Prelim

Mostly just a shifting upwards of my previous ballots. Changes are noted.

1. Sam Crawford (2,x,x)

I'm in the process of reconsideration, though so far I haven't found anybody other than the usual suspects in the backlog who is crying out for election. Actually this year barely qualifies as a backlog year, since Wahoo Sam (who is high on my ballot) and Getty Eddie (who is not) look a lot like shoo-ins. So I've got another two weeks to complete the reconsideration process.

1. Bob Caruthers (2-2-1-3-3 last year)--with Wagner gone, Parisian Bob and Hughie Jennings have the top two available peaks. Made my PHoM way back in 1905.

2. Sam Crawford (x-2)--not really a backlogger, a first ballot winner almost any other time. Makes my PHoM this year.

3. Dickey Pearce (1-1-3-4-4)--second best player in baseball history as of 1870. Made my PHoM in 1913.

4. Harry Wright (4-4-4-5-7)--third best player in baseball history as of 1870. Made my PHoM in 1914.

5. Jimmy Sheckard (15-9-8-9-6)--one of the big surprises of this whole project. The guy's peak is higher than Sam Crawford's for chrissakes. Edges out a couple of the other backlog of already elected HoMers (Joe Kelley and Eddie Collins) to go into my PHoM this year.

6. Sam Thompson (5-5-5-6-8)--made my PHoM way back in 1905.

7. Charley Jones (6-7-7-7-9)--made my PHoM way back in 1921.

8. Grant Johnson (X-6-10-5)--still wavering a bit. My analysis (so far) of Negro League 1B numbers hasn't encouraged me about the overall quality of play in the League (or, rather, among the barnstormers), but OTOH Johnson was not a 1B nor part of that analysis. I'll get to the SSs in the next couple of weeks and think a lot more about Grant.

9. Hughie Jennings (7-11-10-12-10)--the best peak along with Caruthers.

10.Ed Williamson (8-8-9-11-11)--I'm tryin' to keep hope alive here. Pretty much an equivalent career to Jimmy Collins, arguably better. Not in my PHoM yet but should make it during the backlog years.

11. Mordecai Brown (x-8-12)--I may have to take up Paul Wendt's suggestion and go to a pitcher quota. My system doesn't make much of an allowance for the shorter primes and careers of most pitchers, unless he's got an awesome peak like Caruthers. All part of my reconsideration process.

12. Lip Pike (11-12-14-14-14)--I think I've voted for Pike more than anybody, every election except one since 1898, but he's not in my PHoM.

13. Fred Dunlap (14-x-x-x-15T)--one of the most underrated players in this process. I keep finding reasons to discount his accomplishments, even AFTER I discount his UA season in '84 by 35 percent or so. I'm in the process of reconsidering 2Bs right now, and I'm not sure I can justify discounting him that much. Was Grant "the black Dunlap" or was Dunlap "the white Grant"?

14. Cupid Childs (x-10-12-13-15T)--also underrated. Can't convince myself Grant was better.

15. Joe McGinnity (10-x-x-15-13)--see Brown.

16. Plank 17. Grant 18. Wallace 19. Browning 20. Monroe

...with Wagner gone, Parisian Bob and Hughie Jennings have the top two available peaks.

So if peak is what gets Caruthers his #1 slot (which it must, since it can't be his career), why is Jennings so far down your ballot? Just curious.

Jimmy Collins, Eddie Collins, O'Rourke, Kelly, Delahanty, they're all the same to me.

Re. Caruthers and Jennings, first on peak: While Caruthers is #1 and Jennings is #2, Parisian Bob beats Jennings pretty handily. Jennings is closer to #3-4-5 than to Caruthers, depending on the measure.

On prime, they're fairly close but again the edge goes to Caruthers. Bob's prime is 5-8 years depending on the method, Jenning's is 5-7, and Caruthers has a slightly better rate.

And for career, again they're pretty close. I have Caruthers 261-232 on adjWS and Jennings 105-104 on adjWARP1.

So overall, Caruthers beats Jennings on better than 2/3 of my measures. And hey, is #9 "so low"? Two players might seem fairly comparable, but that doesn't mean that a bunch of other guys can't slip in between. If I needed a qualitative tie-breaker, it would be that Caruthers was a more unique talent.

Prelim:

1) Sam Crawford

New candidates are nowhere near the ballot, only Schulte makes my top 1000. Wagner moves off and Welch is back on. Yawn.

1) Crawford--Rumor has it he?s tiffed that we picked 1923 to be a one-electee year, might boycott the induction ceremonies.

John,

IMHO Doyle and Magee are the two best 20th century players not in the HOF (among those who were or have been eligible for awhile anyway, meaning not nec. including players of the '70s and '80s). Yet it appears to me that WS and WARP don't think much of Doyle. Magee OTOH looks OK. But I am trying not to get my hopes up too high for either of them just yet. And for that matter, we've got a good backlog, and a wild guess is that Magee will enter my ballot at #5, Doyle at #12. But both should get elected eventually. Doyle in particular might take until the second real drought period in the '60s.

Magee and Doyle won't make your ballot? That's pretty surprising to me, especially if you take 2 off your ballot for the '24 and '25 elections. I wouldn't be surpised if Magee and Doyle are the electees in 1926.

Obviously, they were terrific players, but they weren't truly dominating at their position, plus the backlog is quite impressive with names such as Thompson, Pike, Tiernan, Ryan, Jennings, etc. I don't see Magee and Doyle as superior to the above names (unless you are timelining).

I'd be shocked if they made it in '26.

Jackson and Cicotte won't be on my ballot in '26 for different reasons, of course.

IMHO Doyle and Magee are the two best 20th century players not in the HOF (among those who were or have been eligible for awhile anyway, meaning not nec. including players of the '70s and '80s).

Of the top of my head, I'd take Hack or Groh over them.

Jim Sp,

9) Griffith?Comp is Marichal, plus he could hit.

Alright, I'll bite on this one. Marichal and Griffith are that close? ERA+ numbers are similar as well as IP. WARP tells a different story with Juan coming out on top. Bill James Historical Abstract has Marichal much higher when ranking pitchers but his bias is toward 20th century players.

Still, I don't see it - maybe you can shed some light on it. I don't want to underrating the Old Fox if there are something that's a big "vote for me". Being equivalent to Marichal wouldn't hurt...

Alright, I'll bite on this one. Marichal and Griffith are that close? ERA+ numbers are similar as well as IP. WARP tells a different story with Juan coming out on top. Bill James Historical Abstract has Marichal much higher when ranking pitchers but his bias is toward 20th century players.

FWIW, Marichal had the greatest run support of any 20th century HoF pitcher I've found - RSI is 115.3 (for comparison Vic Raschi was at 115.6). Don't have season-by-season stuff with me, nor do I have a page open to the yahoo group file with all the RSI stuff avaible, but IIRC, Marichal gained more wins through his run support than any other pitcher I've found. I think he gained 22-23 wins. More like a 220-165 pitcher. Of course the problem here is that instead of building Griffith up I'm tearing Marichal down. Ah well, still the single biggest surprise I've found in the 90,000+ games I've RSId.

Al, that's all there is to the comparison, plus their W/L records are similar. I list Marichal and Griffith as comparable because they have similar ERA+, IP, and W/L career numbers. In context I believe Marichal's ERA+/IP numbers are more impressive because IP counts were higher when Griffith pitched, and extreme ERA+ performances were more common when Griffith pitched. FWIW, I have Marichal #44 and Griffith #65 in my ranking of pitchers. This week, anyhow.

