Here is my preliminary first ballot.

1. Roger Connor

Unlike many, I guess, I have a strong bias toward peak over career. This belief is based upon the strong feeling that even if Charlie Hough was able to pitch until he was 80, and had retired with a career record of 616-616 and an ERA+ of 108 (instead of 216-216), he would still not be a Hall of Famer, because he was only an average pitcher during the heart of his career, and being average for even longer doesn't make you great. Another way of saying this is that, over the course of a 10+ year career, replacement level should be "average".

As such, my ballot may look different from those above in terms of placement, but should include mostly the same names. Also, it will look slightly different than it would have looked yesterday and will look tomorrow.

1. Dan Brouthers

A couple of things:

My ballot:

I don't have a prelim. ballot just yet but I would just offer a counter to many of the comments so far. Like the other Mar(k) (the one who spells his name wrong) I try to balance peak and career value. Few players have the opportunity to strongly influence more than a few pennant races in their career, and a player with a high peak can "more" strongly influence a few of those races. And the ultimate unit of measure, the common denominator in baseball is the pennant. Not so much whether a player won one or not, but how much of an advantage did he give his team in a pennant race or races. All the rest is just hangin' around.

Peak value is especially important in the 19th century and especially the early 19th. Many 19th century players, through the accident of birth, had no real opportunity to accumulate career value like players do today, with careers of 20 years and more of 162 games each. Particularly for a player born before 1850, he's lucky if he got 10 "ML" seasons of 50 to 80 games each. So there's gotta be a way to equalize. Players who dominated 50 to 80 games would have dominated 154 or 162. Peak value, properly calibrated, offers the short season/short career player of that day--which covers pretty much everybody--fair recognition of their athletic and baseball skills as they related to the environment of the time--training, diet, coaching, playing time, strategy and everything else. It's not a question of whether a player from 1875 could plop down in 2003 and be a star or not. It's who he was and what he did when he was playing.

(Besides which, the greatest athletes of the 19th century, given the coaching, playing time, etc. etc. etc. of today would be great athletes today. With modern training Jesse Owens would have run as fast as Carl Lewis. Johnny Weismuller would be a world class swimmer today if he had grown up over the past 15-20 years.)

So I think Bill James goes too far with the 3 and 5 year peaks, I try more for a 50-50 balance and, frankly, as it relates to the 19th century, probably even more toward peak. Whatever that means.

More specifically, a lean toward peak value is meant to give Barnes, Spalding and G. Wright their due. I won't quibble as to whether they are top 15 or not, but surely top 25. Dan b, you seriously want Mike Tiernan ahead of G. Wright?

Just a few follow-ups myself. Of all the players appearing on other people's preliminary ballots (and not on mine), the only one that I would seriously consider adding to my ballot at this time would be John Ward. By the way, his biography makes clear that Ward actually did not like being called Monte.

I will take a closer look at all the 19th century guys before casting my real ballot, so maybe I'll wind up adding Ward, Keefe, McVey, Williamson, Dunlap, Bennett, McPhee, or Start; but I kinda doubt it. No way I'm voting for Spalding, Wright, Thompson, Tiernan, O'Neill, or Barnes (see below). I'll comment on other pitchers below. I don't know enough about George Stovey to vote for him (sorry).

Ross Barnes represents probably the widest disparity in opinions among 19th century players. Some vociferous folks believe he was one of the best players in the 19th century. Most others disagree. I am firmly in the camp that would not seriously consider putting Barnes on my HOM ballot. I don't think he merits it based on the numbers, and if you discount his achievements to any significant degree (due to the fair-foul bunt), he drops far below many others who had longer and more accomplished careers.

I am not enamored by 19th century pitchers, especially before the move to 60'6" in 1893 (someone correct me if I have the date wrong). The only two pitchers eligible for the first HOM ballot that I give serious consideration to are Clarkson and Rusie. Guys like Keefe, Galvin, Welch, Bond, McCormick, and even Radbourne cut no ice with me. Charlie and Joe are modifying the win shares system for the 19th century and I am betting that they attribute significantly less credit to pitchers in the pre-1893 baseball than we are used to in the 20th and 21st centuries.

Thanks much.

I'm trying to balance peak and career, with an era adjustment for playing after 1893 and playing in the NL vs. the NA or AA.

I'm sort of between the two camps on Ross Barnes.

If it were a strict Yes/No vote, I'd vote "Yes." When it came to actually constructing a ballot, though, it seemed difficult to rank him ahead the specific other people on my ballot.

So he's hanging out in the 15-25 category, and will get re-evaluated when more slots open up next "year."

