0.92/9IP v. 0.49 for Spalding). The non-Spalding group gave up only 1 homer in 324 innings, Spalding gave up approximately 1 homer for every 20 complete games. Generally, the DIPS numbers are pretty similar, with maybe a small advantage to the non-Spalding group.

Next, to make sure that the unearned runs to make sure that the defenses were about similar when Spalding was in and when he was out. (It is possible that when Harry Wright is pitching, the defense in centerfield drops considerably.) The UERA (unearned run average) was similar when Spalding was pitching (3.53) and when others were pitching (3.61). I therefore discarded unearned runs as being nearly entirely caused by defense and not relevant to the pitching comparison.

Finally, I looked at ERA of Spalding and his composite teammates. Spalding's ERA was 2.04. His teammates' ERA was a combined 2.91.

Therefore, not only was Al Spalding's career ERA+ 142, but his ERA+ where "+" is defined as the ERA of the other pitchers on his team (who have similar or slightly better "DIPS numbers") is 143. That 143 is not due to defense, since they were all basically throwing in front of the same defense.

So, I respectfully request that all those who left Spalding off of your ballots or had him low (he has received a vote for each of the 15 places the last two years!)because you thought that his numbers were mostly an illusion caused by his good defense reconsider, as the evidence shows that this was not the case.

I think Rusie's being pigeon-holed as a high peak low career guy. According to WARP-3, he has the best career value out of the currently eligible pitchers. He's almost 4 WARPs ahead of Pud Galvin, who many claim is an all career value guy! So you've got a guy whose 5-year peak is better than anyone's(including Galvin and Radbourn) and his career is better than anyone's. He seems like a no-brainer as the top pitcher on this ballot.

Here's a bio of Cal McVey from the Iowa Sports Hall of Fame:

McVey was a member of the first all-salaried club, the Cincinnati Red Stockings of 1869. It was not only the first, but also the most successful.

In just under a century, no pro club has matched the Red Stockings' 65 games without defeat in one season, or their stretch of 92 over two years.

By modern standards, Cal's salary was a pittance. It was a fortune, though, compared to what farm laborers -- or any others -- were being paid around McVey's birthplace in Montrose, Ia.

$100 a Month--McVey, born in the Mississippi River village near Keokuk on Aug. 30, 1850, was commanding wages of more than $100 a month from baseball at the age of 18.

The problem with Rusie is that he's the third best pitcher of his time (and he's not that close to Young and Nichols).

I'm probably going to pencil him at around 10. He needs to wait a few more years.

I said '14)Williamson- I've officially flip-flopped on the Williamson-Sutton debate. They have similar career value, but it took Sutton 6 more years to compile it.'

This is a link to an article on the National Association.

Link to a great site for 19th century pictures.

'The problem with Rusie is that he's the third best pitcher of his time (and he's not that close to Young and Nichols). '

So there's a glut of pitchers in the 1890s. There'll be a glut of catchers in the 30s and CFs in the 50s and SS today. Rusie is the best pitcher on the ballot and should be ranked as such despite his ineligible contemporaries. When Nichols and Young become eligible, that is the time to start comparing Rusie to them(assuming he's not elected already).

Carl, here's a discussion between Matt, Joe and myself about the 1884 season for Williamson and Sutton. I think it illustrates the problems with the BP approach (not that WS is 100% perfect).

Posted 2:09 p.m., June 18, 2003 (#67) - MattB

So there's a glut of pitchers in the 1890s. There'll be a glut of catchers in the 30s and CFs in the 50s and SS today. Rusie is the best pitcher on the ballot and should be ranked as such despite his ineligible contemporaries. When Nichols and Young become eligible, that is the time to start comparing Rusie to them(assuming he's not elected already).

I think the difference between the catchers of the thirties and the centerfielders of the '50s is much smaller than between Young Nichols and Rusie.

My point is that we need to compare Rusie to Radbourn, Galvin, Spalding, etc right now and not Young and Nichols. When compared to those on the ballot right now, Rusie is clearly the best pitcher. We will encounter alot of pitchers who are worse than Young and Nichols over the next 100 elections and a few dozen of those will be elected to the HoM(2 already have been- 3 if you count Ward) and I don't think its fair to discount Rusie just because he happened to play at the same time as those 2.

I would only caution that we be REALLY certain before we rush Rusie into the HOM. Glasscock looks like a lock for election in 1904. Behind him we have the pitchers Radbourn and Galvin (and Spalding) and now Rusie. Right now it looks like Rusie will be the second electee.

