Interesting discussion about Pete Hill on the 1927 ballot thread. Pretty unanimous agreement that he is comparable to Magee and Sheckard. I offered the skeptics POV (and I see that TomH agreed with me), and yet I agree that he is comp to Magee and Sheckard. So it comes down to what you think of Magee and Sheckard, Magee already having been elected with a lot of bonus positions but also a lot who didn't have him in the top 5. So I expect a big spread on Pete.

I like Pete and he may make my PHoM someday. But overall I think we are still (or again) trending toward too many OF and not enough middle IF. So generally I like guys like Childs, and in the Negro Leagues I would urge Pete Hill fans at least to take another good look at Bill Monroe, who apparently hit almost like Pete Hill while playing a stellar 2B. He also was basically interchangeable with Frank Grant.

But overall I think we are still (or again) trending toward too many OF and not enough middle IF.

I don't know...11 of our 48 HoM members are middle infielders. That's probably a dead-on number, considering that's two positions. I do agree with you on Monroe, though...I think he was better than Grant, given the level of competition.

We have 15 of the 48 as outfielders, which is also about right (or even a bit low) since that covers three positions. But the trend you mention is certainly there.

IMO, we should be a lot more concerned about catchers (primarily) and third basemen (to a lesser extent), each of whom only get 3 of the 48 spots. However, I believe we haven't seen much talent at 3b so far, so our options are few.

Sorry, that wasn't about Pete Hill, so I sort of "polluted" the thread.

I've got Hill about where the others have analyzed him...as comparable to Magee, but with a greater defensive contribution. Although in some ways he may look like Sheckard, remember that Hill played a more difficult defensive position, so the high quality defense he contributed means more than the high quality defense provided by Sheckard in left.

Anyway, I voted Magee #1 last time. I've got Hill at #2, since Joe Jackson will take the #1 spot this time. Hill will be the second Negro Leaguer in my PHOM.

I don't think my vote is attributable to Hill being a new candidate...all players get a rating in my system on a scale of 1-100 (based on the % of the HoM standards I've set), and he is where he is. He won't move up or down unless new data is produced or the existing data is analyzed in a different way.

A note on Hill's parks:

We've discussed the fact that Schorling Park, home of the Chicago American Giants from 1911 on, was an extreme pitcher's park.

I've just come across a note in Riley's entry on Turkey Stearnes to the effect that Mack Park, home of the Detroit Stars, was a hitter's park.

So Hill's renaissance as a hitter in 1919 when he moved from the Giants to the Stars was doubly influenced by park effects.

Looking at the Macmillan again (too lazy to go year by year, though):
H/AB - AVG.
1904-13: 76/178 .427
1914-17: 100/428 .234
1918-25: 140/419 .334
Total: 316/1025 .308

Park Effects, 1928 Negro National League:

St. Louis 124
Kansas City 99
Detroit 93
Chicago 57
Birmingham 102
Memphis 101
Cleveland 144

To properly apply, these would need to be prorated by the # of home and away games. For example, for St. Louis, 59 Home games and 39 away games would give an "overal" factor of 115 to be applied to St. Louis player stats.

Chicago 57

!!!!!

That's almost inconceivable, but highly important to keep in mind.

These numbers go against the anecdotal assessment of Detroit's Mack Park, though it looks like an excellent hitter's park when compared to Chicago?

KJOK, what's the source of those numbers?

If Schorling park played similarly in the 1910s, Pete Hill's down years begin to look a lot more respectable. I had been factoring in half of a 90 park-factor in converting Hill's numbers . . .

This could hurt Rube Foster in my rankings a little bit. His 1911-1914 seasons aren't particularly important, to his value, but I'm less impressed by his numbers if he was doing a good part of his pitching under these conditions.

Park Effects, 1928 Negro National League:

Your comment answered my question. These are STATS-like "raw" park-factors (runs-at-home/runs-away) and need to be processed (averaged over each teams actual schedule) to create useful factors that can be applied to player (or team) summary stats.

This could hurt Rube Foster in my rankings

It makes any ERA numbers less impressive but it doesn't affect the majesty of those W-L records.

Chicago 57

In the 20th century, the most extreme "raw" STATS ML park factors I could find (not an exhaustive search mind you) were 68 in Braves Field 1934, 69 in Milwaukee 1958, and 70 in Oakland 1973. OTOH, Cleveland looks like a typical season at Coors.

This could hurt Rube Foster in my rankings

It makes any ERA numbers less impressive but it doesn't affect the majesty of those W-L records.

Sure, especially since they weren't achieved during his seasons in Schorling Park. Holway provides RA data for some of Foster's seasons, 1911-1914, which I had used to assess where Foster stood in relation to major-league average in pitching quality. I took park factor into account, but not quite to that extent . . .

Park Effects, 1928 Negro National League

Any way to "smooth" those numbers over a few years? I think most of the professional sabermetricians agree that park effects for a single season are unreliable.

I think your numbers give us a good sense of hitters park or pitchers park, but I would caution everyone against applying those specific factors.

I think most of the professional sabermetricians agree that park effects for a single season are unreliable.

Good point Jeff, and it applies particularly to these numbers. From what little I know about the Negro Leagues, my guess is that these numbers are from a smaller sample of games. Also, they didn't always come close to balancing the schedules between parks and amongst the teams, which can impact the calculations.

KJOK, what's the source of those numbers?
They are based on the scores from 363 Negro League games in 1928.

Chicago 57

In the 20th century, the most extreme "raw" STATS ML park factors I could find (not an exhaustive search mind you) were 68 in Braves Field 1934, 69 in Milwaukee 1958, and 70 in Oakland 1973. OTOH, Cleveland looks like a typical season at Coors. I think you may be comparing apples and oranges here. If the STATS numbers are calculated like the Total Baseball Park Factors, then they already "factor in" road games, while the 57 means that scoring AT the Chicago Park is 57% of the average league games.

Here are the raw numbers:
Chicago Home Record 48-29-1
Chicago Road Record 15-18
Giants Runs scored at home - 309
Giants Runs allowed at home - 213
Giants Runs scored on road - 202
Giants Runs allowed on road - 186

Any way to "smooth" those numbers over a few years? I think most of the professional sabermetricians agree that park effects for a single season are unreliable.

Without more than 1 season's worth of data, the proper method would probably be to regress the results to the mean about 75%.