Rob, if you're willing to consider Clarkson then you aren't totally discounting pre-1893. And if that's the case, take another look at Keefe. I agree that the old timers overrated Radbourn pretty dramatically (elected to HoF in 1939, all the others [Clarkson, Keefe, Galvin, Welch] in the '60s and '70s) and frankly I don't think Welch and Galvin add up either. But I don't know how you can differentiate Clarkson and Keefe to the extent of an automatic in/out, black/white judgement.

I agree that Barnes mastered the rules of the day. He had no way of knowing which rules would one day be changed. Are we going to refuse to elect a spitballer? A catcher who didn't crouch?

I understand discounting pre-'93 pitchers, but I mean discounting as a percent of value, not reducing it to zero. So it is with fair/foul and other rule changes.

BTW I understand that the adjWS do just that, they discount pre-'93 pitchers by 50%, give the other 50% to the defense, but normalize the whole thing to 162 games. If you do that then Clarkson has 268 WS and Keefe 285, which puts them in the Marichal, Ed Walsh, Joe McGinnity, Bob Feller, Don Drysdale range. You have to decide for yourself how good a Marichal, Walsh and Feller comp is. But I guess you can and should discount their numbers, just not entirely.

Keefe pitched two years in the AA. As for his '84 season, you can discount everybody in every league in '84 if you want to. Placing Keefe even with Welch means exacting a heck of a penalty for (basically) the one remaining season 1883.

This is very preliminary; I do want to note that I deliberately downgraded Anson and Brouthers because they weren't the best at their position; I may change my mind about this.

1) Connor

Andrew, well, what can I say, that was a long long time ago. I've revised my opinion, it's at least a couple of schnipples.

I should have been clearer with regard to Keefe. I currently have him just outside my first ballot. I think he is 17th on my list. I didn't mean to suggest that I *totally* discount pre-1893 pitching performances.

I see McPhee and Glasscock as at least superficially similar. Each arguably the best of 19th century at his position for the whole career. But McPhee was never the best for any substantial period of time. Barnes obviously the best 2B in the '70s, Richardson in the '80s and Childs in the '90s. So, hard to decide which goes to the top of the least and frankly, the competition drags them all down a schipple ;-)

At SS, G. Wright the obvious choice of the '70s and Jennings in the '90s. Not completely sure about the '80s, maybe Ward, maybe Glasscock. But like McPhee he just outlasted the competition rather than clearly outclassing them.

Right now I have Barnes and Wright as the best based on peak value.

Tom H -- There was an extensive discussion of the ABC guys on the 1st basemen thread; I gave my reasons for Connor there; I won't repeat them here. On dispersing them: I had them ranked 1,2,3 overall before I decided to give a "bonus" for being "best" at a position; it dropped A&B some and it didn't get Sutton onto my ballot; I'm not certain I'm going to retain this.

Andrew Siegel -- Clarkson is the best of his generation. I think his peak impact is diluted relative to Radbourn or Spalding due to the increased schedule length by then (126 and 140). He's in that group that just misses my ballot.

Marc -- Yes, Glasscock and McPhee have some similarities. But Glasscock has a much higher peak (this isn't hard because McPhee has no peak). Using WARP-3, Glasscock has 3 seasons at 10 or higher; Anson has 1, Brouthers has 1, Connor has 3, Barnes has 4. I'll admit, he's very inconsistent, but when he's on, he's great.

I just wish I had more knowledge about his not-so-good-seasons. He wasn't a model citizen. He was temporarily blacklisted because he jumped to the Union Association but that didn't last long due to an amnesty. He didn't jump to the Player's League and was condemned by other players for that. He apparently had a temper; BaseballLibrary lists a drunken brawl with the police, and a bat throwing incident (retaliation for head-hunting) amongst his noteworthy items.

Dan b, you seriously want Mike Tiernan ahead of G. Wright?

Yes. I created a ranking of leading players (non-pitchers) on our 1906 ballot using WS with a timeline adjustment for play prior to 1893 in addition to adjustments for length of schedule and play in the AA. Tiernan ranked 5th behind the ABC 1st basemen and O?Rourke. I slotted him as low as I did on my ballot out of deference to the discussion on the right field thread where consensus put him behind Kelly and Thompson although he was seen as nearly identical to Thompson and we were entertained with a spirited debate of Kelly v. Thompson.

If I were to change my ballot now, it would be to drop Gore and add Glasscock in the 15 spot.

Here is my list - preliminary. I haven't looked at what everyone else has said, which I think will change this... nor have I thought through my timeline adjustment exactly so that may move guys up or down.