It's worth noting that Marichal pitched for better teams and got much better run support, on the other hand.

Ups & downs. . . .

Grant Johnson. Feel better about him than ever. Leaps over McGinnity & Plank.

Jake Beckley's going up this week. Previously, my attitude on the lack of 1Bers in the HoM has been to shrug & say (I think I even posted this once) - no big concern, it won't last because there's so many future ones coming up. But I realized that that attitude is antitheitcal to this process. It's 1924 & dang it we don't know our Hank from our Greenberg yet. As is, there's a shortage of 1Bers, I think Beckley is a more than deserving candidate, & he's clearly the best 1Bman between ABC & probably that Sisler fella that's been tearing things up for a few years. So Beckley's going up.

Clark Griffith - first up, then down. The more I thought about it all last week, the more I figured he probably was better than Brown. Similar ERA+, Griffith had more innings, & (this is a factor I ignored in the '23 vote) played in a more competitive time He had the AL '01, but Brown had the Federal League. Between the three adjustments that hurt Brown (unearned runs, RSI, & defense) he ends up a pitcher with a slight edge in ERA+ & a deficency in IP to Griffith. Then, however, I did the rest of Griffith's RSIs & he came out to have better run support than I thought. That being said, I still may put him a little over Brown.

Mickey Welch - up. OK, so it wasn't the team that got him to 300 wins. The problem now is trying to make sense of 1880s pitchers. Normally, with these guys I'm more likely to value prime/peak because I'm less sure that a guy with great career numbers (like say, Pud Galvin) was really great or just slightly-better-than-mediocore with lots of durability (and back then if you could last 2 league-average years, before blowing your arm out, you could end up with great numbers). How impressive was winning 300 back in the 1880s? Well, Clarkson, Radbourn, Welch, Keefe, & Galvin all won 300 (& Mullane would've had he played in '85) & they all had their careers centered almost entirely in that decade. Me thinks there's a structural issue at work here - a combo of the longer schedules & the shorter distance to the plate allowed pitchers to rack up huge numbers of innings.

Still, new info does help Welch. He wasn't a team-aided 300 game winner, but an actual 300 game winner. But where does that leave him? The best analogy I see for him is a pitching version of George Van Haltren - never the best, but in that glut below the best as he was very good & effective for about a decade. Right now I have him penciled in above Brown & Griffith, but still below McGinnity.

Tony Mullane. Poor man's Mickey Welch. Similar numbers but in a crappier league. Can say he would've won 300 with 1885, but part of me wants to argue that that's counterbalanced by 1882. AA was new that year & Mullane wasn't very good. In a normal year with established leagues would he get as many starts as he did? Maybe. Hurt by offense, but helped by defense.

Bob Caruthers. Up. That gaudy winning percentage was mostly Caruthers-created as he's still got a .644 winning percentage with RSI adjusted info (& that doesn't give him any credit for his bat). Similar in innings to Brown & Griffith (though less than either). I have a hard time gauging 1880s pitcher & a hard time with multi-threat guys like Caruthers, so he's really tricky for me. But for now, I've got him just ahead of Welch.

Johnny Evers - thought about him more & figure he belongs below Childs but ahead of Bresnhan. So that's where he'll stay.

Jack Clement. Looked at this good hitting catcher. I'd put him a little over Bresnahan. Couldn't hit as well as RB, but he did more of it at the right position.

New tentative ballot:

It's 1924 & dang it we don't know our Hank from our Greenberg yet. As is, there's a shortage of 1Bers, I think Beckley is a more than deserving candidate, & he's clearly the best 1Bman between ABC & probably that Sisler fella that's been tearing things up for a few years. So Beckley's going up.

Chris, should this reasoning be applied to catchers also? In other words, there is a dearth of catchers in the HoM now too, but you've got Clements and Bresnahan at 30 and 32, respectively, with no chance to get elected, and there are fewer catchers in the hopper than first baseman.

Chris and Jeff, over here! I'm over here! Look at me!

John, I like you and had you on my ballot for the first election, but the big guys during your time were named White and McVey.

Me:

It's 1924 & dang it we don't know our Hank from our Greenberg yet.

For some reason this line made me want to look up Wally Pipp. Pipp's been a good player - sort of the Wally Joyner or Eric Karros of his time. But he was 30 this past season, and turned in an OPS+ of 95. Even though the Yankees ran away with the pennant, they'd better be thinking about a replacement for Pipp. There was this local kid out of Columbia University that they gave a few games to - what's his name, anyway? Something like "Gregg"? Could he be the answer?

That gloom I was speading about Babe Ruth on my ballot last year? Maybe I'd better take it back. 152 games of .393/.545/.764 ? And more in the World Series? What can you say?

On the subject of first basemen: if you'd asked us a year ago, we'd have been sure that George Sisler would blow away everyone we've seen for a long time. But after turning in a .420/.467/.594 (OPS+ 170) at the age of 29, he just missed the entire 1923 season. What's the word - will he ever be back? Perhaps it would be best to just assume his career is over. If that's the case, we're going to have to figure out where to place him on the "short-career hitter" scale with Flick, Donlin, Titus, and Stone.

Q: Is it possible that the Cubbies were SO good, that tehre is some distortion in the extreme so pythag'd W-L records won't match right?

Not that I know of, but I will say that a lot of the best pitchers in baseball history seem to come off as underachievers when you crunch the numbers. Biggest underachiever ever? Walter Johnson by the numbers. Tim Keefe's way up there. Warren Spahn's in the top ten. Brown's a little behind them (as is Vic Willis).

No idea why, but I don't think it's due to the quality of the Cubs or their defense because it's something that happens elsewhere.

a lot of the best pitchers in baseball history seem to come off as underachievers when you crunch the numbers.

What I have to say isn't a direct response, but may be vaguely related. I was using Retrosheet to look up Bob Gibson's 1968 season. This is personal. I spent that summer attached to a radio (and TV, when possible), and I remember it vividly, even 36 years later.

Gibson's season was not at all typical of his career, but his career isn't the point here - I'm dealing with a smaller sample, just the season. Using "underacheiver" the way I think Chris is defining the word, then Gibson had an underacheiving season. His ERA+ of 258 led to a W-L record of only 22-9. If you turn that ERA+ into a record with 31 decisions, he would have been 27-4. Alternatively, as I prefer to do it, turning his 232 RA+ into a record with 304.7/9 decisions, he would have been 29-5.

Now, Chris is going to tell me that Gibson's RSI in 1968 was lousy, but I'm going to guess that he'll tell me that it wasn't lousy enough to account for the 5 or more wins that his record is missing. In the 34 games that Gibson pitched, the Cardinals scored 103 runs and gave up 51. That's 3.03 R/G for and 1.50 R/G against in a 3.43 league with about a 98 park factor. (That's 49 runs allowed by Gibson and 2 by his relief. Divide 304.7 by 34 to get the picture.)

Do a Pythagorean record for 34 games at 3.03 for and 1.50 against, and you get 27-7. (Actually 27.3-6.7). And still Gibson was only 22-9. (The team was 24-10.)

You can get one glimpse of what happened by dividing the season into thirds.

In Gibson's first 11 starts, the team scored a miserly 2.09 R/G while allowing 2.18. Gibson was pitching well, but his personal W-L record was a mere 4-4. (The team was 6-5 in his games.)