OK, here are the same numbers adjusted for the # of Home and Away games, adjusted for STATS/TOTAL BASEBALL convention, AND regressed 75%:

St. Louis Stars104
Kansas City Monarchs100
Detroit Stars99
Chicago American Giants92
Birmingham Black Barons100
Memphis Red Sox100
Cleveland Tigers105
Cuban Stars (West) 100 (only played away games)

I think most of the professional sabermetricians agree that park effects for a single season are unreliable.
I should point out that if you're trying to determine a player's VALUE (and I would argue the HOM project is about value) then you should ONLY use 1-year park factors and NOT average across seasons.

I should point out that if you're trying to determine a player's VALUE (and I would argue the HOM project is about value) then you should ONLY use 1-year park factors and NOT average across seasons.

I understand your point, but the reason NOT to use single seasons is that it seems to be acknowledged that single season park factors are unreliable. They tend to bounce all over the place from year to year, even when the park hasn't changed. It can't all be weather!

Bill James uses a 5-year standard. He weights the focus year at 1/2 and takes the two years before and after at 1/8 each. That way, you still get a unique park factor for each season, but you eliminate some of the randomness of the park factor.

I've said before that, even though I posted the skeptics opinion of Pete Hill, that I consider myself a FONL and a qualified FOPH. I guess we're not supposed to look ahead but I did anyway, just to try to place Hill among his peers in the NL and get some sense of his place in history that way. As a pioneer, it would be easy to overrate (or underrate) him. There's still no standard to compare him to.

But anyway, one thing I have noticed is that the OF pool among NLers is relatively weak (looking out through 1946, which is as much of the list as has been posted). Of course we've got Oscar Charleston coming up, though he's almost 20 years away, but there's your gold standard.

Other than that you've got Torriente, Poles, Lyons, Chino Smith and then way off in the mid-'40s Stearnes and Suttles (expanding the category a bit to include cornermen/hitters). There are others of course--Jimmy Lyons, Heavy Johnson, Fats Jenkins and more--Ben Taylor if you include 1B. But that's not a big pool compared to the pitchers and IFers, I don't think.

And so it struck me that in the early NLs, you had the situation you had in early baseball generally and in below professional ball today--all the best athletes are either pitchers or shortstops (or middle infielders, I should perhaps say). I mean so far you've got Johnson and Grant with Monroe and Foster and maybe Sol White as pretty clearly the best yet.

And then Pete Hill, the first great OF. But then a shallow pool for another 20 years.

So the idea of electing NLers by virtue of a quota method, which might work (and/or might simply be unavoidable) at a macro level, is going to be hard to work with at a micro level. I'm not sure there aren't a good half dozen or more C and IF and another 3-4 pitchers who are as good or better than the best OFers, until Charleston, that is.

In sum, Pete Hill looks to me to occupy a really ambiguous position. Still something of a pioneer, probably a second-tier player compared to Charleston, Stearnes and Suttles, and Lloyd and Lundy and Santop, and Joe Williams and a couple other pitchers. If all were in one big pool I'm not sure he wouldn't sink out of sight. But he's not in the pool with them, he's one of the pioneers. So anyway, this sounds like I'm anti-Pete Hill. I'm not. I just want to understand whether (what with our quota system) the slot we give to him shouldn't really go to Ben Taylor or Dick Lundy or Jose Mendez or John Beckwith some day. Or more to the point, to Jimmy Sheckard or Sam Thompson today.

Hill is no easier than Dickey Pearce to peg, in my book. One of the best (and the best of his kind) at the time. But what exactly does that mean?

Hill is no easier than Dickey Pearce to peg, in my book. One of the best (and the best of his kind) at the time. But what exactly does that mean?

I hear what you are saying, but I don't think Hill and Pearce are in the same situation. I think the league in which Hill played was far more developed than the leagues in which Pearce played. We lack some data for both, but the Negro Leagues were well-established by Hill's time, and the players were drawn from a broader regional base.

We've already elected two negro leaguers who were more in the Pearce category than Hill, in my opinion.

I don't know that we ought to elect Hill because he was one of the first great Negro League outfielders (particularly if your assertion is correct that the best players were infielders or pitchers). It appears to me, however, that he was one of the very best stars of the Negro Leagues during his era...as big as Johnson, Grant and Monroe were. That's a reason to elect him.

Jeff, I'm going to be conservative with Hill at first, but I'm not agin' him, just asking questions. Hill seems to be one of the brightest stars of the NL at the time, and he was, but his luster on the ballot at this time is sorta like Bobby Wallace or Addie Joss in 1919. Walter Johnson and Hans Wagner weren't on the ballot yet.

Hill played with Lloyd and Torriente and Poles and Petway and Lyons and Mendez and Santop. As I said in my first Pete Hill post, his leadership abilities are not disputed, as are (not) Harry Wright's and Rube Foster's. But how did Pete stack up against some of these other guys between the lines? I don't think he was Ty Cobb or Speaker or even Crawford. I'm sure he was not Earl Combs. But maybe Sheckard or Magee or Max Carey?

I don't think he was Ty Cobb or Speaker or even Crawford. I'm sure he was not Earl Combs. But maybe Sheckard or Magee or Max Carey?

I agree with you on each of those guys, except I haven't evaluated Combs or Carey yet. At the Magee/Sheckard level, he's pretty top choice among our current eligibles.

Snap comparisons to Pete Hill

John Henry Lloyd -- no comparison; Lloyd the best Negro-League player who played primarily before 1920. Hit as well or better than Hill, great defensive shortstop, great player past 40. Named an all-star by Holway 13 times. Hill was a 5-time all-star, but Holway starts his awards in 1910, so Hill would probably have garnered at least 4 more if he had the chance.

Torriente -- possibly better than Hill. _Great_ hitter, excellent defensive player. Career fairly short for Negro Leagues: 16 years. 9 time all-star.

Poles -- not Hill's equal. Definitely a star in the first half of his career, but I haven't seen evidence that he was a great player after 1915 or an impact player after 1920. Career shorter than Torriente's at 15 years. Yes, he was better than Hill 1911-1914, but that was his peak and the end of Hill's prime, so it's not an even comparison. 3 time all-star, should be 4 but Holway didn't name an East team in 1912, when Poles definitely would have been on it.

Petway -- probably not Hill's equal. Great defensive catcher, but only a good hitter for about five years, missed quite a lot of time due to injuries in the teens. 7 time all-star.

Lyons -- not Hill's equal. A few great seasons, but no evidence of a prime like Hill's. I'd take the underappreciated George Shively over Lyons. 2 time all-star (Shively was a 7-time all-star).

Mendez -- no clue. Pitchers are especially hard; players who split their careers between pitching and position play are harder than that. I don't have a sense yet of his career. 4 time all-star.

Santop -- probably better than Hill. A tremendous hitter, and unlike Petway his hitting did not drop off after just a few years. No defensive reputation to match Petway's, but no stories of him being a disaster behind the plate. 7-time all-star.