One thing I noticed while compiling my list is how close together the eight best pitchers of the 1880s are. Clarkson, Keefe, Radbourn, Welch, Galvin, Caruthers, Mullane, and McCormick... there's hardly anything at all to choose between these guys and Grasshopper Jim Whitney is right behind them, maybe even with them.

It's interesting to look at Black Ink scores for these guys... Clarkson and Keefe are way out in front with 60 and 58, McCormick had 40 and Radbourn 35. Smiling Mickey Welch only has 3, and I think that's a solid reason to put him below the other guys. Whitney is at 28, and Caruthers and Mullane at just 27 and 25 respectively, and they played a lot in the AA. Man, they are all so close! It doesn't matter *that* much to me, as I give all these guys a significant penalty due to pitching completely before 1893 and so they are down at the bottom of my ballot.

1. Anson

I forgot to mention Frank Grant, who I consider to be about even with O'Neill and Dunlap - who is also in that last-mentioned group.

My preliminary 15:

BROUTHERS

In case it may be helpful, here's the tally of the 12 preliminary ballots posted so far.

1 Roger Connor 275

It is imperative that in the actual ballot, we ask people to use both first and last names... otherwise we're going to get votes for "Stovey" and the like.

Thanks for compiling those ballots, Rob! Fascinating look.

Mark,

I must admit I took the easy way out on Barnes, which is to say I just have no freakin' clue what to do with him. He led the league in freakin' everything (you can't talk him down just because of the fair-foul bunt, I think that's mistake by James... he led his league in *slugging percentage* three times); his defense was superb at second and average at short; he was a great player for one year in the NL and an OK (average) player for three more.

In the end, I simply didn't consider him in the same depth. He played 499 career games, which is ridiculously low for a position player; I know the seasons were shorter but again I don't really know how to adjust for that. (To take this objection to ridiculous extremes, his ninth most-similar player is Jose Vidro. If Vidro's career ended today, would I consider him?)

He played all his career in bad leagues, and often was the dominant player. He has Buzz Arlett's problem in that respect, and frankly I don't know what I am going to do with him either. Rube Foster is going to have similar issues.

I'm still going to keep an open mind on this, and when my final ballot goes in (or more likely several votes down the road) I might begin to rank him. But for now, I just can't make a worthwhile comparison.

I am troubled in turn by Mark M's comment. HOM voters are not permitted to vote "strategically" (vote against their own views in order to effect a more preferred group outcome). Any voter voting strategically is subject to eviction from the group.

One of the major benefits of doing the HOM via an internet discussion group is to learn from others' views. For example, if many voters have Player X rated much higher than I do, then it behooves me to reconsider my view of Player X. Maybe I have overlooked some things, maybe I have some biases that I am unaware of, etc.

This does not mean that people have to, or even should, change their views. But I think the project will be infinitely better if people are open to reconsidering their views rather than simply sticking with their previous views.

If we want to do away with the back and forth of the discussions, then we would simply have each voter submit their "master" all-time rankings. A computer could then constitute each voter's ballots and spit out the year-by-year HOM selections. This is not what we have in mind.

Speaking as a veteran of the Baseball Survivor internet exercise, I can vouch for the fact that members of that group learned a great deal from the discussions and modified their votes on an on-going basis based upon those discussions. This was definitely a good thing and made the exercise a great deal of fun for everybody.

"By the way, I noted that MattB had Buck Ewing listed twice (both 4th and 9th) on his ballot."

Hm. I guess that's why we make preliminary ballots. Maybe when it's time to vote for real I'll just add them together and make him 13th!

I agree that Rob's submission of a "composite ballot" was helpful. The hardest parts of assembling a ballot are (1) making sure you didn't miss anyone important through oversight and (2) ranking the people you do include.

It would be practically impossible to compare every player individually against every other player, so it is useful to have a composite "spot check".

I see, for instance, that Amos Rusie and Jack Glasscock (9 and 11) are the two highest ranking players that did not make my Top 15 ballot. I will give more attention to those two while deciding who to fill in my final ballot with (I have an extra space, having voted for Buck Ewing twice!) but will end up, of course, chosing the person I find most qualified.

I also note that Ross Barnes (who did not make my ballot) is polling higher than George Wright (who did). This is interesting to me, because I assembled my ballot primarily by making Top 5 lists for each position and then choosing the best candidate from those on top of each list. When I made my last choice, George Wright was on top of his list, but Ross Barnes was buried below Richardson and McPhee. When I compare them against each other, though, Barnes has the higher peak, and I'm not sure Wright makes up for it on length and pre-1871 service.

This is going to make me completely rethink how I draw up my ballot. Barnes was the best player in his league. Richardson or McPhee was the best second baseman (for career) in the 19th Century, but was never one of the top 5 or so players in their league. I will have to re-focus on how much weight to give dominance.

