Shouldn't the onus of proof fall on those who think that Richardson (or anyone) is worthy, rather than those who think he isn't?

Thus I am currently trying to put together an argument for Hoss Radbourne. I'll get back with it soon...

Thinking a little more about that Glasscock/Larkin analogy...

Glasscock had 1736 games played, but in a long career (that started young). Our implict standards for how many games a good defensive shortstop should play were formed in our minds in a time of 154-162 game seasons. So, let's do a very simple exercise: take each year of Glasscock's career, divide his games played by the team games played, and multiply by 162 to see what playing that fraction of the time would look like in a 162 game season. If you do that, his the "games" column now adds up to 2414. Not all of it was a shortstop, but that is a number consistent with the notion that he was a very valuable shortstop. I tried "re-weighting" his OPS+ for the variable season length but it turns out that that hardly made any difference. In comparison to some modern shortstops (with their games adjusted for 154-game seasons or strike interruptions):

Glasscock 2414 "games", OPS+ 112

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.

I have Killen in '93 and Nichols in '97.

I have Killen in '93 and Nichols in '97.

Killen edges out Rusie (barely) because of his bat. Rusie, as a pitcher, was better.

I have no reservations about Nichols over the Hossier Thunderbolt in '97, however.

Oops. That last "rusty" post was actually me.

I have looked at Radbourn, Galvin, and Rusie using 7 different measurements, to try and find a clear leader. Unfortunately, it isn't clear at all. I'll show the variances and explain why I stand behind Hoss.

Seasons

John Murphy wrote:

As for 1897, do you agree that Amos was #2 that year?

Yup.

Rusty, your post was good and illstrates the point of what these HoM debates are really about. They are NOT about player A vs. player B. They are about timelines and league quality discounts and pre-1893 (and pre-overhand) pitching value and peak vs. career. Obviously there are some players who cannot be made into HoM candidates regardless, but there are a good dozen to 15 now, and in 100 years there will be 100 candidates not yet elected. And how one voter ranks those candidates has everything to do with subjective principles and only secondarily with actual numbers. In other words, we choose our principles first, then the players and finally the numbers we want to support our case with. (Some may choose the players first and then the rest). But we cannot and do not hereby establish what the many players' objective value was, we can only register our opinions about how we "value" the particular "value" that a player had.

Signed, Plato

I agree with what Marc / Plato ;) says. We should not pretend we are completely objective. We are instead trying to be as objective as we can given out own biases and frames of reference.

The only true objective results are unaltered numbers, and we know how much people love those. :)

<i>Win shares has Killen ahead 42-41.

I havent' mentioned this yet, but I just wanted to share how funny I find the name Bug Holliday. Can't you just picture a family of roaches, sunning themselves by the pool, and calling for another Marguerita? Or would that be Margueraida?

Where do bugs go on holiday, anyway?

Speaking of Joe Blow's Amazing Sabermetrical Formula reminds me of participating in the SABR retroactive Cy Young award survey, back in about 1985 or 86. Pete Palmer's "The Hidden Game" was the state of the art reference at the time.

They only went back to 1900, but if I apply the nascent sabermetric methodology I used back then to 1893, Rusie is an easy choice, because he led the league in IP by a lot and was #2 in ERA. I concluded that Killen's better W-L record was entirely due to pitching for a better team. I think Rusie had more wins above team.

If the BBWAA was voting, they'd probably vote for Killen, since he had the best W-L record, a good ERA and played for a contender.

Doggie Miller, aka "Foghorn" aka "Calliope" is eligible. OK, he's nowhere near as good as Jack Clements, so we're not going to take him seriously. But if he had just retired from your modern team, would you want to hear that he had been hired to be part of the broadcast crew? Or maybe he could be the stadium PA announcer.

What would Cy Young's win total be in today's game? 325-350? Would Rusie be more in the 150-175 area? That Clemens/Martinez comparison looks better and better.

OK, so Rusie is like Pedro and Cy Young is like Clemens. How exactly does this hurt Rusie's case?

OK, so Rusie is like Pedro and Cy Young is like Clemens. How exactly does this hurt Rusie's case?