Ben Taylor -- probably not as good as Hill; doesn't have the reputation as an impact player that Hill does. I haven't studied his statistics carefully yet, but nothing I've seen so far suggests he was a better hitter than Hill, and as a first baseman his defensive value was less. Does have a couple of years as a pitching star early in his career to further complicate the picture. 8 time all-star.

My sense of these players now.

Lloyd
Santop
Torriente
Hill
Taylor
Petway
Poles
Lyons

Mendez???

My in/out line is right around Taylor/Petway. I need to study them both more before I conclude what side of the line they are on.

I understand your point, but the reason NOT to use single seasons is that it seems to be acknowledged that single season park factors are unreliable. They tend to bounce all over the place from year to year, even when the park hasn't changed. It can't all be weather!

Sorry, but your answer indicates you don't quite understand the point I was trying to make. For VALUE, there is no such thing as an "unreliable" park factor. The run environment for a given year is what it is. If in Year A games in Park A average 6 runs per game, then it's irrelevant that the following year they averaged 11 runs per game in Park A. A HR in year A will be more valuable than a HR in year B.

Sorry, but your answer indicates you don't quite understand the point I was trying to make.

Interesting tone. And yes I do. But even if I don't, Bill James has a pretty good sense of what I'm talking about and there's an entire essay on it on page 86 of Win Shares. To quote, in part:

"How do you determine the park factor for a team? This is one of those issues about which, if you think you know how to do it, you don't understand the problem."

For VALUE, there is no such thing as an "unreliable" park factor.

Yes there is. What you are assuming is part of the park may be the result of other things, and the data sample is too small to draw any conclusions about what is park and what is other stuff. If you are going to adjust a player's stats by a park factor, you have to make sure the park factor is only incorporating park characteristics...and you can't do that on a single season basis.

Park factors are there to let us translate stats to park-neutral by comparing one park to another. If you are applying a park factor that has non-park factors involved (e.g., a few horrible pitching performances in a park for a particular year), then the adjustment doesn't produce park-neutral stats. Those "other factors" may be random occurrences that would not have affected a player from a different team playing in the subject park rather than his own.

I agree with both KJOK and Jeff on the park factor issue.

I posted this to SABR-L last month:

The statement I responded to was:

"I don't think the park factors I used to compare the offense of the '48 Indians and Red Sox were questionable. They were based on 5-year averages."

My response:

"This misses the point. My point was that ALL park factors are somewhat questionable, because 1/2 of their basis is from one team - a team that has generally) been designed to take advantage of the park in question. ALL park factors are ESTIMATES, not absolutes, and as such are questionable by definition.

Which doesn't even get into the question of whether 5-year or 1-year averages are appropriate, we'll save that one for another day :-)"

Sorry I'm not explaining this well, but I'll still contend a player's VALUE has nothing to do with either:

a. sample size
b. only park characteristics
c. park neutral stats

If a team plays only ONE game for the season, and the score is 6-5, then ALL value is in the context of the 11 runs scored. Doesn't matter if the wind was blowing out, or below average pitchers were pitching, etc. etc. Also doesn't matter that the other two teams in the league played a 2-1 game in their one game season.

James explained this in the original historical abstract, and even used it in the calculation of the player's offensive winning %.

Using 3 year of 5 year park factors and calculating park neutral stats is appropriate for questions of player ABILITY, not VALUE.

Let me quickly raise a different kind of question about park factors.

Looking at the 1928 pfs calculated by KJOK, the extreme run-suppression effects of Chicago's home park is diminished relative to the rest of the league quite a bit by a) the fact that many more games were played there than at any other park and b) regression to the mean.

If that's corrrect, I would understand this to indicate that when normalizing figures for Negro-League players relative to one another, we conclude that Schorling park had a large effect, but one well within the range of expected park effects.

I wonder, however, how the run-environment of Schorling Park affects the statistics of Negro-League players in comparison to their contemporaries in the white majors. If one of the reasons that Schorling Park's effect is diminished in the Negro-League context is that everybody -- home team and away teams, are playing a large percentage of games there, would that not mean that batting statistics in the Negro National League as a whole are being reduced by these park effects, making the NNL into a strong "pitcher's league"? If those inferences are correct, is there any way to get a "ballpark" estimate of how one might account for that effect in figuring MLEs?

Chris, I don't think that's a real issue.

Between 1913 and 1954, there is always at least one park shared in common between the AL and the NL. If you look closely, during the 10's, 20's, and 30's, it is obvious that the NL is the "hitting league" and the AL is the "pitching league". The average park factor of the non-St.Louis parks (non-Polo Grounds parks during the '10s) is always higher in the NL than it is in the AL, particularly when using factors calculated from more than one season.

But I don't think this really matters. Once you convert the stats to league-centric stats (ERA+, OPS+, BP's more complex normalizations), get away from the raw numbers, then which league favors hitting stats, which favors pitching stats, that is washed away.

In its place, however, is the question of relative league quality. Is 100 in the AL "better" than 100 in the NL or vice versa? And similarly how does 100 in the Negro Leagues compare to 100 in each of the "white" major leagues?

jimd,

Problem is, I don't have league-centric stats for any Negro-league players, and if such stats are available anywhere, I'm not aware of it. If league-centric stats are available, great, that gets us right to the league-quality question.

Right now,though, I have to apply league-quality conversions to raw stats, so I need to know how the pitchers' league/hitters' league issue would affect the conversion of raw stats.

Nobody's added up the stats for the games that are available? They may not be complete, but a nice random sample should allow some decent estimates to be made. I assumed with all of these partial player numbers floating around that there would at least be some sketchy league numbers.

I have complete league numbers for 1928 only. I hope to have more league numbers in the next few months.

I don't have my "stuff" with me, but I'm 99% sure that Schorling Park is the same park as American Giants Field which is the same park as South Side Park III, which is where the White Sox played from 1901 thru 6/27 of 1910, so we certainly can get a feel for the overall "hitter friendliness" of the Negro League parks thru a direct comparison with AL numbers for the White Sox 1901-1909...

From 01-09, South Side Park decreased run scoring by 13% and decreased HR's by 75%!

Hill looks to be closer to Sam Crawford than Magee or Sheckard, IMO. I'm still figuring out what his slot in the top ten will be, though I know he'll be ahead of Jackson.