Barnes on defense... he was great. A big part of defense in those days was not making errors... Barnes was excellent at that; over his career he made 23% fewer errors at second than the league average *and* he was even better than that in his best years. He also had excellent range factors, but who knows how much Spalding (who pitched all the time for Boston) had to do with that.

Wright... he made 26% fewer errors at second than the league average over his career, and his RFs are also good. Both RFs might be indicative of Spalding being a groundball pitcher; it's very hard to disentangle old fielding numbers.

The best case for including Grant or Stovey is the glowing accounts of some of their contemporaries. Unfortunately, for me there's just not enough of that to make the inclusion. The statistical record for both is thin at best... what there is supports the notion that both were very good players. Grant is described on several occasions as one of the best second basemen in baseball, possibly the best. He led the IL (a very good league) in hitting at 20, in home runs at 21; that's quite an achievement.

I'd say that Grant (on his day) was probably near the equal of Hardy Richardson, and at least as good as Dunlap.

Stovey had one terrific yearin the IL, going 34-14 (and also hitting well). He was probably the best player on a black team ("Genuine" Cuban Giants) in the 1880s and early 1890s... but frankly the team wasn't all that good - they couldn't even dominate the Middle States League, a not-very-good minor league that they played in for three seasons.

As for Hines, I'm pretty sure that a lot of his contemporaries thought he was stark raving nuts. (Re the Washington Monument stunt) :)

Another point in favor of a composite ballot (I think) is that it helps us know in what direction to lobby. (Although there will likely be more than 12 ballots cast. I'd be interested to see how the final one compares to the first 12 preliminaries).

For example, King Kelly is currently #6, just off the cut-off. He's also pretty widely dispersed on many ballots (on 11 of 12 ballots, but placed as high as third and as low as 13th). In fact, he is #6, even though only 3 voters picked him higher than sixth (one #3 and two #5s). If I were a big Kelly supporter (I'm not -- I had him 11th), this would be a good time for some pro-Kelly posts. If you convince a couple of people to bump him up a few notches on their ballots, then he'd have the honor of being a "first ballot HoMer."

I am also very curious to see whether, as the votes progress, the ballots become more or less consistent. My guess is less so. When criticizing the Hall of Fame, the few sore thumbs that stick out are easy to rally around. Assuming that we all have similar (although certainly not indentical) criteria for electing a HoMer, in any given year in which a superstar is not added to the ballot, the elections will become a debate on the best of borderline candidates from throughout an increasingly larger slice of baseball history.

The elections will never be Barry Bonds v. Willie Mays, but they may increasingly become Hardy Richardson v. Ken Boyer.

I don't see Barnes, Spalding and Wright as ability vs. value issues. I try to rate and rank based on value, not ability. Ability is quicksand. So the question is not what their abilities were (at least not any more than for a player from any other era), the question is how much value they had relative to the opportunity to create value. Their situation is more analogous to that of players who missed time to fight in WWII or due to the color line. How much (if any) additional value do we allow for them because of the unusual circumstance that constrained their value?

In the case of Ted Williams, I am willing to infer a lot of additional value. As Bill James says, it's not a question of whether he would have been a great player in 1943-44-45. He WAS a great player. In the cases of Jackie Robinson or Minnie Minoso the answer is less clear; we don't really know how great they WERE, much less woulda/coulda been, before they finally got to the majors.

Finally, for the NA players, specifically Barnes, Spalding and Wright, we know that they were great players 1871-75 (and a little bit beyond). We know that all three played and played well prior to 1871, though we don't know if they were "great." Well, I think Wright was. But in any event, I don't tend to hold their short seasons against them (adjust to 162) and I tend to rate them against their contemporaries as to their short careers. That is to say that short careers were common, but a few like Anson and Deacon White transcended that. But a 10 year career in the 1870s is not as big a drawback as it is beginning soon afterward. Wright in particular was 24 in 1871 and could easily have racked up big numbers for four years prior. Barnes and Spalding were 21 in 1871 and probably would not have done a whole lot, though, again, they did in fact play before 1871.

So I suppose in the final analysis it really is about ability as well as value. The analysis sort of takes the form of: how much value did they have the ability to create if given an opportunity more like what players of other eras had. Obviously, it ain't easy, and there's only a few circumstances in which I would worry about this stuff--like I said, WWII, segregation and the 1860s-70s.

Oh, and a (highly) provisional ballot.