Gibson's next 11 starts were The Streak. First five shutouts in a row. Then, going up against Drysdale in LA with a shot at Drysdale's brand-new scoreless inning record, Gibson wild-pitched in a run in the first - but went on to a complete-game 5-1 win as his teammates knocked out Drysdale. Another shutout, then a run allowed in a late inning with a big lead, then two more shutouts, then a 7-1 win. In these 11 games, Gibson and the Cardinals allowed only 3 runs, 0.27 per game. At the same time they scored an above-league average 4.18 runs per game. Of course, they won all 11.

In Gibson's last 12 starts, the streak was over and he had about three bad games. The team scored a below-league-average 2.83 R/G, while allowing 2.00 R/G. Gibson was 7-5 in these 12 games, as was the team. There were four 1-0 games in this stretch, with Gibson winning three.

The bulk of Gibson's offensive support happened to correspond to his streak, when he didn't need much support, and was poor in the first and last thirds of the season. You could mismatch the support to a a pitcher's pattern at any level, but the special feature of being at the level Gibson was at in 1968 is that there was a big chunk of the season in which he was going to win whether he was well-supported or not. Something like that may well be behind your observation that superior pitchers are more likely to "underachieve".

Although I am quite highly appreciative of Chris's work with RSI, I do have some reservations about the idea. Is there really anything in a pitcher's job description other than "prevent runs"? His offensive support is just stuff that happens. Of course, we should adjust pitchers for defensive support as best we can, and we should adjust for the general environment (league and park). I'm less sold on adjusting for a pitcher's offensive support - but then, I've never put that much stock in actual W-L records.

If I had to vote today, Brown would be fighting for a 15 slot with many other pitchers. I realize I might be way off consensus with this, so somepne, if you wish to crunch numbers and convince me I'm off my rocker, go ahead.

Tom, I had my say on the 1923 Ballot Discussion thread in post #203. In fairness to the two Chris's, I'll also point to posts #144 and 185 in that thread. Those two pretty much agree with you, and I disagree - at least I disagree with your rating of Griffith > McGinnity > Caruthers > Brown.

Using "underacheiver" the way I think Chris is defining the word,

May not be the best word to use. I use it for lack of a better one & I wish I had one, because that word has a certain moral conotation (you shoulda done more!) that I am actually trying to steer clear of.

The fact that so many first-rate pitchers tend to win less than pythag says they should makes me wonder if there's some bug somewhere in the system. Or that a straight pythag version of RA/9IP & league average run support would be fairer.

The reason I try to point out under/overachieving is mainly to point out that there may be a bug in the system & this is how the bug affects that pitcher. Sorry if that hadn't been made clear.

The fact that so many first-rate pitchers tend to win less than pythag says they should makes me wonder if there's some bug somewhere in the system.

I'm not sure it's a bug with YOUR system, but rather, with the danger of putting too much stock in W-L records. I think it is useful information as part of the overall mix, but I wouldn't start scrapping traditional measures, Win Shares and WARP to focus solely on a modified W-L record.

If I used WARP3 and total career value, Mordecai wouldn't make my ballot. And that's not even counting his "unclutch" W-L record!

All the more reason to use WARP1, since no one knows what kind of adjustments are being made between WARP1 and WARP3. If you take Brown's 3-year peak, 5-year consecutive peak, 7-year peak and career WARP1 numbers, and compare them to pitchers who are actually in the HoF (eliminating the ridiculously good numbers like Young, Mathewson and Alexander, and the extremely bad numbers, like Marquard, Haines and Bender), he's right at the mean. He does better in WS and LWTS. Maybe you don't give him a #1 vote, but it's hard to say he isn't deserving if he's as good as the average HoFers that most agree were decent selections (Dazzy Vance, for example).

The fact that so many first-rate pitchers tend to win less than pythag says they should makes me wonder if there's some bug somewhere in the system. Or that a straight pythag version of RA/9IP & league average run support would be fairer.

I think OCF's account of Gibson's 1968 season offers a pretty good explanation of why great pitchers are more likely to underperform than overperform pythagorean expectations. My statistical skills aren't the best, but wouldn't the fact that a great pitcher's runs allowed sticks closer to the lower bound of zero mean that run variance will sometimes give the pitcher _many_ more runs than he needs to win, but the pitcher will almost never give up _many_ more runs than he receives? This would be esp. true in low-scoring environments. Would not such a pattern lead systematically to Pythagorean underperformance by great pitchers? It's not that he could have done better, but allowing more runs in blowout wins, or getting tagged for a blow-out loss or two, would help bring pythagorean expectations down into line with the actual record. If that analysis is correct (and the real stats people in the group could confirm this), then it should only be expected that the pitcher with the most career shutous and the most runs saved above average would miss his pythogorean projection by the largest margin, and that's exactly what has happened.

Thinking the matter through, it looks to me like we should expect great pitchers, esp. in low-scoring environments, to miss their Pythagorean projections somewhat. As long as we recognize that to be the norm, and to be a product of the statistical environment in which they performed, I don't think it should have much effect on how we rate individual pitchers.

I might consider, when comparing pitchers to position players and using win share values that I've derived from actual wins, adding some portion of the difference between a pitcher's pythag wins, given his run support, and his actual wins, to the pitcher's career value. That wouldn't radically shift my evaluation of pitchers, but it would give them a small boost relative to position players in my system.

The more we work on evaluating pitchers, the harder it becomes . . .

I think OCF's account of Gibson's 1968 season offers a pretty good explanation of why great pitchers are more likely to underperform than overperform pythagorean expectations. My statistical skills aren't the best, but wouldn't the fact that a great pitcher's runs allowed sticks closer to the lower bound of zero mean that run variance will sometimes give the pitcher _many_ more runs than he needs to win, but the pitcher will almost never give up _many_ more runs than he receives?

That's brillant. Seriously - it makes big heaping gobs of sense & answers a question I've been wondering about for months. We ain't in clutch hits, but screw it - I'm nominating that post for a primey.

This would be esp. true in low-scoring environments. Would not such a pattern lead systematically to Pythagorean underperformance by great pitchers?

IIRC, I large propotion of the worst underachievers were from the 1900-20 deadball era (Willis, Walsh, Brown, Plank, Johnson). And modern guys like Johnson, Clemens, & Maddux are doing OK (overachieve or at least not much of an underachieve). So that all fits it.

Stuff like this is why I love this site.

Life. Liberty. Baseballprimer.

PythaganPat is ((R + OR) ^ .286)

Now to back up my claim above that my statistical skills are limited, could Joe or someone else who knows this formula explain what it means (including what OR stands for) and how one would apply it to modify or replace the pythagorean method?

Happy April 7 FIRST ANNIVERSARY of the inaugural voting for Baseball Hall of Merit!!

Seems like only, oh, a year ago that we were debating whether Ross Barnes had enough longevity to "merit" a place in the HOM.

I think I stick to my view that Win- derived stats (by all means with appropriate adjustments) are more accurate than ERA-derived stats in assessing the Merit of pitchers.

For example, consider Pedro Martinez, 2002 and 2003. His ERA+ went up from 196 to 212 between the two years, and his IP dropped only 5%, yet any Red Sox supporter, even before Game 7, would have told you that he was much more valuable in '02 than '03 -- in the latter year, Grady Little was babying him riduclously, making him more or less a 5 inning pitcher, and his ERA after 90 pitches went through the roof. This was well reflected in his W/L record, which went from 20-4 to 14-4 -- while remainining nominally the Sox' "ace" he was becoming irrelevant to a large number of games in which he pitched, because the real crux of the game became not Pedro but the middle relievers who had to be brought in in inning 6 or 7.