And to follow up on Chris Cobb's astute question about a large % of total league games being played in Chicago, here's the breakdown for 1928:

TEAM %of Lg Games Played in Home Park
Chicago 21%
St. Louis 16%
Detroit 15%
Memphis 14%
Cleveland 11%
Kansas City 11%
Birmingham 10%
Neutral Sites 0%

So yes, Chicago would probably tend to pull the leage numbers down, but really not any more than St. Louis and Cleveland together would pull them back up...

we certainly can get a feel for the overall "hitter friendliness" of the Negro League parks thru a direct comparison with AL numbers for the White Sox 1901-1909...

If it were only that easy.

1) Parks change over time. Fenway was a pitcher's park in the 1910's. The configuration didn't change that radically (the bullpens were added in the late 30's, shrinking right field; and Duffy's Cliff, a grassy knoll in front of the wall like Houston's but from foul line to center field, was flattened) but it did change. More importantly, the hitting background changed. In old Fenway, the Monster was covered with ads; it was very difficult to pick up the ball. When Yawkey redid it in the early 30's, the ads were painted over and the background improved dramatically.

2) Park factors are relative. Wrigley has not changed much over 50 years, but it's gone from a neutral park in the late 40's to a hitter's paradise (when the big pitcher's parks like Dodger and Candlestick replaced the bandboxes like Ebbetts and Polo Grounds) back to a neutral park when bandboxes have come back into fashion.

3) The style of play can influence park factors. The Polo Grounds was a pitcher's park (in the NL at least) until players started pulling the ball down those short foul lines for easy home runs. The HR factors for the "Hitless Wonders" are probably irrelevant.

jimd:

Good points all, however:

1. As far as I know, South Side Park did not have a major remodeling (and the Fenway changes were MAJOR.)

2. This actually supports what I said. By assuming Wrigley is relatively unchanged over time, the fact that it's Park factors have changed quite a bit tells us a lot about the OTHER parks in the league now vs. earlier.
Accordingly, the Park Factors of South Side Park in the AL vs. South Side Park in the Negro League might tell us something about the parks in the AL vs. Parks in the Negro Leagues.

3. I agree 100% that style of play will have an impact, but not that HR factors would then necessarily be irrelavent. I'd say it tells us a little bit about why the "hitless wonders" were called the hitless wonders - their park surpressed offense, espeically HRs! On the road, they actually hit with some power!

Now, how that would impact offense in the more HR happy 1920's needs more data to be determined.

I mentioned elsewhere that I have an opening in my 1933 PHoM selections. Walter Johnson will go in, of course, but I am way behind on LFers (or you guys are way ahead). And with Zack Wheat being a good strong contender, I just thought I would ask the question. If you could pick one of the following for your PHoM (or for the HoM for that matter), and only one, which would it be and why?

Harry Stovey
Joe Kelley
Jimmy Sheckard
Sherry Magee
Pete Hill
Zack Wheat

Or maybe somebody wants to back G. Burns or Bobby Veach. Thanks for your help with this dilemma.

Marc:

I would go with Hill over all of them, though Stovey would be next.

I'm not that enamored with Wheat and don't see him in any way as a no-brainer selection. I have no idea where he'll be on my ballot.

1921 Pete Hill
NNL Detroit Stars

G-42 (team 54)
AB-137
H-41
D-6
T-5
HR-3
R-20
W-18
HP-5
SH-4
SB-3
AVE-.299 (NeL .263)
OBA-.400 (NeL .324)
SLG-.482 (NeL .361)

Holway has Hill on his western all-star team for '21, but given what I know about the season, I can't agree with that. Charleston (.440/.525/.741) and Blackwell (.430/.504/.715) are obvious, but Torriente was clearly better than Hill.

If you divide up outfielders strictly by position, Charleston's the all-star cf, Blackwell's the rf, and either KC's Hurley McNair (.336/.381/.520) or Chicago's Jimmie Lyons would be the lf.

About three weeks ago, I calculated MLEs for John Henry Lloyd. I thought it might be useful to look at a couple of other players of that era who've never received the MLE treatment, including Pete Hill.

I've followed the same approach that I used for Lloyd: (a) The MLEs are based on sources for which contextual data such as league averages are available—the HoF data that were released at the time of the special election, the Cuban League data, and the leagues and series that Gary Ashwill has tabulated. (b) I did not attempt year-by-year MLEs; instead, I split the career into periods for which the number of plate appearances is large enough to produce reliable rate statistics. (c) I convert the batting line based on walk rate (actually walks plus hit-by-pitch per plate appearance), batting average, and isolated power. (d) I adjust these rates for differences in context and league quality of play, using proportional adjustments for context and quality-of-play factors of 0.9 for batting average and 0.81 for walk rate and isolated power.

The first question is determining the starting and ending dates of Hill’s major league equivalent career. According to Riley, he started playing in 1899 and ended his career in 1925. Riley lists his first team (1899-1900) as the Pittsburgh Keystones; however, this team isn't listed in Clark and Lester or in Holway, so I'll assume it was probably a semipro team. Riley says he then played for the Cuban X Giants in 1901-02, but Clark and Lester and Holway don't show him on the X Giants roster for those seasons. His first appearance in Clark and Lester and in Holway is in 1903; he also appears with the X Giants on their tour of Cuba that fall. So, I'll start his career in 1903. At the other end of his career, in 1921 he was playing for the Detroit Stars and hit .329/.424/.507 (including inter-league games), making him one of the best hitters in black baseball. In 1922, however, he played for a non-league team and when he returned to the league as a player/manager in 1923-25, he was a below average hitter and his playing time was reduced. Therefore, I'll assume that his major league equivalent career would have lasted 19 seasons, from 1903 to 1921.

Note that Hill’s date of birth is not a settled question. The Hall of Fame reports his birthdate as October 12, 1880, which would make his 1903-21 playing ages 22 to 40, but Gary reports that recent research by Patrick Rock suggests a birthdate of October 12, 1882, which would imply that his ages were 20 to 38 during the summers of 1903-21.

I've split Hill's career into two periods: 1903-13 (ages 22 to 32 if he was born in October 1880, or 20 to 30 if he was born in 1882), and 1914-21 (ages 33 to 40 or 31 to 38).

1903-13. For this period, I rely on Cuban data from Gary Ashwill and from Figueredo. From 1906-12, Hill played five winter seasons in Cuba. Using data from Gary for 1906-07 and from Figueredo for the other four seasons, Hill's Cuban League batting average and slugging percentage were .300/—/.375 in 613 AB in a league context (excluding pitchers) of .235/—/.287. For three seasons, Gary provides walk (and hit by pitch) data for a subset of games; for these seasons, Hill's walk rate was .154 (403 PA) in a league context of .106. In addition, Hill played with several Negro league teams that toured Cuba and played against Cuban League opponents (the 1903, '05, and '06 Cuban X Giants, 1907 Philadelphia Giants, 1908 Brooklyn Royal Giants, and 1910 Leland Giants). In these series, Gary reports that Hill hit .284/.369/.358 in 232 AB in a context of .214/.285/.239. Finally, Hill played in Cuba in four series against visiting major league teams (the 1908 Cincinnati Reds, 1909 and '10 Detroit Tigers, and 1910 Philadelphia Athletics). In 23 games, 75 AB, Hill hit .347/.456/.467 in a context of .227/.297/.273.