1. Dan Brouthers

Re. AA, apologies to whoever posted this originally (in other words, the "I" in "I referenced all to 1881" is not in fact I) but I copied it and it is extremely helpful. Note that even the NL was not permanently as good as it had been in 1881 until 1891, and it was better in the late '70s than it was in 1884-'85-'86 and '90. The AA was better than the NL in 1886 and, of course, the elite AA teams competed very effectively in the Temple Cups.

1800s lg BWA

I do want to emphasize this. I don't see any right or wrong ballots posted here.

I drew up 4 different ballots to create a composite ballot: by WARP-3 career value, WARP-3 "peak" value, PDWS (its Win Shares based) career value, and PDWS "peak" value. WARP-3 has an 80's bias due to it's timeline; PDWS works for Boston's NA stars at Spalding's expense.

There are 20 spots in the top 5 of those four ballots, and I have 16 different candidates mentioned there. I think each has a case that can be made for inclusion, and most have a weakness that can be picked on when arguing for somebody else. It's not at all cut-and-dried, because it very much depends on the definition of "value" used, emphasis on peak vs career, opinion on quality of opposition, etc. There's a lot of room for subjectivity and opinion.

There is no Honus Wagner here, that only an idiot can ignore, demanding to be let in. I think the best we can do is advocate our own positions, and point out inconsistencies in each other's reasoning (if you emphasize career and omit O'Rourke, you might be called on that; if you emphasize peak but don't have Radbourn, that's worth asking about.)

This exercise has made me appreciate better why the original Veterans Committee in 1936 failed to elect anybody from the 19th century because a 75% consensus was required; they didn't clear that hurdle in their first ballot and then never got a 2nd chance.

I went over my first preliminary ballot and realized I had never factored in NA estimates for my (modified) career Win Share numbers. This boosted some of the 1870's guys a lot (and pushed some other guys off the WS career ballot). I also decided to not penalize positional abundance (value is value). The top dozen guys are all pretty close; I could change my mind again by this weekend.

My current preliminary ballot:

1) R. Connor (2: 88,89 NYG)

To answer how I am handling NA performances, I would say that I am discounting them by about 1/3. I have also tried to credit pre-1871 performance, and I discount them by about 1/2.

I have reviewed the case for Ross Barnes and am now confident that I will include him on my first ballot, but he will probably be around 13 or 14.

The one player I am having the most trouble with is Amos Rusie. I really want to list him high on my ballot as being the first great modern pitcher (at least the first to retire). But there are so many position players with long productive careers that I need to place above Rusie. My current compromise is to rank Rusie around the 12th spot. Anybody want to make a compelling argument why Rusie should be higher?

Joe wrote:

"I lean with Marc on the vote tallying. It is helpful, but it really opens up the can of worms for strategic voting."

This may be correct. On the other hand, it will only be an issue on the first ballot. Subsequently, the vote tallying will be there for all to see in the results of the last election.

Ed, it doesn't work that way. When you are taking points away from pitchers pre-1893 (or 1884 (overhand delivery) or any other significant milestone) you should be giving most of the extra credit to the fielders, not the hitters, since the fielders assume a consequently greater role in run prevention.

But yes, in fact, I do award a small part of the decrease in pitcher value to the hitters, because I just have a hunch that the variability of hitters was larger then, and hence the differences in offense are more important.

Rob, the argument for Rusie being higher is his "high-peak" dominance over the league, his two years "in the wilderness" due to the Freedman dispute (I give him a small boost for that but not too much since his arm was probably about to fall off), his incredible reputation, his large numbers of innings, the strikeouts, and so on.

The best argument against him is that he was the Nolan Ryan of his day, and in fact reputationally overrated. A lot of Rusie's "pro" arguments can also be made for Wild Bill Hutchison, and ain't nobody making arguments for him. Rusie also pitched only a little after 1892; nearly half his innings were before 1893.

I had him eighth, which if anything is too high.

MattB - Good point. Given our open forum style of voting, any of us could compile our own tally anyways. Thanks to Rob for saving me the trouble.

ed - I will respond to your comments this evening.

I agree with Craig on the discounting issues. Keep in mind that WS is based on a 20th century-centric (say that 3 times fast) model in which we believe the right equilibrium of hitting, pitching and fielding seems to have been reached. There are dramatic changes in the balance of offense and defense at different times (high offense in mid-1890s, 1930s, 1990s; low offense 1900-1919, 1960s). The model seems able to handle the variations, though we all make subjective judgements about dead ball pitchers and 1930s sluggers even after looking at the OPS+ or ERA+ or other league-adjusted numbers.

Post-1893 hitters don't need to be discounted because post-1893 the model has their value pegged more or less correctly. There's a good argument to be made, however, that pre-1893 the model begins to fail, so you adjust going backward. Or to put it another way, compared to pre-1893, the hitters ARE being discounted by virtue of the pitcher discount being removed (it's just that compared to the 20th century, they're not).