Because I don't think I would have Martinez near number one on my ballot (depending on who is there, of course) at this moment (unless you're ranking 100% peak). And Pedro is still better peak wise than Rusie.

I doubt Koufax will make my the top few slots in the early seventies, either.

'Because I don't think I would have Martinez near number one on my ballot (depending on who is there, of course)'

Would you rank him 10th in a weak election year?

Would you rank him 10th in a weak election year?

Definitely. I probably would have Martinez above 10 for a strong election year. His peak is so much better than the Thunderbolt that he can move up higher even though his career is not that long.

If you take Win Shares totally seriously in the 19th century, you will have to conclude that most of the greatest pitchers in the history of the game played before 1900, and the 19th century catchers were mediocre at best. Thats one reason why BJ cut himself didnt take the numbers seriously and cut the pitchers WS in half for the Historical Abstract, and also why the best catcher of the 19th century (Ewing) is rated roughly at the level of the 5th or 6th best catcher of the 1970s

But if you compare a certain player to the others at the same position (which I do) for each season, it doesn't really matter then.

Rusie had 3800 IP, Radbourn 4500 (fyi Pedro has 1900). A big difference, but not enormous, especially since Rusie has 10 extra points of ERA+ and put up some huge performances in a very high offense era (1893-95).

The problem again is that you are comparing Rusie and Radbourn as if they pitched during comparable conditions. They didn't. The dynamics of endurance and career length were radically changed during the 1890s. Of course, this ebbed and flowed through out the last century.

You have to compare the pitcher in question with his peers. I can't repeat this enough. A guy like Pud Galvin would have added at least five more years to his career if he had started a decade later.

But Rusie pitched 3,800 innings! So he's not a 5,000 or 6,000 inning warrior like Pud Galvin. So what? Very few pitchers are.

Amos Rusie pitched 3,800 innings of 130 ERA+ ball. Res ipsa loquitur.

res ipsa loquitur - Latin. "the thing speaks for itself"

Thanks to Google for aiding us denizens of these post-literate end-times.

This is part of a study that I started last "year", comparing the 2B/SS/3B based on defensive usage between 1886 and 1986. It didn't get any reaction last time, but I think it's important to consider as Richardson/Glasscock/Sutton rise to the top of some ballots. I'll copy my conclusions on top so, people can decide whether it's worth reading further:

DanG,

"Res Ipsa Loquitur" is actually a burden shifting mechanism in the law.

If I walk by my factory and a barrel rolls out of your second story window and hits my on the head and I sue for damages, your defense might be, "Hey, it's not my barrel!" Then, I'd have to go and prove that it's your barrel, which will be hard because I don't know a lot about barrels or what kind of barrels you keep in your factory.

Res Ipsa Loquitur allows me to say, "Hey, the barrel rolled out of YOUR window. I think we can reasonably assume that it was your barrel." Because of the doctrine, the burden now falls on you to prove that it was not, in fact, your barrel. If you do so (perhaps by proving that all of your barrels are red, and this was a blue barrel), you can still win. But, unlike a normal case where I have to prove you hurt me, if the doctrine of Res Ipsa Loquitur applies, you instead have to prove that you DIDN'T hurt me -- essentially proving the negative.

James Newberg's invocation states: "Sure, it's possible that Amos Rusie is not an All-Time Great Pitcher, but he pitched 3800 innings with a 130 ERA+. What those facts, it is not for the FOAR to prove his case. Rather, it is for the EOAR to disprove it."

Be very careful with ERA+ (or any other stat) during high offense era. Have you noticed the increase in leading ERA+ figures since the home run explosion of the nineties? You need to use standard deviation to draw back those outliers.

As for Rusie, all I'm asking is that you have to compare him to his peers. Not just Cy Young, but to all pitchers that started around the same time as Amos. Forgetting about the limitations of the stats in question. John Clarkson was a much more effective pitcher (peak and career) for his time than Rusie was. Yet, if you didn't compare them to the others around him, one might not pick that up.