Today Pedro is a useful Intimidator, to those batters still foolish enough to fear him, but without a fastball over 90 he will struggle to win 10 games, even though his ERA will still be excellent if the manager yanks him early enough. (Therefore, the Sox shouldn't re-sign him, but that's a separate discussion!)

You can pitch a lot of low-ERA inings, yet they can be worth little in helping your team win, because they're against weak lineups, or in the later innings of blowouts. Conversely, the "Pitching in a pinch" philosophy results in a higher ERA, but more Wins. So I'll stick to the latter, thank you, as my primary evaluation method.

Andrew, karl's point was that Pedro pitched just as many innings. i.e. instead of pitching 7-8 innings in 25 starts and missing starts with a sore arm, he pitched 5-6 in 35 starts. Obviously, if you incorporate his G or GS, you'd know something was different.

But, karl, is he really more valuable pitching in fewer games? Or only more valuable in those games, and therefore less valuable (in 2002) overall? What do WS and his WARP have to say? Could we not ignore ERA+ AND W-L and just consult with WS and/or WARP?

And if I missed this entire discussion, what would I have missed? It seems that you have all decided that the discussion was not helpful? Can somebody give me the 5? summary of what we now know?

No, for the same number of IP, he would be more valuable if they were concentrated in 7-9 inning starts, because the team would not then have to rely on the weakest part of its pitching corps in those starts. He could even let up in late innings of big wins, which would not affect his W/L but might increase his ERA. Pedro's value advatage over a replacement level pitcher is minimised by his enforced usage pattern. W/L reflects this; ERA doesn't.

Pedro might well at this stage be most valuable as a 70s type "fireman" -- fewer IP but more leverage

Marc wrote: And if I missed this entire discussion, what would I have missed? It seems that you have all decided that the discussion was not helpful? Can somebody give me the 5? summary of what we now know?

Marc, to which discussion were you referring? There are just a few going on now :-) .

If you are referring to the "Pythagorean wins and deadball era pitching greats" discussion, I could attempt a five-cent summary.

I think if you use the PythaganPat exponent, instead of 2 or 1.83 you won't have this problem - or if you do it won't be nearly as problematic.

PythaganPat is ((R + OR) ^ .286)

I'm really not sure if I have any idea what this means.

The normal Pythag is you take runs scored to the power of two over the sum of runs scored to the power of two & runs allowed to the power of two.

With PythaganPat - am I replacing 2 with 1.83 or with .286? I think it's 1.83, but then I have no idea what that .286 is doing at the end.

Last time I took a math class: June of 1993.

> No, for the same number of IP, he would be more valuable if they were concentrated in 7-9 inning starts,

Chris J: If I understand the math whizzes correctly, you use the PythaganPat formula, which is

(R + OR) ^ .286), or translated into words -- the sum of runs scored and runs allowed, raised to the .286 power

to calculate the proper exponent to use in the basic Pythagorean formula. This is what TomH shows in post #36:

<i>Where a team scores and allows 6 run sper game. exponent = 12^.286 = 2.035

I think I've voted for Pike more than anybody, every election except one since 1898, but he's not in my PHoM.

Not so fast! He's been on my ballot every year and probably will until 2050! Never lower than 12th (the first two years) and in my pHoM since 1908.

Philip, I meant to say that I've voted for Pike more than I've voted for any other player. Ah, the ambiguity of words, I hope it will last.

Chris, is the converse also true: That "pitching in a pinch" is also an illusion?

OK, assuming Cobb got it right, I think I understand what's being said. New question: how the heck to you figure something to the .286th power? This is more a calculater question than a math question. Or for that matter (& more importantly here) what do I type into a excel spreadsheet to do this? Right now I have in something like:

=SUM((A*A)/((A*A)+(B*B)))

How'd one type in a command for the power to the .286 in excel?

Chris, is the converse also true: That "pitching in a pinch" is also an illusion?

Maybe. It depends on what you mean by "pitching in a pinch."

It's well accepted in baseball lore, and I think welll-demonstrated statistically, that "clutch pitching" exists, that pitchers can change their strategy or raise their game, when the circumstances demand it, or order to prevent runs. If that's what pitching in a pinch means, then yes, it exists -- it's no statistical illusion.

If by "pitching in a pinch," you mean "pitching to the score" as Karlmagnus speaks of it, it's not clear to me that this ability exists, or _if it does_, that it varies in meaningful ways from one pitcher to another. If it did exist and varied meaningfully, the pythagorean formula and its more refined versions would not be as accurate as they are. It's possible that some pitchers are better at pitching to the score than others are, but there's no clear evidence about it.

Take the example of Mickey Welch. He gave up many fewer runs than his component statistics of hits, walks, strikeouts would suggest he should have. That's pitching in a pinch. However, his won-lost record does not miss (I think, one could recheck this) a proper Pythagorean projection of it based on runs scored and runs allowed. If he _is_ pitching to the score, he's not any better at doing so than pretty much any pitcher. That's my take, anyway.

that's B2**0.286 for you Fortran programmers out there.

No love for us COBOL college programmers from the eighties? :-)

Here is my prelim. No changes from last year. I still can't get over the lack of support, and indifference toward, Andrew Foster. His career is poorly documented, but there is reason to believe he had a ten year peak (approx. 1901-1910), not the two year peak (approx. 110 wins in those two years) many seem to attribute to him. He was "a" or "the" star pitcher on the best blackball team in the country in both 1901-1902 and 1907-10. Given his age it is not improbable he had a career even before 1900.

1. Sam Crawford

Re Foster:

Here is a cut and paste from outoftheshadows.net. It shows he started his career in 1897 and it went until at least 1910. If you read this and don't think this guy is a Hall of Meriter it must because you doubt the quality of play of blackball prior to 1910. That's fair enough. But how do you reconcile that with the comments of Wagner and McGraw, his actual competition against the major leagues and the fact that he is accepted as the best black pitcher of his time. I don't get it.

Young Rube's career began in 1897 with the Austin Reds of Tillotson College. The following year, he joined the semi-pro Yellowjackets in nearby Waco. The new century found Foster signing with Frank Leland's Chicago Union Giants for $40 a month and 15 cents meal money. Here, he developed a nasty screwball thrown from an unique submarine delivery.

In 1903, he headed East to play for the Cuban X-Giants. That fall, he posted four victories in a seven-game series against the Philadelphia Giants for the so-called "Colored Championship of the World." The wandering Foster switched over to the Philly team the next season and met his old teammates for bragging rights to the "colored" title. Foster accounted for both victories in the best-of-three series. He struck out 18 batters in one game, beating the current major league record of 15 set by Fred Glade of the St. Louis Browns.

Andrew Foster's reputation as a fine pitcher continued to expand with a victory over Rube Waddell and the Philadelphia Athletics in 1904, earning him the nickname of Rube. He later assisted John McGraw, New York Giants manager, with a few pitching tips for Christy Mathewson. Until 1903, Christy's won-lost record was 33 wins and 37 losses. After a sermon from Foster, Mathewson won 30,33 and 31 games, and led the league in strikeouts the next three seasons.

In 1907, he returned to Chicago and the Leland Giants, leading them to a 110-10 record, including 48 straight wins. In 1909, the Giants enter the tough integrated city league. In Foster's first eleven starts, he won eleven games, with four shutouts. Such a dominating force that season, that after one commanding victory an Indianapolis newspaper's headlines simply stated: "FOSTER PITCHED, THAT'S ALL."