After converting to a 1903-13 National League context and adjusting for quality of play, the MLEs suggest that Hill would have hit:
BA/OBP/SLG––LgBA/LgOBP/LgSLG (x pit)––BA+/OBP+/SLG+/OPS+
.313/.393/.433–.262/.329/.345––––––––––119/119/126/145

Among dead-ball era Cuban League hitters, Hill may have been the best; his only competitors are Torriente and Julián Castillo (a slow slugger with a short career and a defensive liability—the Frank Howard of the aughts).

Besides his hitting, Hill was also known for his speed. In this period we have stolen base data for him in 214 games, in which he stole 82 bases (or 59 SB/154 G). I assume that his major league stolen base rate would have been somewhat lower—perhaps 45 to 50 SB/154 G.

So Hill's MLEs during his prime show a very balanced set of skills—above average in hitting for average, power, and drawing walks, as well as in speed. Bill James compared him to Sam Crawford, which may be about right in terms of value, but these MLEs show Hill having less power and drawing more walks than Crawford. Some left-handed batters who may have been more similar to Hill (at about the same ages) include the following:

Player – (seasons) rel avg/rel obp/rel slg/ops+
Hill (1903-13) 119/119/126/145
Collins (1908-1918) 126/128/125/153
Flick (1898-1907) 119/121/130/152
Clarke (1895-1905) 116/115/121/136
G Stone (1905-1909) 125/121/130/151

Hill's MLEs project him as a fine hitter, maybe one of the top ten hitters of the era, but definitely a step below the all-time greats like Cobb, Speaker, Wagner, and Lajoie.

1914-21. There are several gaps in the record for this period. We have data for one season of Cuban League play, 1915-16, when Hill hit .373/—/.412 in 51 AB in a context of .280/—/.334; the 1916 season compiled by Gary, when Hill played for the Chicago American Giants and hit .302/.397/.395 in 172 AB in a context of .258/.331/.336; the 1920 season with the Detroit Stars, recorded in the HoF data, when he hit .271/.457/.427 in a context of .250/—/.328; the 1921 season with the same team, for which I use Gary's slightly more comprehensive dataset, and when Hill hit .329/.424/.507 in a context of .270/.330/.368.

After converting to a National League context and adjusting for quality of play, the MLEs suggest that during 1914-21 Hill would have hit:

BA/OBP/SLG––LgBA/LgOBP/LgSLG (x pit)––BA+/OBP+/SLG+/OPS+
.280/.372/.386–.266/.324/.354––––––––––105/115/109/124

These statistics seem consistent with a typical aging pattern—declines in batting average and power, partly offset by a higher walk rate. Hill's base stealing was also diminished—he stole 21 bases in 147 recorded games.

Career totals. To combine the two periods, I need to assign playing time to each period. Based on playing time of a few major league outfielders with similar length careers, I assigned him 9,585 AB, with 5,747 AB during the first period (or 522 AB/yr) and 3,838 during the second period (or 480 AB/yr). The resulting career line is

BA/OBP/SLG––LgBA/LgOBP/LgSLG (x pit)––BA+/OBP+/SLG+/OPS+
.300/.384/.414–.264/.327/.349––––––––––114/118/119/136

These MLEs imply 2,873 career hits (which would tie him with Babe Ruth at # 39 on the all-time list), 3,972 total bases (# 78), 4,192 times on base (# 29), and 598 stolen bases (# 18).

Regarding his fielding, Riley says he was "one of the fastest outfielders in the game, fielded flawlessly, and had a deadly arm." However, as I noted on another thread, Hill did not play exclusively in center field and spent quite a bit of time as a left fielder. From rosters published in Clark and Lester, Hill's main position was:

CF – 1907 (Leland Giants), 1908, 1909, 1910, 1911, 1912, 1913, 1914, 1916, 1917, 1919, 1920, 1921 (12.5 seasons)
LF – 1904, 1905, 1907 (Philadelphia Giants), 1915 (3.5 seasons)
RF – 1918 (1 season)
OF – 1903, 1906 (2 seasons)

However, for the two seasons for which we have Gary's data compiled from box scores, Hill's time was split as follows:

1916 – CF 46%, LF 39%, RF 12%. C 2%
1921 – CF 36%, RF 33%, LF 16%, 1B 15%

In his series with Negro league teams visiting Cuba, Hill played:

LF – 1905, 1907, 1908 (3 series)
CF – 1906, 1910 (2 series)

In 3 Cuban League seasons, he played:
1906-07 – LF 87%, CF 13%
1907-08 – RF 45%, CF 37%, LF 17%, 1B 1%
1908-09 – LF 100%

Chris Cobb,

I notice that Cristóbal Torriente doesn't have full-blown MLEs. Are you planning to calculate MLEs for him? If not, I can run him through my simplified system.

Brent,

The Pete Hill study is great!

Thank you for reminding me about Torriente. I can and will run MLEs for Torriente for 1916, maybe 1917-1919 depending upon how much data Holway has, and 1920-27.

I think it would be very valuable for you to run him through your system, too. You can give a fuller view of the 1913-20 period in his career, and I think you have a better handle on the Cuban data than I do: I don't normally use it, but I work it in for players, like Torriente, who could be viewed as having the main part of their career in Cuba.

Brent,

Great stuff. Hill was elected right before I started with the HoM so this gives me a much fuller picture. Thanks.

Chris,

Ok, let's both run Torriente's MLEs. Hopefully they'll tell about the same story, at least post-1920. If not, it will give us something to talk about.

Chris & Mulder & Scully

Thanks for compliments.

In Cuba for the American series (plural), is one "context" derived from all of the American series --Cin 1908 plus Det 1909, etc?


the 1916 season compiled by Gary, when Hill played for the Chicago American Giants and hit .302/.397/.395 in 172 AB in a context of .258/.331/.336;

On the continent before league play, is the "context" defined by a set of "major" teams --Gary having compiled all available box scores for the major teams?

--
When was the Schorling Park of the Chicago American Giants constructed?

[These questions may be useful to readers even if KJOK has flown the coop and no other visitor has that data.]