Does that make sense?

I did consider Caruthers' hitting very strongly (as well as the hitting of other pitchers). Jim Whitney gets a boost because of his hitting as well, though not to the degree that Caruthers does.

What works against Caruthers is how short his career was by comparison with the others. He pitched about 40% less than the other members of the "Group of Eight", and that militates against him, and he wasn't the most effective of the group either although his W-L record is outstanding. I think Clarkson, Keefe, and Radbourn were more effective and given that they pitched much, much more than Caruthers, they get the nod for me despite Caruthers' hitting.

In a sense, Caruthers-as-hitter had the equivalent of four full seasons of very effective play... as good as his pitching was. If you were to add four solid years as an 1880s pitcher to his resume, he would have about 4200-4300 innings in total... which would put him just a hair below the other members of the Group of Eight, at a marginally higher effectiveness. I would say that puts him in front of McCormick, and Mullane, but still leaves him behind Clarkson, Keefe, and Radbourn. This is in career value, of course.

Galvin is the weird one in all this, because he pitched a really, really long time.

I forgot to mention Guy Hecker in all the excitement. Hecker also gets a significant hitting boost, and comes close to the bottom of the "Group of Eight" (like Whitney) as a result.

One of the reasons I discount these guys a tiny little bit, actually, is that there are just *so* many players like them. It's not just the ten I already mentioned (Hecker, Whitney, Mullane, McCormick, Caruthers, Galvin, Welch, Radbourn, Keefe, Clarkson)... there's Will White, and Tommy Bond, and Charlie Buffinton, and Clark Griffith (a little later on, it's true), and Larry Corcoran...

Yeah, that's the problem with these 1880s pitchers, there's just so many of them. By the time you get down the list a ways you're looking at guys who are comparable in my opinion to the dead ball pitchers with the gawdy ERAs--the Ed Ruelbachs and Hippo Vaughans and Joe Woods and Doc Whites and Addie Josses and Jack Pfiesters and Eddie Cicottes and Jeff Pfeffers and etc. etc. Obviously, pitching was a pretty good deal in the '80s and 1901-19, and the stats compiled, whether it's 40-50 wins or an ERA <2, have to be discounted. Just like HR hitters of the 1990s. I don't care if Fred McGriff hits 500 HR, that doesn't make him Jimmie Foxx.

Some people overreact, however, and say that John Clarkson, e.g., can't be taken seriously. I would say that the best at any given time deserve consideration, but you can't isolate on anybody's numbers across generations and leap to any conclusions. The Heckers and Hutchisons and Corcorans and McCormicks (and the Tommy Bonds, who you missed) are just too numerous. They're all interesting but their numbers don't mean much, you gotta look at how they rank in their time. I'm not even sold on the Hoss. Clarkson and Keefe are pretty much it for me.

"Once we catch up, around 1960 we'll only have X number of spots available and the end result will be that the players from 1910-50 will get more spots than they deserve, taking some spots from 19th Century players, while later players (who also were probably better) won't be able to take any of their spots."

I think your concern is misplaced. Even as we debate now, we are giving credit to players who thrived in the 1870s (Barnes, Wright), even recognizing that even though they may not have been as "good" (ability), they still helped their team win pennants (value).

The same should hold in later years. A 19th century player who was the best at his position for a long period of time (say, McPhee or Richardson), should be competitive with any 20th century star because they will each be considered on their own merits.

Many of those who will become eligible in 1907, 1908, etc. actually will be 19th century stars whose career leaked over into the 20th century. Is it really inappropriate to compare Cupid Childs to Richardson and McPhee? Richardson was the best second baseman of the 1880s, Childs was the best of the 1890s (retired in 1901) and McPhee was the best whose career spanned the majority of both decades.

I know that I will give all three considerably more weight for their "best of" status when comparing them against the fifth best second baseman of the 1930s.

Just as the debate on this thread has likely pushed up 1870s star Ross Barnes on my (and your) ballot despite later century stars who competed against tougher competition, a top 19th century star should remain competitive throughout the ballot.

I am prepared to try to weigh Hardy Richardson against Ken Boyer if neither have been elected in the 1970s. That should be part of the fun!

JoeDimino worry too much? Naaaaahhhh.

:-)

Eleventh hour commentary:

Despite MattB?s words of assurance, I think Joe is right to be concerned. The problem is the old Silent Majority. There are what, about 70 registered voters on the Yahoo site. How many do we frequently hear from here, maybe one-quarter? IMO, the more astute members of our electorate are most often heard from; I worry about the infamous SM. At best, they will tend to fall in line with the consensus of the public discussion; at worst they will fall in line with traditional stats and players? reputations.