I guess the problem I have is seeing Rusie as inferior to all of our other first-ballot picks. Perhaps Amos needs more than a couple weeks of discussion to really figure out where in line he falls. If we rush him in, we may never be quite sure.

I just checked BP.

I'm not comparing Rusie to Nichols and Young. I am comparing his to Radbourne and Galvin. Thus far I see him as good, but not as good as the two of them.

'I am comparing his to Radbourne and Galvin. Thus far I see him as good, but not as good as the two of them. '

Even though some of you guys think Bill James is an idiot (insert smiley face) and to quote his NHBA rankings is a sure way to draw disparaging remarks, a ballot constructed from his rankings would have Rusie number one with no one close.

On WS vs WARP3:

I certainly can't say that for any given player WS is more accurate than WARP3, but because WS is an open-source system, I can identify its limitations in rating 19th-century players and correct for them. In addition to being open-source, WS is both more adjustable and more reliable because its ratings are bounded by team performance, as the discussion of Williamson and Sutton recently re-posted demonstrates.

Here are the things I try to adjust WS adequately for. Much of this will be familiar to many of us, but I thought it'd be worthwhile to try to sum up, in a sense, what we've learned from our collective study of the 19th-century game.

1) Season length and league quality. _Much_ discussed. W3 makes adjustments, WS doesn't; the adjustments W3 makes may or may not be appropriate. I fully adjust all seasons except 1871 and 1872 to 162 games, though I take _very_ high peak values from short seasons with a grain or two of salt; I accept, for now, the W3 proratings of AA quality, and I prorate NA statistics somewhat, though I don't have WS to adjust there.

2) Ratio of defensive WS assigned to pitching and fielding: by far the biggest issue right now. I think this is the main thing that WARP3 does better than WS, though I don't know how W3 does it, of course. It makes more difference for the pitchers than the fielders because their shares of the defensive WS are already divided 8 ways. My ratio for pitchers/fielders right now are about 25/75 for NA, 40/60 for 1870s NL, 50/50 for 1880s. Rusie coming onto the ballot forces me to start looking at the ratio for 1890s baseball, which could be strongly affected by the change in pitching distance -- I just don't know.

3) Ratio of defensive value of the various positions. James makes some adjustments for 19th-century, but some more may be necessary for pre-glove first-basemen. If pitchers' pitching share of defense is accurately rendered, they should receive independent credit for fielding (James just folds pitchers' defense into the pitching Win Shares). I suspect early outfielders WS may be a little bit high; I'd like to see some work on percentage of chances handled by IF vs. OF. James notes the much greater number of independent PO by catchers in the 19th-century game. Does more pop-ups mean fewer fly balls to the OF? Through the mid-1890s, I adjust defensive WS of catchers and infielders upwards by 20%, outfielders by 10%. This could be more finely tuned, but I think it works well enough for getting players in the right order on the ballot.

4) Value of speed & baserunning. WS probably already accounts for this better than W3 (especially after SB numbers become available), as it sees Hamilton and Stovey, for example, as better players than W3 does, but some boost for exceptional baserunners is called for. Because of the way WS spreads value across a team, my sense right now is that WS does not _generally_ undervalue this component of offense; it only misses the value above average of exceptional players. I suspect that some of this value may show up in the WS of the top hitters, but I don't have any direct evidence of that right now for the National League. I know that's what happens when one tries to compute offensive WS for the NA without incorporating a baserunning adjustment, but I don't know the extent to which that effect carries past 1875. I don't have a simple metric for altering WS to account for baserunning, but Stovey gets a bump for it now. He's the one player that we're looking at right now for whom I'm certain this is an issue. So far the indications that I've found are that he may well be the best baserunner pre-1890 among players who are otherwise good players.

Suggestions from those who've spent a lot more time thinking about this than I have on the rightness of rule-of-thumb adjustments would be welcome!

I think of it as Young is in the Uber-tier with Nichols and Rusie making up the top-tier.

I consider Young more in the Spahn/Carlton mode. Obviously, he was a great pitcher (and will be at the top of my ballot). I'm not confident that he was a GREAT!!!!!! pitcher. Those 511 wins are hard to put in their proper context.

As I have mentioned before, I don't buy the 1893 dividing line argument. I consider that year something akin to juicing up the ball for a given year. The one difference is a change in the pitching rules can affect the attrition rate.

The reason I compare Rusie with Young and Nichols is to get an idea of what type of environment he pitched in. Young wouldn't have had remotely the length of career he actually had if he had started ten years earlier. There are many things that need to be factored in.

As for the comparison between Knichols and Rusie in BP, I have no idea what they are doing. My opinion of their methodology is getting lower the more I analyze it.