KJOK on the 1928 NNL ballparks, seeking some indication about the Chicago ballpark in the teens
#14
Here are the raw numbers:
Chicago Home Record 48-29-1
Chicago Road Record 15-18
Giants Runs scored at home - 309
Giants Runs allowed at home - 213
Giants Runs scored on road - 202
Giants Runs allowed on road - 186

#16
16. KJOK Posted: May 27, 2004 at 03:18 PM (#647818)
OK, here are the same numbers adjusted for the # of Home and Away games, adjusted for STATS/TOTAL BASEBALL convention, AND regressed 75%

Are there neutral games in the record? If not, then the Cuban Stars record should match the difference between home and away for the other seven teams.

At what level of detail to you have data --only the season totals as above for all eight teams?
If season totals only, then the relative offense/defense quality of the Cuban Stars and the (im)balance in their schedule distorts all the calculations, I think.

Chicago may have played extra home games against the other six teams with ballparks, but that 78:33 home:road count suggests that Chicago played a lot of Cuban Stars games too.

In #32 and #34 KJOK gives 99% sure
- construction date : 1901 (or 1900) for Comiskey's White Stockings - 99% sure
- neutral sites: none for league games

This doesn't really have to do with ballparks or league context, but in a study I'm conducting for the Strat-O-Matic Game Co., Pete Hill pencils out as almost equally effective against left-handed and right-handed pitching. Regardless of who was on the mound, Hill's power, BA and OBP stay consistent, with a slight edge (of course) against right-handed pitching. Not the same can be said for a number of the NegLg HOMers, where some interesting splits have developed...

In Cuba for the American series (plural), is one "context" derived from all of the American series --Cin 1908 plus Det 1909, etc?

Yes. For each of the four series, Gary provided me with the series averages, excluding pitchers. (Thanks, Gary!) To calculate the "context" for the combined four series, I weighted the rates for each series by Hill's AB (for batting average and isolated power) and by his AB+BB+HBP (for walk rate and on-base percentage). I used similar calculations used when combining other leagues and/or series. I believe that's how bb-ref does similar calculations.

On the continent before league play, is the "context" defined by a set of "major" teams --Gary having compiled all available box scores for the major teams?

Yes. Gary's compilation for 1916 is available here.

For 1921 (after league play starts) I use his data here. I used the version that includes games against some major non-league teams (mostly eastern teams, which weren't yet organized into a formal league) because it's a larger sample.

If you look at these spreadsheets, you'll see that they provide detailed information on home-road splits, etc. I didn't make park adjustments to the U.S. play of Hill and Lloyd, though I probably should have. In Cuba, I feel safer not making park adjustments since most games of that period were played in Almendares Park in Havana (though a few games were played in Matanzas).

I’ve been giving some serious thought and study to the Chicago American Giants of the 1910s lately, and since this thread is alive and well, perhaps this is the place to post them.

One of my favorite ways to try to crack open the puzzles of Negro League history is to look at the instances in which there is the widest gap between a player’s anecdotal reputation and the statistical evidence. On the timeline of NeL research, this started almost immediately, with the realization that Satch Paige may not in fact have been the greatest blackball pitcher ever. From that, it took a couple of decades to realize that Cool Papa Bell was also immensely overrated, not because he wasn’t a fine player, but just because he had the largest public relations portfolio of any Negro Leaguer after Paige--who himself was a large part of Bell’s PR campaign.

There are also many fascinating examples of being underrated. One of the most common--players who toiled in lesser markets--is still very much in force in baseball today. In NeL history, this would apply to guys like Verdell Mathis, who pitched virtually his entire career in Memphis.

On the same side of the ledger, you have players like Charles Blackwell, Branch Russell and Pythias Russ who have scintillating statistical profiles, but get little respect from blackball historians. Now, two of those three are St. Louis Stars, so the park factor clearly plays a role there. Yet, Willie Wells‘ St. Louis homerun numbers are still taken to be meaningful, despite being distorted by playing in that same park. One of my personal favorites in this category is Walter Davis, who was a monster on offense, and is never taken particularly seriously.

In a word, there’s still a lot of work to be done in this regard: slowly but surely, I believe we will begin to be vigilant about closing those gaps between anecdotal reputation and statistical reality. In fact, the work you guys are doing is helping a whole lot.

OK, to make a long story, umm, long, this brings me back to the Chicago teams during Rube Foster’s glory days as owner and/or manager, which run roughly from the late 19-oughts through 1922.

The problem is that many of his players from that era have been overrated--not the team, which was a consistent winner, but the individual players. The poster boy for this problem is not so much Pete Hill, but Bingo DeMoss. The DeMoss thread, which has been quiet for about 4 years, reached the conclusion that, in order to make him a legitimate HOM name, he has to be “the Mazeroski candidate” at 2B. That’s true enough, although the problem isn’t so much (as was posted) that there weren’t excellent secondbasemen after DeMoss (Allen, Scales, and Hughes, to name three), it’s that the kind of ball that DeMoss played simply doesn’t translate outside the context of Rube Foster’s managerial style. Said another way, even with his defensive credentials and his base-stealing ability, how exactly does a .236 hitter get mentioned as one of the greats (much less the very greatest) at his position?

I’m going to suggest three theories for the overrating of Foster’s players. The first is that the success of the team as a whole showered little bits of gold dust on the individual players. That’s a natural (and to some extent legitimate) phenomenon. But to give an example of how it lurches into excess, John Henry Lloyd named Jap Payne as the starting rightfielder on his all-time all-star team. Oh please. Payne was a decent ballplayer, but your all-time team? Others, like Pete Duncan and Wes Pryor also tend to get the star treatment from historians, despite the fact that they were pretty ordinary.

The second theory flows from the stature of Rube Foster himself. To compare Foster’s position in blackball to Babe Ruth’s in the white game is entirely fair--except that Foster was more important. He was a great player, a legendary innovator as a manager, and he was personally responsible for the construction and promotion of league blackball. In other words, what Foster said pretty much had infinite weight. He was always quoted as saying that his 1910 team was the greatest baseball team ever, black or white. Of course it was nothing of the sort. But since Rube said so, it‘s taken seriously.

The third theory is about Foster the manager, and how he thought about his team’s offensive production. He had a lot of ideas that would curl any sabermetrician’s hair, and for good reason: they worked far better in the scruffy blackball environment of the 1910s than they would today. Basically, he encouraged OBP, and then a very extreme form of the running game, in order to force defensive mistakes--which it did. In this respect, he was entirely content with what we would consider to be low-percentage plays, like bunting a mind-bending number of times, and then taking brazen risks on the basepaths, even when runners looked like they were going to be thrown out. The low percentages didn’t bother him, because there were inevitably going to be one or two situations in every game in which he got some runs out of creating chaos on defense that he hadn’t earned--and probably wouldn’t have been able to earn--with his bats. Given that he always had great pitching on those teams, that’s all it took to win.