I?ve hashed out something that may be helpful. Or not.

After considering some of the recent posts here, I?ve come to the opinion that we would be better off if we moved our first election back eight years, to 1898. Some of the reasoning behind this:

The argument has been made that pre-1893 (and especially pre-'84) the pitcher just served up the ball like a slow pitch softball pitcher as a means of putting the ball in play. The concept of a ball not being "in play" (Ks,HRs, whatever) were all pretty much unknown. So more of the burden of creating outs rested with the fielders.

And the fielders, not wearing gloves, were highly variable. FAs were under .900 and the difference between good fielders and bad ones was much wider than today, so the impact of a good fielder in terms of competitive differential was greater.

WS divides success into 48% offense and 52% defense, and defense into 67.5% pitching and 32.5% fielding. The pre-1893 discount doesn't change the former, only the latter.

I don't think the changes happened suddenly and dramatically in 1893 but it is as logical a time as any to use as a milestone.

If you look at historical strikeout trends, you'll note the following:

DanG wrote:

"? Is this going to change whom we ultimately elect? No, probably not; D. White, O?Rourke, Clarkson and Kelly will just get in a bit sooner. But I think we can do a clearer assessment of the pioneers without the shadows of Anson-Brouthers-Connor-Ewing-Rusie obscuring the landscape."

I think this is the main point. There will almost certainly be unexpected problems, but I do not think moving back 8 years will change who is elected, and therefore would not solve any particular problem. I think it is best to proceed as planned, and keep our minds open to a "Veterans Committee" later on if the theoretical Silent Majority rears its ugly (but quiet, and at this point still theoretical) head.

Re: pre-1893 pitchers.

I think there is something there to be discovered, but I don't think it will be found by looking at strikeout rates and DIPS type things. I think it is an as-yet-undiscovered phenomenon, and I don't know which way it will end up cutting.

Consider Al Spalding and look at his BP player card:

http://www.baseballprospectus.com/cards/spaldal01.shtml

In the second group of stats, labeled "Advanced Pitching Statistics," there are a series of Deltas. Delta H is essentially the Voros DIPS number. Al Spalding gave up 33 fewer hits on balls in play than his teammates did. Is this meaningful when he's throwing most of the innings? Maybe, maybe not. But eyeballing a bunch of the top players, there appears to be a small ability to control balls in play, but not significantly different from the Steve Carltons or Roger Clemenses.

See, e.g., Pud Galvin, 106 fewer hits on balls in play, but 102 of them can in two of his 15 seasons. Radbourn, 50 fewer hits on balls in play, but wide season to season variations. McCormick, about 8 fewer hits on balls in play per year in 500-600 IP seasons.

http://www.baseballprospectus.com/cards/galvipu01.shtml

Note, also, regarding my response to moving back the date of the first election:

Cap Anson played in 1871, the first year of the major leagues. Jim O'Rourke started in 1872.

Moving the first election back to a point where they cannot be considered seems counterintuitive when the goal is primarily to compare players against their peers.

Matt, 1) pls interpret your pitcher post in English, I think you are getting to the point here but I'm not sure. Is preternaturally good luck really a preternaturally good defense? Or something else.

2) For players who started in the 1870s but played pretty much throughout the '80s (in addition to Anson and O'Rourke I think White pretty much completes the cohort), I see their peers as the '80s guys because you have vastly more comparable numbers of games played, and thus their numbers largely reflect the style of play of the '80s. The guys who started and finished their careers for all intents and purposes in the '70s are unique in the small numbers of games played compared to the number of pennants pursued (10 years, 500 games) and unique in the patterns found in the numbers. I like to consider those short season/short career guys as a cohort and the longer career/longer seasons (recognizing we're not talking 154 games generally) as a different cohort. Another way in which the 19th century is different than those that followed.

"Matt, 1) pls interpret your pitcher post in English, I think you are getting to the point here but I'm not sure."

I'm not sure what my point is, really. I was hoping others could help with interpretation.

What the numbers seem to say is, that while the average pitcher who gives up 5 singles, 2 doubles, a walk, and homer will give up 6 runs, Al Spalding only gave up 5. (I'm making up those numbers.) This would not have anything to do with defense, unless by some chance there is such a thing as a "clutch infielder" who can stretch a couple of inches more when the game is on the line.

Essentially, what these numbers show is that Al Spalding's linear weights were different than his peers' were.