Caveat emptor. Quo vadis, Et tu, Brute? Veni, vidi, venti. Anno Domini. There, I think that's enough Latin for everyone to digest. :-)

Speaking as a FOAS, it seems to me that there are several objections to his candidacy, all of which I disagree with. Here's why.

1. First, there are those who say his achievments in the '60s don't count for squat because we do not have a "normal" (ie. detailed) statistical record thereof. The record is anecdotal and narrative, not statistical. This camp would also be EOJS and others. There is no logical argument to be made in response other than to say that baseball clearly was played in the '60s, championships were claimed (at least) and the identities of the best players generally agreed to. The numbers would be nice, but the lack of numbers does not mean the '60s didn't happen.

2. Second, there are those who say his achievements in the NA don't matter because the competiton was not comparable to competition that came later--ie. the timeline argument. Some also say as a corollary that some of the better players were playing in other circles rather than in the NA. To these arguments, the "pennant is a pennant" argument carries the day for me. The gentlemen who played in the NA sought out and played against the best possible competiton of the time and their achievements really did occur. They deserve recognition. And on a slightly more tangible plane, there is evidence that the later NA, at least, was pretty comparable to the early NL which in turn may have been better than the mid-'80s NL due to the larger number of teams of the '80s. The competition argument, in other words, is questionable on the evidence even if you don't agree that it is irrelevant as a matter of principle.

3. Then there is the argument that pitching just wasn't very important at the time, becuase, after all, they pitched UNDERHANDED! Or whatever. Well, I cut pitching WS in half pre-'93. But if we are talking about Value and not about Tools, then you have to adjust for the massive numbers (or percent, at least) of innings they (he) pitched. Surely the pitcher was still more important than the SS or the 1B. Surely.

4. Besides which Spalding threw 2900 innings in the documented portion of his career alone ('71 and after). Even if you don't adjust to 154 or 162 games, surely this represents a "career" sized sample. His achievements/quality cannot be attributed to an insufficient sample.

5. Which brings us to the short career argument. And for those who reject the '60s and the NA, sure, he had by those standards one good year. If you are willing to consider his entire record, then, you still may feel that ten years as a top pitcher is not enough. I don't know what to say to that other than this--sure we can construct a list of durable pitchers, pitchers who threw more than 2900 innings and more than 10 seasons. And sure we can construct a list of the most effective pitchers (ERA+, whatever). How many pitchers were more effective over 2900 innings? I haven't constructed precisely such a list, but we know what the answer is. Not many.

6. Then finally to reject AS you have to reject contemporary opinions. And of course many of you do. I have seen posters say that, hey, we have statistical analyses now that make contemporary opinions obsolete (just wait til we get to G. Sisler). Do we seriously believe that if Harry Wright had hired us like Theo hired Bill James that we would have helped him build a better team? C'mon. Contemporary opinions should count for something. And there is a world of difference between the opinions expressed about Ed Williamson (several years after his retirement as he lay on his deathbed) versus those concerning AS, expressed while he was active in the form of $$$. Contemporary opinion has to count for something.

So to reject AS altogether, all you have to do is: Blow off the '60s. Blow off subjective evidence, including contemporary opinions. Blow off the NA. Blow off pitching pre-'93. Blow off peak value. Claim that 2900 innings and 10 seasons makes a small sample. If you want to claim all of those things as a matter of principle, by all means do NOT vote for AS.

Marc,

I, in fact, do discount #6, but your other points are great.

In your opinion, what would a conservative estimate of some of Spalding's numbers over his career, like WS or WARP?

Tom,

I'm not how this analysis changes my rankings of Richardson/Glasscock/Sutton. I was just noting that it seems like an important thing to get a handle on as we enter their portion of the ballot.

WARP3 gives us:

Glasscock: 103.9

Even I would not attempt to render AS' pre-NA efforts in terms of WS. But for the period '71-'76 I have him down for 352 adjWS (adj to 162 then X .50 with the other .50 going to the defense).