By 1910, the Leland Giants were the talk of the Midwest. They christened a ball park in an all-white neighborhood near 69th and Halsted, as the Leland Giants Base Ball Park. Foster now serving as player-manager, amassed what was in his opinion the greatest team of all time. Featuring such stars as John Henry "Pop" Lloyd, the notorious streak hitter Pete Hill, Grant "Home Run" Johnson, catcher extraordinaire Bruce Petway and great pitchers like Frank Wickware and Pat Dougherty, the Leland team won 123 of 129 games. This team compelled McGraw of NY Giants fame, to announce, "If I had a bucket of whitewash that wouldn't wash off, you wouldn't have five players left tomorrow."

Your points are well-made, daryn. I'm hoping that we can focus more on Foster this year now that the "obvious" choices from last year (Wagner, Crawford, and Plank) have found their niches.

Pythagoras - Pretty good explanation here: http://gosu02.tripod.com/osusaber/id69.html

Basically, for pitchers like Bob Gibson 1968 or Pedro Martinez, the run environmnent that applies when they start a game is different (lower) than the "league average" run environment, so if you just use a "normal" Pythagoras type formula they will almost always appear to "underperform" relative to Wins and Loses.

Prelim ballot, pretty much same as last year, minus Wagner. A few guys jumbled around, especially in the 16-30 range.

1. Crawford (2)

Preliminary Ballot - Not much changing yet:

Using OWP, playing time, and defense (Win Shares/BP) for position players, applied to .500 baseline. Using Runs Saved Above Average and Support Neutral Fibonacci Wins for Pitchers.

1. SAM CRAWFORD, RF . .686 OWP, 401 RCAP, 10,596 PA?s. Def: AVERAGE.

2. MORDACAI BROWN, P . 295 RSAA, 282 Neut. Fib. Wins, 138 ERA+. Certainly benefited from good defense behind him, but '06-'10 Cubs were the BEST EVER at preventing runs, so pitchers probably deserve SOME of that credit too! Better pitcher than Walsh, Radbourn, & Galvin.

3. EDDIE PLANK, P . 278 RSAA, 288 Neut. Fib. Wins, 122 ERA+.

4. JOHN McGRAW, 3B. .727 OWP. 459 RCAP. 4,909 PAs. Def: AVERAGE. Similar to Lajoie except only 60% of RCAP, but 60% of Lajoie is still excellent. He didn?t have a long career, but he?s being discounted for low playing time way too much as he provided more value in those few appearances than all of his contemporary 3rd baseman. When he retired, he ranked 12TH all-time in Plate Appearances by 3rd basemen:

PLAYERS 1876-1906 PA OWP

Chris, my take on "pitching in a pinch" is that pretty much any decent pitcher can pitch "down"--i.e. relax, pitch below his ability--when given a big lead. OTOH I don't believe that a pitcher can be better than who he is or rise above his level of ability.

"Pitching in a pinch," IOW, is an oxymoron meaning exactly its converse--that the pitcher in question sometimes did NOT pitch to his ability. And I don't know why I would elevate a player's rating or ranking because he did that.

My understanding of Chris' comments on pitching in a pinch can be summed up by this. Kerry Wood, lord of the strikeout, can change his approach with men on base, sacrificing his natural tendency to cause whiffs for a shot at a ground ball double play. It isn't so much pitching better, but more a case of pitching differently.

But it also seems pretty well documented that certain pitchers can work at 90-92 mph and really "reach back" for something extra nasty when the situation calls for it. Pedro and Schilling get cited for this all the time (normally in the context of "in the bottom of the ninth with the bases loaded and the game on the line", but that's for another time).

Any chance of us getting vote totals up today?

Any chance of us getting vote totals up today?

Any chance of us getting vote totals up today?

In the spirit of reconsidering anyone who was overlooked, I thought this was interesting...leaders in fielding win shares among eligibles. Only Wallace, Tinker, Cross, Long, and Leach are still on the ballot. This list missed anyone with under 250 Win Shares total. Leach in particular I need to reassess, I don't think I've been giving him enough defensive credit. Looks like he was A+ at both 3B and CF.

Bill Dahlen 143.35

My response to the collection of posts on this thread from #27 to #46 is going onto the "Pitchers" thread. I'm going to re-do some calculations with using (R+OR)^.286 instead of 2 as the exponent.

Based on this thread I've readjusted the W/L records for the pitchers. I've got a list of 30 major ones on the pitchers thread here.

Evan, my position on this has always been: But how do we know who was "pitching in a pinch" 75-100-125 years ago from seasonal data?

all-time HOM point-vote-getters.

Tom,

I know that was just for fun (and, I agree, Wow!) but it is easier to get a high ERA+ in a higher scoring era. A shutout is much more valuable now than in 1905 (in terms of ERA+) and you cant give up negative runs by pitching better.

Still, it deserves a wow.

<i>Posted 8:42 p.m., April 7, 2004 (#69) - Chris J.

1924 prelim. More or less everybody moves up from last week.

I have now tabulated the available statistics for the early Negro Leaguers. When time becomes available, I will try to analyze this data in the same way that I analyze it for major leaguers. Namely: to try to quantify and blend together answers to the three core questions of the Keltner list: Was this player ever the best in baseball? Was this player ever the best at his position? Was this player good enough to play past his prime? In the meantime, IRS and work deadlines are looming...

1. Sam Crawford

After posting the last comment, I realized to my chagrin that the last paragraph could be construed as an attack on Marc. Please be assured, Marc, that I like your contrarian ballots! :-)

By my own calculation, my 1923 ballot was the second farthest from the consensus, after Marc's.

Correctly identified. Summing the last three years, you're the third furthest from consensus, behind KJOK and yest. The five candidates who contributed the largest amounts to your distance from consensus were Waddell, Bresnahan, Johnson, McGinnity, and Grant. (In my own case the five "furthest" would be Evers, Grant, Sheckard, Thompson, and Caruthers. So I got a little carried away with Evers.)

But no one is really out of line and everyone has their reasons.

I should wait a few years until be get into the "new candidate drought" years ahead before saying this, but we don't really seem to be fragmenting the candidate pool all that badly. The number of candidates getting serious consideration isn't growing very fast, if it's growing at all. The four you cite would be visible even without your support.

No offense taken, though I didn't set out to be a "contrarian" here at HoM. But I'm stickin' to my no-timeline peak-oriented ballot, I like it too.

1924 ballot. I'm fairly set this year - most everyone moves up one spot. Still mixing up the 16-30 positions so those are the people I need to firm up ordering when we start electing off our backlog.

1. Sam Crawford (2). Alltime leader in triples. On the bad side he didn't even hit 100 home runs. Top 10 in Slugging % 14 out of 15 years (1901-1915).

2. Sam Thompson (3). Personally feel a deserving HOMer. During 10 year period (1886-1895) was top 10 in league in total bases 9 times. .308 EQA, .684 OWP, man could hit a little.

3. Eddie Plank (4). Outstanding win totals with the needed underlying stats to go with them. Would be the type of pitcher any staff needs - steady, longterm option just not super. Oldest regular player in the 1915 Federal League - and still won 21 games. Whether that upgrades Plank or downgrades FL is your call.

4. Joe McGinnity (5). Quanity of work for the 1900s impressive to say the least. Helped the team by taking the ball often. Top 5 in IP for six straight years (1899-1904). Nine year span (1899-1907) of 123 ERA+ over 3235 IP. Good thing he threw sidearm with that workload.