Ratio of defensive WS assigned to pitching and fielding: by far the biggest issue right now. I think this is the main thing that WARP3 does better than WS, though I don't know how W3 does it, of course.

This probably my biggest complaint. At least with Win Shares, we have the formula. With BP, I have to have total faith in the numbers. I understand the entrepreneurial reason for this and have no quarrel with it. However, it makes it extremely difficult to analyze. At least with Win Shares, I can fine tune it (as Chris is doing).

Joe (and other Spalding bashers),

Can you make a compelling argument that he was NOT the best pitcher of his era (1869-76)? I think a lot of his support hinges on the belief that he was #1 in his time.

On Rusie, a 245-174 W-L record, with a peak of 36 wins, with Clarkson having won 49 five years earlier and Chesbro (198-132) winning 41 ten years later is just not that impressive. Nichols won 361, Galvin won 364 and Young 511, all within a few years of him. The guy was pitching on the main NY franchise, in most years a strong one, and had the NY media touting his accomplishments. That's why he's in the HOF, but he's close to the bottom of my list of 15 for the HOM. Isn't he Drysdale, not Koufax?

With regards to not having the BP formulas: Has anyone asked for them? The PECOTA cards are blocked to non-members but the WARP stuff isnt. Is anybody sure that the reason it isnt available is for entrepreneurial reasons? I know the defensive formulas were published in the 2002 BP, although the numbers dont exactly match the warp defensive numbers, so there has been some changes. Anyone interested in the defensive formulas should probably start there.

>The problem with Rusie is that he's the third best pitcher of his time (and he's not that close to Young and Nichols). I'm probably going to pencil him at around 10. He needs to wait a few more years.

John, that couldn't have been you who objected to comparing Glasscock to Dahlen, Davis and Wallace.

re Spalding:

Can anyone tell me who the silver bat winner (best hitter) for pitchers would be for the years 1871-76?

Re. Spalding, Joe said that pitching is...

>Not much more important than being the best defensive SS or catcher of his era. I think if we are going after 1870s players, I'd want Joe Start, Ezra Sutton, Cal McVey and Lip Pike in first, then Spalding. He was a very good player, but pitching (as Matt showed us) really just wasn't that important back then. Yeah it made some difference, but not much.

Well, Joe, was it more important than SS or C or 1B or not?

>He should probably go in at some point, but he's got to get in line.

Joe, Spalding is never going to make it. Between those who think pitching was only a little more important than SS (or less important, I'm not sure what your point is) and those who think the '70s don't matter, he has two chances. Slim and none. In other words, he has categorically no chance because many people are not even getting past the category to look at the man.

John, that couldn't have been you who objected to comparing Glasscock to Dahlen, Davis and Wallace.

Here's the post in question (Clint's statement is in bold; my answer is in italics):

James has Glasscock ninth among the shortstops we've seen so far by 1903: Wagner (#1), Davis (#14), Jennings (#18), Dahlen (#21), Tinker (#33), Long (#34), Ward (#35), Wallace (#36), Glasscock (#43).

Glasscock was better than Long and Ward; probably better than Tinker; comparable to Wallace and Jennings (by my seat-of-the-pants ranking). The probables and comparables could change.

Wagner, Davis and Dahlen were definitely better.

I think that was my only statement comparing Glasscock and the other guys. I don't think I made any objections about comparing the four shortstops here, do you? My only objections centered on James' rankings.

Besides, they're not direct contemporaries. They didn't play in the same environment (as Young, Nichols and Rusie did). Therefore, I'm not inconsistent, anyway (good try, though).

re Spalding:

Can anyone tell me who the silver bat winner (best hitter) for pitchers would be for the years 1871-76?

Rynie Wolters for '71, but Spalding for the other years.

John, I'm sure you recall that somebody objected to that whole conversation on the grounds that players on the ballot should only be compared to one another, not to players who came on the ballot subsequently. Yeah, I thought it was a good try ;-) I personally have no problem comparing Glasscock to Dahlen and Davis (not as good) or Rusie to Nichols and Young (nowhere near as good). Yet, Glasscock and Rusie will probably be #2 and #3 on my ballot after you know who

Signed, a FOAS the greatest 19th century player not yet in the HoM and greatest 19th century player who will never get in the HoM

I thought I'd repost this from last week with a few changes.

Defense-adjusted ERA+ (AdjERA+) is one of the tools I use in evaluating 19th Century and Deadball Era pitchers because it's tough to figure out where pitching stops and defense begins. This stat is only as good as how I apply the Baseball Prospectus numbers, but it gives a better approximation of how pitchers actually did.