By comparison, I have Clarkson at 268 and Keefe at 285, both of which were good enough to rank highly on my ballots. Among others under consideration I have Hoss at 278, Welch 258, Galvin 284, Caruthers 202 (though I rate Freedom Bob highly on peak), Bond 259, Mullane 254, Foutz 177, Corcoran 168, McCormick 284 (!), Hecker 176, S.King 167, Hutchison 165.

I rank on peak to a fairly large extent, so the above numbers are not the be-all and end-all. Welch and Galvin did not have high peaks, Caruthers did. But Spalding quite obviously had a very high peak what with 352 adjWS in 6 years. He is the only one with a very high peak and high career value.

I am not unaware, BTW, of the argument that his rate (percent of team's IP) could not have been sustained over even 80 much less 162 games. The 352 is taken with a grain of salt from that perspective, likewise Bond's 259, etc. But even if you reduce Spalding again (above and beyond the 50 percent penalty he has already suffered), I don't see how he is not still equal (or better; better IMO) to the top guys--Clarkson, Keefe, Radbourn, McCormick, Galvin....

Marc, great job stating the case for Spalding. I assume you also rate him above Rusie?

My Pitcher rankings for this year had been:

1. Radbourn

Dan, I am also a FOAR but yes I have Spalding above Rusie. It is hard to say precisely why (it is indeed meaningless to compare their raw totals), but I think it boils down to this. A 10 year career was somewhere around the average for the best players in Spalding's day. In Rusie's day, 10 years is indeed a short career. I guess my most basic principle is that you rate players based on how far above the norm they were for the time in which they played, and (speaking purely figuratively) I would say that AS was about 2 SDs above the norm of his day, Rusie was about 1 SD above the norm. That is how I would try to compare the two based on how I see this whole process of ranking and rating across eras.

I guess my most basic principle is that you rate players based on how far above the norm they were for the time in which they played, and (speaking purely figuratively) I would say that AS was about 2 SDs above the norm of his day, Rusie was about 1 SD above the norm. That is how I would try to compare the two based on how I see this whole process of ranking and rating across eras.

Bingo. That's how I would do it, too.

'I guess my most basic principle is that you rate players based on how far above the norm they were for the time in which they played, and (speaking purely figuratively) I would say that AS was about 2 SDs above the norm of his day, Rusie was about 1 SD above the norm. That is how I would try to compare the two based on how I see this whole process of ranking and rating across eras.'

I assume most of us here would agree that the 'norm' should be Replacement level, not league Average and certainly not the best pitcher in the league. Using W1(I adjust W1 up to a 70 game season to adjust up for the season length and down for Spalding's high usage and the lesser quality of the league) for Spalding and W3 for Rusie(which seems reasonable to me), Rusie show a slightly better peak compared to RL. If we assume 4 6 WARP seasons for Spalding before the NA(I not willing to assume anything more than that given that he was only 16 in the 1st of those seasons), there career values come out very similar as well(84.2 for Rusie to 81.5 for Spalding). This is about as liberal as I'm going to get with Spalding assumptions and I'm still going to put Rusie higher on my ballot. I think that to put Spalding ahead of Rusie, you have to be more liberal than I with NA adjustments or Pre-NA guesses or you just believe that the best pitcher in 1 era should be ranked higher than the 2nd or 3rd best pitcher in another. I just don't think that the difference between the best pitcher and a Replacement level pitcher was that great in the NA. Alot of it is due to the time they played, but it seems to me that Rusie was more important to his team than Spalding was to his. I think I may raise Spalding compared to some of the hitters and I still have to re-do this evaluation to figure out where Radbourn and Galvin fall into the mix. This is not that much of an upgrade since I believed Spalding belonged in the HoM to begin with, but now I think he should go a little earlier than I previously thought.

Jason, I agree that the weaker teams in the NA had peers outside the NA. But the toughest competition that a team or individual could possibly face was in the NA. A team that stayed out of the NA was by definition NOT playing the best competition.