5. Jimmy Ryan (6). He passed away this last year. Tip of the cap to a fine old ballplayer.

6. George Van Haltren (7). And whereever little Jimmy Ryan went George Van Haltren was sure to follow...

7. Three Finger Brown (8). Has stood up pretty well to analysis of Cub defense, team quality questions. Don't forget his key stat - Wins per Finger.

8. Frank Grant (9). Nothing new to report; scant evidence points to a player worthy of mention, possible induction.

9. Pete Browning (10). Don't know why I soured on him to such an extreme. Even with league discounts he swung some mean lumber. Let me throw out some numbers. Career OPS+ = 162 which puts him in company with names like Foxx, McGwire, Frank Thomas. Discount it because of AA play? At the OPS+ = 147 level you're talking Heilmann, McCovey, and Schmidt. That's some pretty lofty company. Batting Average Placement within league 1882-1891: 1, 2, 3, 1, 2, 2, 3, -, 1, 3. Had a career .341 average in a league environment of .257 . Offensive Winning Percentage - .745. His guy was a WOW when hitting. He might not be the ideal multi-dimensional player but when you are this much an outlier at part of the game its going to get a bonus from me.

10. Jimmy Sheckard (13). Another well-rounded player, successful for the great Cub teams.

11. Rube Waddell (11). His strikeout numbers were well ahead of his time. Suffered from inconsistent offensive support. Had the sizzle of a star player in contrast to the substance of Plank.

12. John McGraw (12). Limited playing time but what he did with it is nonetheless outstanding. Positional bump as well.

13. Home Run Johnson (14). Weighed the evidence, can provide some ballot support. Probably ranking him mostly as SS.

14. Bob Caruthers (15). My tug-of-war with his value continues. I'm more sure now that if push came to shove Freedom Bob deserves in before the rest below him.

15. Jake Beckley (16). Tougher and tougher to ignore with dearth of 1B for a number of years; career totals eventually add up to quite the player despite lack of peak.

Coming back in future years:

16. Tommy Leach. Career took off after Jimmy Williams exited Pittsburgh for the new American League.

1924 Preliminary Ballot

1. Sam Crawford

In Daryn's post #56 above re Rube Foster, he wrote:

He was "a" or "the" star pitcher on the best blackball team in the country in both 1901-1902 and 1907-10.

This does not match the accounts of Foster's career that I have seen. According the to the Foster bio _The Best Pitcher in Baseball_, Foster did not come north to play for the major black teams until 1902. He joined Leland's Chicago Unions at first. He had some success, but was also hit hard, and left the team mid-season to play in the Michigan State League, where he again had mixed success. He was reputed to have great stuff, but he was still learning how to pitch. Near the end of the season, he joined the Cuban X-Giants, and it was in 1903 that he really established himself as "the star pitcher" on the best black team in the country.

I give Foster credit for 1902, but not star credit. He was a rookie, making rookie mistakes.

He was certainly a star pitcher in 1909-1910, but he missed more than half of the 1909 season with a broken leg, and it's not clear that he took a leading role on the 1910 team. He was surely a gate attraction and an effective pitcher, but he wasn't pitching nearly as regularly as he was in his greatest seasons, 1903-1905.

The information in Darryl's post #57 mostly fits with the information in the biography, except for the year of Foster's arrival in the north (1902, not 1901), and the Christy Mathewson story:

He later assisted John McGraw, New York Giants manager, with a few pitching tips for Christy Mathewson. Until 1903, Christy's won-lost record was 33 wins and 37 losses. After a sermon from Foster, Mathewson won 30,33 and 31 games, and led the league in strikeouts the next three seasons.

Both the author of _Best Pitcher in Baseball_ and John Holway regard this story as apocryphal (though that doesn't stop them from telling it!), and a closer examination of Mathewson's record bears this out. He lost a lot of games in his first seasons because he was pitching for very bad teams. Given the level of offensive and defensive support he received, his record was outstanding. His won-lost record leaped in 1903 because John McGraw had arrived in New York halfway through 1902, and by 1903 he had begun to put a great team together around Mathewson.

Having looked at middle infielders recently as part of a grand "reconsideration," I am as puzzled as ever re. the appropriate placement of Frank Grant, Grant Johnson, Bill Monroe and Sol White, even among themselves. All four were apparently great players. Probably by the time the fourth of them (Johnson) retired, they were the four best black players who had completed their careers. But other than that, what exactly do we (I) know?

Well, we have "expert's" lists.

SABR Negro Leagues committee--35T. White. The other three are not listed among the top 40 and ties.

Pittsburgh Courier's all-time team (1952)--none made the first or second team, but Bill Monroe was listed as the 3rd 2B.

Holway's Complete Book of Baseball's Negro Leagues--has Johnson as the #2 2B, none of the others is listed.

Bill James' Historical Abstract--Monroe and Grant are the #5 and #6 2B in that order, neither is among the top 100 players (black or white), and White and Johnson are not listed.

The book Cool Papas, etc., etc., by McNeil contains a number of lists. A group of 28 former players was asked who were the best Negro League players not in Cooperstown. None of our fellows were among the 24 who got 10 votes or more (obviously the panel voted for multiple players each). But at 2B, Grant and Monroe tied for 4th place but with just 1 vote each. Home Run Johnson got 1 vote at SS, good for 8th place.

Then an "expert" panel voted on the same criteria (not in Cooperstown). They came up with 16 votes for Frank Grant, putting him in 16th place among all players. The others were not among the 21 players who got 13 or more votes. Grant's 16 votes also made him the top 2B, while Monroe finished 5th with 6 votes, and White 7th (tie) with 1 vote. Grant Johnson tied with Dobie Moore (he of the 7 year career) for 3rd at SS with 7 votes.

Then the two groups voted on an all-time team. None of our fellows made the team and in fact none of them received a vote from the players.

But among the experts Grant was 4th and Monroe tied for 5th at 2B with 3 and 2 votes respectively. Johnson and White were not mentioned.

That is that. So whose list you gonna trust?

Then you've got some numbers, though they're either pretty selective (not through any fault other than that complete numbers are not available) or they're "projections."

We know that Grant hit .337 across parts or all of 6 seasons against mid- to high minor league competition. KJOK projected 5 years of data--unfortunately Grant's first two and last three years, not his peak or prime--at .280 as a MLE with a .345 OBA and .424 SA. Not bad for a good defensive 2B outside of his peak or prime years. Do ya trust 'em?

We also know that Johnson hit .389 across parts of 4 seasons (1910-13, age 36-39) and that he hit comparably to teammate Pop Lloyd for at least a short while, also while he was in his later 30s and not hardly his prime. KJOK projects him to .308/.390/.501 for 15 years of MLE, which coming from 1901-15 would make him second to Wagner and better than anybody else around, if you trust the projection.

The only real numbers for Monroe was .348 in 1914, his final year, age 38, but KJOK somehow projects him over 14 years at .311/.380/.440, not quite in HR Johnson territory (and for one less year) but Monroe apparently was the best glove of them all. (Except of course Johnson played most of his career at SS, or did he?)

Finally, we know that White was very comparable to Frank Grant whenever they crossed paths, though it was Grant who was athletic enough to make the shift to SS to accommodate them both on the same team (though White also played some SS later on). But they hit comparably except one year when they were on different teams and White's team won the championship with Sol hitting .356 and Grant a mere .349. White also hit .381 against low-mid minor league competition, and then .452 at his prime against colored competition. But KJOK projects White, in just 5 seasons late in his career, at just .247/.329/.380.