Perhaps this is because he was a "clutch pitcher" who could could "bear down" and consistently scatter hits throughout the game without letting them cluster into a single inning. This is not impossible, since assumedly pitchers could not give it their all over 600+ innings. Perhaps this is due to dumb luck. Perhaps due to other factors I have not considered, or due to an error in the statistic itself, or whatever.

If its the case, though, that Spalding didn't "really start pitching" until he had a couple of runners on, and was substantially better in those situations, then that ability is a point to his credit.

Ed, my point about balls in play was not to say you shouldn't use the concept in your analysis. It was that balls NOT in play was almost an unknown occurence then--and it has been pointed out that the same might be said right up through the 1920s or 1930s. I don't know. But certainly in the 19th century there were so few Ks and so few HRs and so few BBs (and the difference between 2 and 3 Ks is not really significant) that every time the pitcher "served up" the ball, the expectation was that it would remain in play. That's all I was trying to say.

Ed, my point about balls in play was not to say you shouldn't use the concept in your analysis. It was that balls NOT in play was almost an unknown occurence then--and it has been pointed out that the same might be said right up through the 1920s or 1930s. I don't know. But certainly in the 19th century there were so few Ks and so few HRs and so few BBs (and the difference between 2 and 3 Ks is not really significant) that every time the pitcher "served up" the ball, the expectation was that it would remain in play. That's all I was trying to say.

Also I don't think the "dramatic upswing in the number of bunts" happen after the mound distance changed. I mean with the league creating so much run with mound distance further, I don't think teams would try to manufacture runs with bunts.

I had the year wrong with the rule change; it was 1894, not 1895, after one year at the new pitching distance. I don't think the rules committee was just looking for something to tinker with; I think they were reacting to a significant change in the game. Batters weren't dragging bunts for sacrifices; they were dragging bunts for singles, taking what the defense was not yet prepared to stop. Trading outs for bases using the sacrifice bunt became a necessity instead of a choice when the offense dried up a few years later.

Making a foul bunt into a strike was a radical concept then. Strikes were when you swung and missed, or the umpire thought you should have swung. Foul balls were not strikes (unless the umpire felt like penalizing you for delay of game, hitting deliberate foul balls; I don't think that rule was used too often, but I could be wrong). So they penalized bunting down the line by making it a strike if it rolled foul which limited your attempts at laying down the perfect bunt.

(and the difference between 2 and 3 Ks is not really significant)

And the difference between 4 (1880's) and 6 (1990's) is? I'm not trying to pick on anything here. My point is that pitching change has been evolutionary (James has made that point many times in different essays). There is no magic date at which the game becomes suddenly "modern".

Refresh my memory on what "1898" means. If I played in 1893 and then retired, am I eligible or not?

Not to sound too conspiratorial, but if 1893 and later players are excluded, there go 9 out of 10 players on the "composite top 10" on this thread.

That leaves one eligible, and look who his b-r sponsor is:

http://www.baseball-reference.com/w/whitede01.shtml

(Cue spooky music here)

Not that he wouldn't get in anyway . . .

Like I said a few months ago, how you structure the voting process will inevitably effect who is elected. It's just unavoidable.

Of course, it's your game and we can play however you want, but it would be a shame to waste all the work we've done on ranking the top players. Months of Anson v. Brouthers v. Connor fun goes down the drain if they all become eligible in different years.

It would certainly be a shame to push back the first election any further than opening day (it has been over a year already), especially since it looks like there is building momentum now, which cannot be started and stopped at will.

This is something that affects every championship team (and cellar dwellar) to some degree or another. The more extreme the team, the larger the effect.

The "park factors" calculated in Total Baseball also include a "quality-of-opposition" calculation. (The description in my old Total Baseball (1993) assumes that teams are playing balanced schedules; I don't know if they attempt anything more sophisticated for the NA.) That's why there is a separate BPF and PPF; the batters and pitchers face potentially different quality-of-opposition when the team is the best offense in the league and paired with a sub-par defense - the pitchers catch a break while the batters do not. (I think similar "park factors" are used at baseball-reference.com.) So OPS+ and ERA+ have it built in, assuming that the math is correct.

To the best of my knowledge, Bill James does not compensate for quality-of-opposition when calculating Win Shares. (One reason why his park factors are different.) Severely unbalanced schedules may cause WS a small problem if the team would have won a game or two more or less than it actually did; I don't think balanced schedules have a problem because they are inherently fair at the team level. (Poor quality opposition inflates the stats but not the win total.)

WARP-3 calculations are not public so who knows how they compensate for this.

Sorry, I was referring back to Tom H's post.

On changing our hypothetical start point, whatever you guys decide is OK. I'm enjoying the process; who actually gets elected on the fringe doesn't matter much to me.