On your second point, once again you are building one conjecture on another. IF there were players over 29 year of age, they were playing somewhere other than the NA, and this proves that all the best players were not in the NA? Many of the very very best players (Spalding, Barnes, etc.) did not play past age 29. It is likely that almost ALL players were making a respectable living by that time, not continuing to play for non-NA teams. The game had only been played professionally for three years at the time of the NA, why would there be old players. The lack of 30-somethings in the NA does not prove that the 30-somethings were playing elsewhere. You're asking NA supporters to prove that they've stopped beating their wives.

Maybe it's just me, but I'm inclined to put Galvin AND Radbourn AND Rusie AND (probably) Spalding in the HOM (in that order). Start is my key hitter not yet in, and I have no problem with Glasscock, either. But pitchers are moving up on my radar, I think...

'The lack of 30-somethings in the NA does not prove that the 30-somethings were playing elsewhere. '

Joe, it's now clear that you and I have a different model of what's happening where in the WARP1 & WARP2 calculations, and that this explains much of our disagreement.

The following are references/links to Davenport's Statistics Glossary:

Carl and Jason,

I can't refute your timeline, it is a philosophical issue though there is some evidence that the NA was not much poorer than the NL of the mid-'80s if not '80-'81. But with your timeline (population) argument, how are players from the 1940s going to stack up against players from the '50s and early '60s--ie. before and after the color line. I can't imagine that if I timeline/demograph one era vs. the next that any '40s guys beyond DiMag, Ted and Musial are going to amount to much.

Which is why I say that timelining, in my "pennant is a pennant" philosophy, is irrelevant even if the evidence of a large change is irrefutable. But when the evidence of a significant change is circumstantial (not enough 30-somethings, e.g. based on who is NOT playing as opposed to who IS) then the timeline fails on empirical as well as philosophical grounds.

The other argument against the timeline is that someday the population of ball-playing men will be three times what it is today. At that point the players of the 20th century will also look like 3rd rate bums. I'm not prepared for my heroes to be 22nd century bums.

Oops. Posted this on the wrong thread:

Saw this link over at Primate Studies about the correct Replacement Level to use:

http://premium.baseballprospectus.com/article.php?articleid=2032&mode=print&nocache=1056622106

The article, and the discussion at Primate Studies:

Charlie Bennett versus Jack Clements:

The two of them did play roughly the same number of total games, but Bennett, whose career included shorter seasons, played in a substantially higher proportion of his team's games than did Clements.

Bennett had a 5-year run, 1881-1885, in which he played in 88% of his team's games and had an OPS+ in the 145-150 range. That's a very high proportion of games for a catcher of his, or any, times.

The best we can do for Clements is a 4-year period, 1890-1893, in which he played in 73% of his team's games and had an OPS+ in the 130-135 range. For the next 5 years after than, 1894-1898, Clements hit very well but only played in 50% of the games. Before 1890, Clements's offensive stats don't look that strong (although he did once lead his team in BA with a .245).

In defense of Clements, it's easier to play in 90-95% of the games in an 85 game season than it is in a 155 game season, and the mid-80's NL in which Bennett played was probably thinner competition than the early 90's NL of Clements.

I wasn't looking at defense. Both of these guys were always primarily catchers, and were presumably good at it. A lot of the arguments for Bennett a couple of debates ago dealt with defense, and I would guess that it would be hard to make the defensive arrow point in Clements's favor.

Conclusion: wherever the catchers fit in absolute terms, I'm going to put Clements below Bennett.

I also will be away on vacation with no net-access for the next 10 days, so this is both my preliminary and final 1904 ballot.

Read previous ballots if you want more depth on my reasons for all but the latest eligibles.

1) J. Glasscock -- A long and valuable career. Clearly the best of the fully documented careers here, plus a good peak. Going with him #1 given the murky distinctions between the pitchers. The NHBA rankings of any player peaking 1886 or before are seriously hurt by James' decision to not adjust for season length, and doubly hurt if they played well a key defensive position that should receive the Win Shares that the pitcher would no longer get.

The case for Amos Rusie is a peak value, short career case. The single best year of Rusie's career was 1894, in which he went 36-13 with a 2.78 ERA (ERA+ near 180). The following doesn't really have much to do with anything, but what was his 1894 team like?