OK, let’s take a minute to review. When the hitting stats first started to come out on those players who had big reputations on Foster’s teams, a lot of us couldn’t believe our eyes. DeMoss hit what? Jelly Gardner hit what? Putting those three theories together explains a lot. With the exception of Hill and Torriente (and Lloyd and Monroe for a few years) Foster really didn’t any bats. The combination of good pitching and defensive mistakes by the opposition was his formula for success.

Now, on to Pete Hill (and thanks for sticking in here with me). One of the other prominent clichés that came out of the Foster era was that the integrated all-star outfield of the deadball era would have been Hill, Speaker and Cobb, left to right. On that basis, I assumed that Hill would have been comparable to Speaker as a hitter--an assumption for which there turns out to be no evidence. Hill was a fine hitter, no doubt, but he was not a great hitter.

In my overall ranking of blackball position players from the 1910s, I would put Hill 7th, behind Lloyd, Charleston, Torriente, Grant Johnson, Santop and Poles, in roughly that order. You could even make an argument that he belongs 8th, behind Ben Taylor. So geez, how can I put a guy who ranks 7th in his decade in my integrated all-star outfield. In a word, I can’t.

I think the cases for my first four picks make themselves. I’ll probably get something of an argument regarding Santop, but I’m sticking to my guns there. Santop hit the dead ball harder than most players have ever hit the lively ball. There is no--repeat NO--white comparable for Santop in the 1910s (and don’t give me Gavvy Cravath--that’s a non-starter). As to Spot Poles, it depends how much emphasis you want to put on defense, since Hill was clearly superior to Poles in the outfield. But Poles was just as fast or faster, was probably a better hitter for average, and had surprising extra-base power for a small man. Hill himself was not a power hitter--he was a gap hitter. So his offensive line is going to end up looking more like Poles’ than one might suspect.

So, in summary, the awe in which Pete Hill’s reputation is held draws in part from each of the three ideas I’ve suggested. Among them, I believe he stood out primarily because he was--like Torriente--an aggressive and dangerous hitter on CAG clubs that bunted the ball and stole bases. When Hill or Torriente stepped to the plate, pitchers had to get real serious real fast. So that part of his reputation is entirely earned. And, just to be perfectly clear, I have absolutely no problem with him being elected to the HOM. But his legend is nonetheless inflated. In blackball history as a whole, I think he is of comparable value to other five-point players like Alejandro Oms and Wild Bill Wright. That feels to me like it puts him in the proper place in the blackball constellation.

--David Lawrence

Torriente was a younger player whose peak we should expect to find in the teens.

For J. Preston Pete Hill some of the crucial exaggerations should be those by Sol White who co-hired him in 1903, taking the Philadelphia Giants to the next level in their second season. I understand that Phil Dixon has reconstructed the 1905 Philadelphia Giants from newspaper coverage that includes box scores for most games. I suppose that means he has published season records for the Giants (Phiants? no, Philly) and their opponents. That alone would be painfully meager. In fact via Brent we have instead the American Series records from Cuba, which include a "context" that cannot be derived from the record of one team even if complete. If I am right about Dixon's work it would yield only comparisons among batters and among pitchers on that team.

The Negro Leagues Book rosters show Hill with Philly through the glory years 1903-07, matching White's tenure as manager. Foster departs in 1907 to be co-manager of the Leland Giants in Chicago; Hill follows next season.
. . . So he played one Cuban League season while in Philly (pre-1907) and played one season in Chicago before the first American Series (post-1908 v Cincinnati Reds).
He was a teammate of Foster from 1904 except 1907 and he played under Foster & Leland from 1908. My point is not so sharp as I thought 30 minutes ago because he may have been a close observer throughout Hill's prime, except in Cuba and in 1907. But discussion specific to "Foster's Giants" may miss his peak seasons. Comparison with white batters of the deadball era would go on to note that the greatest white batters were still super after age 30.

I’ll probably get something of an argument regarding Santop, but I’m sticking to my guns there. Santop hit the dead ball harder than most players have ever hit the lively ball. There is no--repeat NO--white comparable for Santop in the 1910s (and don’t give me Gavvy Cravath--that’s a non-starter).

I don't know the precise point and can't guess why Cravath is a non-starter. Probably he put several years of hard hitting into the Pacific Coast and American Association. If you mean consistent hard hitter with many home runs inside the big parks, Sam Crawford may be your man.

--
Poles is one of the players knocked down from his pedestal here. Like Bell in type but his pedestal was down in the lower quarter so he didn't get many votes for the Hall of Merit.

If you mean consistent hard hitter with many home runs inside the big parks, Sam Crawford may be your man.

Joe Jackson, too.

Poles versus Hill? Interesting argument. Obviously there isn't a lot of data for either one (especially Poles), but in a limited number of box scores, Poles is abysmal against left-handed pitching...

Scott Simkus,
What time period are you researching in detail?
Is the pitcher's hand known for most games covered in box scores?

Paul, I have Poles box scores from 1909 to 1923. Obviously, we don't know the handedness of every pitcher (especially during the pre-Negro League dead ball era), but in boxes against those of whom we know for sure were lefties (the usual suspects: Juan Padron-thanks to Gary A., John Donaldson, Dave Brown, etc.), Spot Poles really struggled. Crazy question, but was he really a switch-hitter or did he bat lefty exclusively? Or does everybody already know the way he hit and I'm just the last one to the party? I'm curious. Between the two of them (Hill and Poles), I've collected approx. 350 boxes (Negro vs. Negro team)... real small sample. But for what its worth (and I'm not suggesting its worth much), I'd take Hill over Poles based on having read every stinking box score and game account.

In the major league biographical database maintained by SABR, a player could be listed "Bats: B" (both) if he switch-hit one winter and spring, perhaps working at it because he had been batting poorly against same-hand pitchers.

What about Hill? Riley may be right about 1901-02 (Cuban X Giants), although right may be insignificant if he played only occasionally.

Brent #40
Riley says he then played for the Cuban X Giants in 1901-02, but Clark and Lester and Holway don't show him on the X Giants roster for those seasons.

Clark and Lester, The Negro Leagues Book does not provide any roster for the X Giants, either season, and none for J.M. Bright's teams between the 1900 "Genuine Cuban Giants" and the 1905 "Famous Cuban Giants". Anyway I believe that some of their season rosters and some of Holway's are generated from only a few box scores, maybe only one box score on occasion.