The formula is ((NRA/DERA)*lgERA)/ERA.

Here are the six-season peak and the career numbers for the four pitchers we've discussed, plus Rusie.

PUD GALVIN

1881---474 IP---128 adjERA+

A try at a preliminary ballot:

Mark,

As another original FOPG, I think you have something seriously wrong with how many pitchers you think should be in. Here are the numbers I use (this is rough, and while 3B is behind and 1B is ahead of schedule, I dont have a problem with it. I just use this to see if we are getting completely out of whack):

From the first 9 electees, 1 from each position

Joe, you're entitled to your own conclusions but I think you're completely wrong about Spalding. He was the highest paid player on the Boston team (which included Ross Barnes, and the "GM's" brother, George Wright). If Harry Wright thought he deserved that money, than that is a contemporary opinion worth heavy consideration, because Harry built the teams that won 8 championships in 10 year (1869, 70, 72, 73, 74, 75, 77, 78), George's playing and Harry's managing being the only constants throughout that period.

***

TomH wrote:

"As a FOHS and an EOST, I'd make the case that Stovey rates far ahead of Thompson (as does Hamilton, who ought to be a first-ballot HOMer) because his speed and his ability to get on base enabled him to manufacture runs."

"But he only played 8 or 9 years (was it 1868 or 1869 when he started?)."

Al Spalding and Ross Barnes were both playing for the Rockford Forest Citys at least as early as 1867, when Rockford dealt the Washington Nationals their only defeat during Washington's western "barnstorming" tour. That would give him at least 10 All-Star calibre years.

Are we sure that Spalding didn't get his birth certificate in the Dominican Republic?

I'd want Joe Start, Ezra Sutton, Cal McVey and Lip Pike in first, then Spalding.

I think the pitcher who outhit the 1b-man head-to-head has a better case. Joe Start had a 113 OPS+ 1871-77, Spalding a 116 OPS+ during the same 7 year period. Is anyone arguing that 1B is more important defensively than Pitcher?

***

Ranking by WARP1 the 11 NA players that are either in the HOM or being seriously considered. (decimal points removed for brevity)

1871

>The formula is ((NRA/DERA)*lgERA)/ERA.

That's really great but nobody has yet to explain what the DERA component is. Nobody. So therefore the conclusion that every pitcher of the 1880s has the same adjERA (114) is not a credible assertion.

And in case anybody saw Joe's post and missed Mark's response, Spalding was one of the top 2 to 4 players in the land for TEN years. Not 8 or 9. Subtle distinction, yes, but one of the subtle mis-distinctions keeping him out of the HoM. Ten.

Finally, if you cut the pitching component from WS in half, which I do, the pitcher is still 23.75% of the defense. It has been said that the pitcher is no more important than the SS...well, OK, maybe a little...but, you know, not really...well, OK, sure, a bit more important (but not really). Does anybody think the SS was 23.75% of the D?

Thank you for all the great discussion today!

jimd (and Marc and Matt): Your evidence has raised Spalding a couple pegs in my estimation.

Chris: Very fine work to reveal some of Stovey's "hidden" value on the basepaths.

Tom and Joe: I'm beginning to think that Rusie may be the best pitcher candidate after all.

Preliminary thoughts: I am sold by Rusie and his strikeouts. Amos will likely wind up either 3rd or 4th on my ballot. First of the modern pitchers, so to speak.

The only other newbie I'd consider is Billy Nash but he'll probably fall just outside my ballot.

Other pitcher comments: We ushered Ross Barnes into the HOM but have no love for Al Spalding. I don't understand it (in fact I'd be happy if it were reversed). I will continue to be a member of the FOAS club until AGS is elected.

Count me among those who sees Galvin slightly ahead of Radbourn. Both being right around my personal HOM threshold. I know this is not exactly kosher, but if you remove (or discount) his 1884 special circumstances season, I don't think anyone would consider Radbourn for the HOM or HOF.

But how can you (not you specifically) not have a pitcher in your top 4 on THIS ballot?

To be in the top four, the player in question should be considered the best of all-time or the best for his time (such as Glasscock, Sutton, Spalding or McVey). Other than Spalding, I'm not sold on the other guys. Galvin possibly (I'm still not convinced), but Rusie was not the best of his time. He won't be near the top five on my ballot.

We're going to have the same problem comparing the 1960s and '70s pitchers as we are having with the pre-1893 and post-1893 pitchers. There was a big difference starting your career in the mid-sixties than starting in the mid-seventies.

John, I'm sure you recall that somebody objected to that whole conversation on the grounds that players on the ballot should only be compared to one another, not to players who came on the ballot subsequently.