The 1894 Giants finished 76-56, in second place 3 games behind Baltimore.

Here's the everyday lineup:

C: Duke Farrell - 18 year career; will draw some HoM debate.

OCF, are you tellin' me that when Rusie or Meekin wasn't pitching the Giants went 7-34? My god, the GM should be shot.

Not quite that bad. The rest of the staff went 19-22. B-R.com has the Giants at 88-44 that year, with a Pythagorean of 76-56.

No, the Giants were 88-44. OCF sited their pythagorean W-L.

So, they were 19-22 in the non-Rusie non-Meekin decisions.

I think the real question here, though, is how the heck did the team out-perform Pythagorus by 12 WINS!

>I think the real question here, though, is how the heck did the team out-perform Pythagorus by 12 WINS!

That Rusie fella musta really beared down when he had to! And as an aside, whoda thunk that a team with Ward and Davis would have a SS platoon going that included neither one!?

Sorry, guys. I misread that bbref page and didn't proofread. jimd and MattB fixed my mistake.

My current inclination, but this ballot in particular requires a LOT of thought:

1. Joe Start

And throwing in some utterly un-sabrmetric observations, a category that no one wants to win...

Speaking as a FOFD, I still gotta ask choc...6th? Surely you realize his '84 season was in the UA!? I like Fred...in 13th to 15th place, depending on the competition.

And throwing in some utterly un-sabrmetric observations, a category that no one wants to win...

I don't know if it is un-sabermetrical, but it surely was fun. Thanks, Howie!

2 comments about John C's data--OK, 3. 1) fun and interesting.

2) of course the guys whose position was "batter's box" dominate and everybody who actually played defense is lower on the list. SB data or no, I don't think a quality C or SS, etc., could possibly be hurt by this data. If Pete Browning's number was 20 you still couldn't play him there.

3) would it help to know the run environments in which these numbers were produced? Just off the top, we had very high run environments at selected times (through '76, '94) and relatively low run environments ('78-'82, '88) with the rest shades of gray. Players on the list whose peaks were '78-'82 would be more valuable than a player with the same RC and a peak from '89-'93, right? It seems that we ought to normalize these numbers a bit to really understand what they are trying to say.

I am also a bit confused as to how to react to Start's and Sutton's low numbers. Given that they played in a high run environment pre-'76 their numbers look especially bad. But a lot of runs in that time were unearned. So are a lot of the runs scored not accounted for in RC? How to interpret???

Any strong feelings on Radbourn vs. Rusie? Seems like a good "peaker vs peaker" debate. Galvin and Spalding are different animals entirely. The R men also are different pitching eras, but their relative dominance is at issue..

Howie, yes. Radbourn's reputation rests largely on one season, 1884, in which competition was diluted to a degree matched in only one other ML season (1890) before the 1970s, or considering population, a degree never matched thereafter. Here are the OPS+ numbers for each from peak to valley (only in seasons in which he is eligible for ERA title):

Hoss 206-151-134-133-122-113-109-106-99-89-79

Wow, this will be a hairy voting week. Upon getting back from vake on Friday, I figured I'd see lots of prosletyzing about candidates for this crucial election.

Glasscock, Radbourn, Richardson, Start, Galvin, Sutton, Spalding, Thompson, and Stovey all would seem to have MUCH to gain or lose from this election, and it's not at all clear that there is any consensus whatsoever on where Rusie fits on all this.

We don't appear to be making a waiting list because if we were Hoss would be a shoe-in this week (with Glasscock). Unfortunately (imo) that doesn't seem to be the case. Hopefully (again imo) some people will coem around on him before casting the actual ballots.

We don't appear to be making a waiting list because if we were Hoss would be a shoe-in this week (with Glasscock). Unfortunately (imo) that doesn't seem to be the case. Hopefully (again imo) some people will come around on him before casting the actual ballots.

Oops. Sorry 'bout that.

This is somewhat related to some things I want to think about for the 1905/06/07 elections. It's a question about whether data exists or whether certain kinds of research have ever been done.