So based on the numbers you'd kinda have to rank them on offense:

1. Johnson--career proj. 308/.390/.501 with a .389 at age 36-39

As always, when I go back and read my own post, I want to amend it. In this case it is to say that the problem for Johnson (in not making more lists) is NOT a timeline bias, because Monroe and Grant make more of them. It is Pop Lloyd and Willie Wells and Dick Lundy. There never was a Negro League 2B who could carry any equipment for any one of the three of them. So of course the various rankings by position are a lot harder for Johnson, than for Grant and Monroe and even White, to crack.

Finally if I had to choose one "expert," it would be McNeil's expert panel who voted for Grant, Monroe and Johnson, in that order (and not for White) across their two votes. Can some body who has the book tell us this: Was the voting by position??? Or was it general, and then it was organized by position after??? That would make a difference in interpreting their rating of Johnson versus the 2Bs!

Having completed my grand "reconsideration" of C-1B-2B-SS-3B, that is as far as I'm going to get this year. Now to integrate all the results into one list, and I will also be listing only those players who are now eligible. This is just the five "infield" positions. First, who was best on the various different measures?

1. I use adjWS to determine who were the best peak players at any given time. In a sense, this is probably the best test of greatness. I run 3 and 5 year rolling totals and any player within 10 WS of the lead is regarded as among the best/greatest players at that time. Hughie Jennings is best with 4 "mentions," a mention behind (e.g.) best on the 3 year total one time or the 5 year total one time. Both in the same year is two mentions. Four mentions is not a lot, BTW, which says that, hey, this is a backlog.

2. 3 year adjWS peak--John Clapp benefits greatly from the adj to 162 games, with 118 to 115 for Jennings.

3. 3 year adjWARP1 peak--Jennings 53.0-Fred Dunlap 45.6. This is with ^1/2 adj.

4. 3 year adjLWTS peak (TB7)--Jennings 20.7-Dunlap 16.6, and Dunlap's score is with a league adjustment (-65%) for the UA.

5. WS 5 years--Williamson 192-Jennings 179.

6. WARP 5 years--Jennings and Dunlap.

7. LWTS 5 years--Jennings and Dunlap.

8. Cumulative MVP voting (I supply my own hypotheticals where missing) over best 5 years--Jennings and Harry Davis.

9-10-11. WS prime--Beckley (17 years) and Wallace (14) for length; Williamson and Beckley for total WS; Williamson and Jennings for rate.

12-13-14. WARP prime--Beckley (17) and Wallace-Herman Long (14) for length; Wallace and Long for total WARPs; Jennings and Dunlap for rate.

15-16-17. LWTS prime--Beckley (17) and Wallace (16) for length; Wallace and Dunlap for total value; Jennings and Dunlap for rate.

18. Cumulative "all-star" points--I provide my own hypotheticals where missing--H. Davis and Childs-Dunlap.

XX. Pennants Added--only used to compare players both whom I have the numbers--Dunlap and Williamson.

19. Career adjWS (adj to 162 game seasons and with AA and UA discounts)--Wallace 366, Beckley 363, Leach 339, Williamson 318, Long 301.

20. Career adjWARP1 (adj to ^1/2 with league discounts)--Wallace 152, Long 134, Dunlap 126, Beckley 121, Childs 118.

21. Career adjLWTS (ditto WARP)--Wallace 37.9 Dunlap 34.5 Childs 32.6 Jennings 28.5 Beckley 24.9

22. Reputation Monitor--this is a number I developed a couple years ago, and now use to build consideration sets. It includes adjWS peak and career, LWTS, my own MVP (peak) and all-star (prime) rating systems, HoFM, HoFS, Black and Gray Ink, defensive WS (or offensive WS for pitchers), OPS+ or ERA+, and peak BBWAA HoF votes earned. It is a measure of reputation, not of value. I suppose I could throw it out and just use HoFM but it took a long time to develop it, so I use it. It works just like HoFM, however, in that every player with 200 points is in the HoF, most players from 150-200 are in, some from 100-150 are in, and anybody below 100 just doesn't belong, though that includes a dozen or so HoFers. Anyway, the best Reputation Monitor scores among this year's eligible C-1B-2B-SS-3B are Jennings 183, Wallace 180, Bresnahan (!) 163, Dave Orr (big OPS+ assist) and Tommy Leach (155).

So those are the various measures. I then ranked 15 players from 1-15 on every measure and calc. Then of course I have to mix in the non-numerical candidates like Pearce and all the Negro Leaguers. Here is the order that these particular players will be on my 1924 ballot, though of course the OF and P will be mixed in but without all the rigamarole.

1. Dickey Pearce--right where he started, the only one who was clearly the very best middle infielder of his time

Sorry, I'm a longtime lurker and a registered voter. but my knowledge of 19th century baseball is pretty poor, so I sat out those elections until now. I feel I've learned enough on the players to put togther an educated ballot. I would've jumped in on the last ballot, but I took the week to do more research knowing that Honus Wagner would get the easy induction.

That said, I have one question which may have already been covered, but I could not find the answer in the archives or in the HoM Consititution. Does Bresnahan get extra credit for his equipment innovations? Because purely on his statistical line, he's low on my ballot, but once you include the invention of shin guards and the face mask, he rockets up my list. That's a huge impact on the game he had. Thanks.

Welcome, Stephen! Looking forward to seeing your ballot.

It's not up to me, but I agree with TomH. Otherwise, I'd have Candy Cummings on my list. And then there's Tommy McCarthy. And...

Thanks. Ballot still under construction, but I spent the good part of Easter annoying my wife with Win Share comparisons of turn-of-the-century shortstops. The top of the ballot's pretty clear for me, but once you get into about five and below, it gets real muddled, real quick. I thought it would be easier.

The top of the ballot's pretty clear for me, but once you get into about five and below, it gets real muddled, real quick. I thought it would be easier.

We all did, Stephen. Welcome to the club! :-D

You guys have to be kidding. This year is easy. I just put all the guys in my PHoM but still eligible for the HoM at the top of the ballot and go from there.

STILL too swamped to do this myself, but I keep seeing Crawford-Beckley comparisons.

For those who have Crawford ranked 1 or 2 and Beckley below 10 or off the ballot, what makes Crawford so much better?

Howie:

Leaving defense and position out of it ...

Both Beckley and Crawford are selling career length and stability. Beckley played 2386 games, Crawford 2517 games. Crawford played in a time of 154-game schedules; in Beckley's time, the schedules were often closer to 132 games. Hence Crawford's extra games mostly come from extra opportunity. But there's a compensating element: Because Beckley played in times of big offense, he got to bat more times per game. Indeed, the plate appearances are Beckley 10470, Crawford 10594 - nearly a wash.

Compared to the averages of his time, Beckley's BA/OBP/SLG was +.033/+.020/+.069. Crawford looked at the same way was +.046/+.036/+.109. Such a consistent across-the-board advantage may not seem large, but it becomes large when you multiply it by 18 or so seasons. Crawford has what appears to be a small edge over Beckley in SB - but there were more SB in Beckley's time than in Crawford's.

If you normalize them to the same offensive context, I've got Crawford creating about twice as many runs above league average than Beckley. Crawford had a number of individual offensive seasons that were much bigger than anything Beckley ever had.

I wouldn't hesitate to take Crawford ahead of Clarke, Flick, Kelley, or Keeler.

For us peak voters, there is no comparison. Keep in mind the adjustments for season length.

adjWS peak 3 consecutive seasons--Crawford 100 Beckley 64

Somebody questioned my sanity since I'm apparently the only voter who didn't have Eddie Plank on his ballot. Well, I've never voted for Mickey Welch either and not too many people have questioned that.