I think you might be confusing arguments of mine. I think it's irrelevant that Jack Glasscock might only be half the player that A-Rod is today (for example). What should be relevant is that he was a terrific player in his own time. He deserves to be honored.

BTW, when I'm comparing Rusie with the other two, I'm only using the comparison as a guide. It shouldn't be the final answer.

I'm concerned with the downgrading of Amos Rusie because he played at the same time as Cy Young. Will I have to wait because I played at the same time as Roger Clemens?

No, because the difference between you and the Rocket is small. The difference between Rusie and Young is Grand Canyonesque (relatively).

Young is to Rusie as Clemens is to Pedro Martinez (if his career ended now). The modern guys are of better quality (relative to their time) than their 19th century counterparts.

'Joe Start had a 113 OPS+ 1871-77, Spalding a 116 OPS+ during the same 7 year period.'

Young is to Rusie as Clemens is to Pedro Martinez (if his career ended now). The modern guys are of better quality (relative to their time) than their 19th century counterparts.

Excellent comparison (though I would take Pedro's peak over Rusie's anyday).

I have Rusie as the best pitcher in baseball for only one year (1894), BTW.

I agree that we should jsut be picking the best players available, and who cares who else is playing at the time. If they aren't currently eligible, then they have no bearing on the current vote.

On the other hand, I still don't think Rusie has the credentials to get in this year. I have him about 5th.

If the HoM standard for pitchers is Cy Young, just say so and we'll kick out Clarkson and Keefe and only elect 5 or 6 pitchers. I shouldn't be penalized more than other pitchers just because I played at the same time. The project is the select the greatest players of all time and sometimes they play at the same time as each other. If we want to select the greatest player at each position in each decade, that's fine(It'd be a fun project, also), but it is a different project.

The problem you are having, Amos, is that there is a huge difference in career length pre-1893 and post-1893 for pitchers. Clarkson and Keefe would have lasted that many more seasons if they had started in 1889. Conversely, Rusie's career would have been that much more shorter if his career started in 1880 (this is all relatively speaking).

As I mentioned before, there were seven pitchers that started their careers from the sixties that won over 300 games, while there were none during the seventies. We're only talking a decade's worth of time, but there was a big difference in the pitching environment for each era.

The pitching attrition rate is extremely variable throughout baseball history and makes analysis that much more harder.

PS Stop your whining, you big baby. :-)

John Murphy wrote:

"I have Rusie as the best pitcher in baseball for only one year (1894), BTW."

How about 1893 also? And wasn't he #2 in 1897? Remember, the years 1892-99 were years of one 12-team league. #2 in the league meant #2 in the world, basically.

WARP1 tells you the value of the player in the context of the game that he was playing; it is vaguely analogous to Win Shares in that respect. Using WARP1 it is clear that Spalding had the most value of any player during his career (or close to it). Just like it is clear that Mullane had a lot of value within the context of the AA alone.

WARP3 tells you Davenport's quantitative opinion of how Spalding would do in a game without the rule restrictions that the NA was played under. His "opinion" is that the hitters of that time would adapt pretty well to the pitching revolution, and that the pitchers weren't good enough to compete and would be forced into an early retirement or into relearning how to pitch (like Mathews did), because this is what actually happened. WARP1 shows the season within its own league context; WARP3 shows how it fits into a larger context adjusting for league quality and schedule length. And there's no doubt that NA pitching could not compete with 1880's pitching, which is why there have been no Spalding style pitchers in the majors during the last 120 years.

But I don't agree that this conclusion means Spalding had little value to his team, relative to the other pitchers in his league, because it is obviously false. Davenport's WARP1 numbers show that Spalding was much more valuable than any of the other pitchers (or players other than Barnes or Wright) playing under the NA rules. They back up Harry Wright's monetary opinion (and show that Barnes maybe should have had a better agent :-).

You can hold WARP3 against Spalding if you like, but I believe that shows a misunderstanding of what it is attempting to measure. IMO, doing that is like arguing that nobody before (pick your date) belongs due to timeline issues, or Barnes doesn't belong due to the fair/foul rule; these go against the spirit of the HOM, which is to try and evaluate the pitchers and players of each era within its own context and honor the best. You can argue that his career is too short and that you don't place much weight on peak, but you can't use the BP numbers to argue that he had little value to his teams, because they show the reverse, that he was arguably the most valuable player of his time.

In support of the FOHR, here is the declassified listing of the Enemies Of Hardy Richardson:

JeffM: Hardy enemy #1, the only voter to leave him off his ballot in 1903.