We know that fielding errors were very important in 1880's/1890's baseball. How else could they have scored 6 runs a game (it was over 7 in 1894) with those BA/OBP/SLG numbers? That in turn must mean that reaching base on an error was a common enough outcome of a plate appearance to matter.

So:

Our SABR chapter in Connecticut is thinking of doing a book of bios similar to the Green Mountain Boys of Summer, so I was looking up Wild Bill Hutchison.

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+.

SABR had a pub called Nineteenth Century Stars that came out in 1989. There was an article on Hutchison. He was a Yalie, OCF. I'm guessing that he came from money because he prepped before going to Yale. While an Eli, he starteed out as an infielder before being converted to pitching. After some postgrad work, he moved out West to be a speculator. He still played baseball, but mainly as an amateur. He did pitch a couple of games for KC in the UA, but I'm getting a mental picture of a guy who was a throwback to the Cartwright era. That didn't last. At some point he realized that he was better at pitching than at business, so he turned pro. That explains the late start.

Where there any particular methodologies that you guys used to rate pre-1893 pitchers that were unique to that era?

Heh.

I jumped into this thread thinking it was the current discussion for 1997. :-D

Where there any particular methodologies that you guys used to rate pre-1893 pitchers that were unique to that era?

I personally didn't, Jon. AFAIAC, the top pre-1893 hurlers were just as valuable as the Rusie generation.

Where there any particular methodologies that you guys used to rate pre-1893 pitchers that were unique to that era?

IIRC, the Win Shares voters needed to deal with those unrealistic pitching numbers. Some followed James lead in the NBJHBA and cut them in half, some did sliding factors over time. Some reassigned the "missing" Win Shares to fielding, some (quietly) ignored them.

Many voters used mostly ERA+. Some debates about comparability of ERA+ from the early "single ace" era, pre-1882 or so, with the 2 starter "rotations" of the mid 1880's, and 3 starter "rotations" of the late 1880's/early 1890's. There was a massive expansion then from 8 starting jobs in 1880 to 48 starting jobs by around 1887 due to schedule expansion (84 to 140 games), the addition of the AA as another major league, and the legalization of overhand pitching (resulting drop in IP by the aces).

There was also a lot of discussion about head-to-head matchup analysis (who pitched against who) when that data became available from retrosheet. (It wasn't available when the earliest votes were taken.) IIRC, Welch got a significant boost when it was discovered that he tended to draw the other team's ace, and did pretty well against them in W-L.

Unfortunately, I don't remember exactly when many of these discussions occurred. Also some of the longer posts did not survive the "baseballprimer to baseballthinkfactory" site migration that occurred during the mid-1920's. Skim through the old threads and look for stuff.

For Al Spalding, pitching for major teams 1867-1876 or so, some of the same issues were covered within the 19c context. EAS (enemies of AS) wondered, if we halve pitching win shares in 1892 or 1885 should we halve again for 1872? Compared with Ed Begley and Hank O'Day, oh say Tony Mullane and Jim McCormick, didn't Spalding rely that much more on his fielders --and enjoy that much greater advantage because his fielders were good to excellent.

Amos Rusie arrived young and his career was short. There was discussion in his first year eligible (1904) whether a man should be compared with his contemporaries rather than only with those whose mlb careers end when or before his own ends. For Rusie that is Nichols, Young, and McGinnity. Does he enjoy inappropriate advantage in coming up 5, 10, and more years before his age contemporaries?

There was also a lot of discussion about head-to-head matchup analysis (who pitched against who) when that data became available from retrosheet. (It wasn't available when the earliest votes were taken.) IIRC, Welch got a significant boost when it was discovered that he tended to draw the other team's ace, and did pretty well against them in W-L.

Thanks, I was wondering if that was something that folks looked at. Also, while I'm not a huge believer in pitching to the score, I thought that that might be more of a factor in those days.

while I'm not a huge believer in pitching to the score,

That was also discussed WRT Mickey Welch (and Clark Griffith). IIRC, some of that discussion was during the elections during the late 1930's (1938-40,42), a backlog period just before the deluge of 1920's/30's stars. If it's not there, check the early- and mid-1930's.