Plank 326 wins (194 losses, .627)

preliminary ballot, feel free to criticize, particularly the ranking of the Negro Leaguers, which I'm pretty sure are totally off base.

1 SAM CRAWFORD. A pretty easy one to start off the ballot.

2 THREE FINGER BROWN. A great peak, and he's 17th all-time in adjusted ERA. I'm not sure I understand the argument against him. If his greatness was a function of the great defense behind him, why aren't Tinkers and Evers getting more HoM support? Because somebody deserves enshrinement for those Cubs teams.

3 EDDIE PLANK. Brown had a better peak, but Plank had the longer career. If I was picking a team, and I had to choose between them knowing what they would produce, I'd take Brown. They pitched at the same time, and Brown outperformed him most years.

4 HOME RUN JOHNSON. I'm real sketchy on the early Negro League players, but I know Johnson was real good.

5 JOE MCGINNITY. His peak was close to Brown's, but he didn't have the same career record. I hate to say I'm a strict peak or career voter, though I do tend towards peak, but sometimes you have to say, the guy only pitched 10 years.

6 BOB CARUTHERS. Jesus, was this guy's peak off the charts. He was also a pretty good hitter, which doesn't hurt the cause. But outside of the peak, there's nothing. It's like five years of complete brilliance, and then three years of mediocrity. I'm not sold on his competition either. The toughest player to rank, I had him everywhere from 2nd to about 15th.

7 HUGH DUFFY. Another peak player. His 1894 campaign was great, and he accumulated 144 win shares in his five-year peak. Good power and speed combo.

8 FRANK GRANT. I'm not sold on who is better, Grant or Monroe. With a sketchy statistical record to go by, I end up lumping them together, so...

9 BILL MONROE. Somebody, help me out here.

10 HUGHIE JENNINGS. Any guy who holds the all-time record for HBP's is my kind of player. Another player with an awesome peak, but he was pretty much done by age 30.

11 JIMMY SHECKARD. Rates well in career Win Shares, but how much of this is simply playing 17 years? Strikes me as a guy who was always pretty good. He merits a mention for consistency, but I'd have trouble putting him at the top of my ballot.

12 RUBE WADDELL. Led the league in K/9IP eight different times. 10th all-time in ERA (22nd in ERA+). Won the pitching Triple Crown in 1905. So how does he have a career record of 193-143? I know won-loss record is a team-dependent stat, but come on. He wasn't even the best pitcher on his own team. Once Plank's in, I'll reconsider him.

13 GEORGE VAN HALTREN. His peak's not bad, but it's nothing that'll put him in the Hall. what I like about his career is that he was so good for so long. Of his 17 seasons, he had an OPS+ above 100 15 times, bookended by two weak perforances. And it seemed almost every season he was near his career average of 121.

14 JIMMY RYAN. A better prime than Van Haltren, but not quite as consistent. Hit 118 home runs, back when 118 home runs in a career meant something. I'm probably ranking these two backwards now that I look at it again.

15 ROGER BRESNAHAN. Just not enough time as a catcher to rate solely as a catcher, and his numbers don't match up with the outfielders on the list. Still a fine player.

stephen, post #88 above is more than everything that I know about Grant, Monroe and Johnson.

stephen,

With regards to your prelim ballot its kinda hard to criticize. No Jack Chesbro so it can't be all bad : ) Looks like you've read over the group's work quite a bit as well as completing your own analysis.

Only question I have is where are the holdovers candidates from the earliest years of baseball? Specifically I thinking of ones like Sam Thompson, Pete Browning, Dickey Pearce, Lip Pike, Mickey Welch. Are they just below your top 15 or are more "modern" players just outside the ballot?

TomH, that quarter of a run per game works out to 4 percent on ERA+. Yes, '80s ball was different than '00s ball--Welch pitched in higher run environments.

I agree that the comps show Plank to be superior, just not by a whole lot. I just wanted to supply some context for those who say that we can NOT fail to elect a guy because he won 300 games. Well, if that's true of Plank, what about Welch? Plank is closer to Welch than to Cy and Matty and Walsh.

Pike, Pearce, and Browning are in my top 20 along with Clark Griffith and Beckley. I admit a very skethcy knowledge of the earliest days of baseball, and they could rocket up my ballot as I re-evaluate. Outside of the Negro Leaguers, the very first players are the ones I need to do the most research on.

Which means I'm open to persuasion on all of them. I consider myself the blank slate on these players, they will get moved up my ballot by virtue of the quality of the argument on their behalf. Pike was the last guy I cut, though. Thanks for the compliment, I've been going through a lot of archives and probably didn't process everything.

stephen, a good way to start is to look at the 1923 results and read up on anybody who got votes. Along with Pearce and Pike, I would commend for your consideration Charley Jones (very comp to Browning), Ed Williamson (comps Jimmy Collins) and Sam Thompson (a shorter version of Sam Crawford).

Al Peterson's ballot reminded me of this (which maybe Joe never got in his inbox). Better late than never, the list of players passing away in 1923:

HoMers

Cupid Childs--I can't see why he's not on everyone's ballot

I definitely agree. Check out how many times he was the best (or near best) at his position in the majors. Which player eligible compares to him? How many inducted compare to him?

How about me for instance?

I have you as the best player in the majors at your position once. I have Childs as the best second baseman in the majors seven times (and very close for another one). Sorry, but the comparison falls short.

John,

Which 7 years do you have Childs as the best 2B in the bigs?

I've got him as the best in '92,'96 (both by a bunch) and '97. Runner-up in '93 and '94 (both to McPhee).

Just wondering what you're using to determine "best", not suggesting that I'm right and you're not.

Which 7 years do you have Childs as the best 2B in the bigs?

I have him as the best major league second baseman for 1890, (almost in 1891), 1892, 1893, 1894, 1895, 1896 and 1897.

Just wondering what you're using to determine "best", not suggesting that I'm right and you're not.

That's a perfectly legitimate question, Max. My base is seasonal Win Shares and seasonal WS per 162 games.

That would do it. I use the basics of Warp data, then modify from there. I imagine the difference between Warp and WS views of 2B defense would be enough to drive Childs up or down from best to 3rd or 4th in any given year. This is assuming as given that Childs is an offense-first man, while McPhee is the glove man...

Childs gets beat up by WARP3 by the linear aspect of the method. I've mentioned this "years" ago before you joined this project, but it's illogical, IMO, that someone who plays 162 games in a season could have a negative WARP3, while a player who batted once during the season could have a positive WARP3. This is a major philosophical problem for me (though I did agree with it for a short time after The Hidden Game of Baseball came out in '84), though I do use the WARP numbers for the NA (without using negative numbers - they come out as just zeroes).

John,

I probably could have been much clearer. When I'm determining "best" by position by year (I call them All-Stars and Reserves), I add up my adjWARP2 FRAA and BRAR earned at that position. Note: I adjust basic WARP2 for games played by a particular team up to 162 (at a 2/3 factor). I also adjust for position based on Joe D and Tango's comments on FRAR.

This assumes 2 things;

Max, do you have a formula handy for calculating BRAA from the other elements on the BP player sheets?

Jimd,

Because I'm using the fielding numbers, I only have EQR and Outs. I use:

EqA = (EqR/Outs/5)^0.4

Jimd,

Because I'm using the fielding numbers, I only have EQR and Outs. I use:

EqA = (EqR/Outs/5)^0.4

Joe's #125-126 belong on the "Pitchers" thread and my response is over there.

I have quoted TomH and replied in the Catchers thread.