Paul,

My comment about Hill and the 1901-02 Cuban X Giants may have been misinterpreted. My intention wasn't to question Riley. I was trying to explain (not very well) why I picked 1903 as the starting point for the MLEs. Let me try to fill in some more explanation.

Before 1905, we simply don't have much Negro league data, so it's impossible to be sure precisely when Hill demonstrated that he was a major league quality player. It was surely before 1905, because by then, he was regarded as one of the top players in black baseball. On the other hand, unless there were compelling evidence that he was an outstanding player before 1903, the uncertainty about his age would make me hesitant to start his MLEs pre-1903. (With the younger birthdate, he would have been only 19 years old in 1902.) So that leaves 1903, or maybe 1904 if you're more conservative, as plausible start dates. Holway has some data for those two years, but they are based on only a handful of box scores, so I didn't pay much attention to them. My choice of 1903 as Hill's starting season is admittedly arbitrary, but given the uncertainty about his birthdate, it seemed like the season that would be reasonable with either birthdate.

Crazy question, but was he really a switch-hitter or did he bat lefty exclusively? Or does everybody already know the way he hit and I'm just the last one to the party?

No, Scott: there's still a discussion going on about whether Poles batting both. The problem, as I understand it, is that nobody's found a picture of him batting right-handed. Either that, or an entirely credible newspaper description of him batting from the right side (very unlikely--why would you write about it?) would be needed to establish that he switch-hit. But that would still leave the problem Paul Wendt describes in the first paragraph of post 57--which is real enough. I have a picture of Honus Wagner batting left-handed, which he did from time to time just to level out his swing. That doesn't make Wagner a switch-hitter, in truly meaningful terms.

I have Poles box scores from 1909 to 1923. Obviously, we don't know the handedness of every pitcher (especially during the pre-Negro League dead ball era), but in boxes against those of whom we know for sure were lefties (the usual suspects: Juan Padron, John Donaldson, Dave Brown, etc.), Spot Poles really struggled....Between the two of them (Hill and Poles), I've collected approx. 350 boxes.

First of all, Scott, if half of those boxes are for Poles, and the majority of those are pre-league, you may have more than anybody else. You're in touch with the NLRAG, right? Second, the primary problem with your Poles hypothesis may not be scant data, but rather that you just named three of the ten best lefthanders in blackball history. I would assume everybody struggled against them--not just Poles.

But please, for my edification, let's hear your argument for choosing Hill over Poles.

Crazy question, but was he really a switch-hitter or did he bat lefty exclusively? Or does everybody already know the way he hit and I'm just the last one to the party?

Riley lists him as Bats: Both, but in his narrative says "A left-handed batter, he watched the ball all the way to his bat and consistently hit for a high average."

So, there's certainly some doubt about him switch-hitting.

Burnis and KJOK, thanks for your input on the handedness of Poles. I appreciate it. It’s a subject of keen interest to me because it has to do with my job… namely, creating the most accurate player profile of Spot Poles possible for a game company, and I appreciate the insight of those like you who are passionate about the subject.
Brother Burnis: Unfortunately, the majority of box scores are Pete Hill. Most Poles are pre-league, of course, but I seriously doubt I’ve collected more Spot boxes than the 50-man army of researchers. I’m approaching 100 for Spot against Major competition, which is probably close to what they have. But who knows?
One of the things I dislike about these open forum blogs (and the primary reason I participate more as an enthusiastic observer than contributor), is how quickly these things can degenerate into electronic fistfights. In light of the Santop/Cravath comments, it almost seems as if we’re piling on you Burnis. And worse, for me, it seems like I’m jumping on top of the pile at the last minute to get in a quick sucker punch. In reality, I’ve read a number of things you’ve posted over the past year and find them fascinating. If you can write 300 pages of it and find a publisher, I’d buy it! I’m simply curious about Hill versus Poles question for professional purposes and value your insight.
I don’t know if I’m equipped to provide any “edification,” but as for the Hill versus Poles issue… When evaluating two players who are so similar in ability (and I think we can agree these two players were similar enough so that there is no clear cut favorite), I tend to favor the player I know more about. We know Hill had slightly more power, higher OBP, higher BA. We know he had a longer career. We know he was slightly more versatile… play more positions defensively. We know their defensive numbers are so similar as to be a draw. We know they both had speed, with possibly an edge to Poles. We now know they both hit against the same high-quality lefties and that Hill was clearly better. FWIW, I now know they both hit against many of the same righties, and Hill was better.
I’m certainly open to the possibility that Poles was better than Hill, but we just need more data. Gary Ashwill is sitting on a treasure trove of Cuban Lg boxes, and he made hold the key for explaining away Poles’ apparent weakness against lefties. Maybe he hit well against lefties in Cuba, who knows? The problem with Spot is a lot of his reputation was established against semi-pro competition. You know, the .440 BA with 41 stolen bases in 60 games. I think 10 years from now, semi-pro numbers are going to become part of the discussion, but we just don’t know (or have time) to figure out what they mean. We’re still too busy tracking down Negro League and pre-league boxes. I’ve studied the Chicago semi-pro league (certainly as tough as Poles’s NY semipro scene?) from the 1910 era, and Nixie Callahan (the White Sox outfielder who quit organized ball at the height of his career, then later returned successfully to the big leagues), would steal over one base per game! This was in a circuit with Rube Foster’s team, and dozens of former Major and Minor leaguers, Jose Mendez’ traveling Cuban ballclubs, college stars, etc.. Maybe 41 sb in 60 games isn’t that great? Maybe it is?
Anyway, my preference of Hill over Poles is simply a preference of penicillin over prayer. I’m going to go with what I know, but I’m certainly willing to be swayed with more data.
Thanks again, guys. I’ve got to go golfing now… its my wife’s birthday…I’m going to break 80… we’re only playing 9 holes.

Scott Simkus
how quickly these things can degenerate into electronic fistfights. In light of the Santop/Cravath comments, it almost seems as if we’re piling on you Burnis.

?

Brent
an entirely credible newspaper description of him batting from the right side (very unlikely--why would you write about it?) would be needed to establish that he switch-hit.

I have read only a few such descriptions and they all concern either concern "other side" experiment or practice. So and so batted from the left yesterday or so and so quit the switch-hitting experiment this week.

Post 63....? Gotcha, Paul Wendt. Truth be told, the baseball people here have strong opinions, but tend to be a rather civil, erudite bunch.
As it stands, due to a fast approaching deadline and a lack of any new information, Spot Poles is going to be carded as a switch-hitter in the SOM game.

The only thing I'd add is that the American Giants' Schorling Park (formerly White Sox Park) really cut down on batting averages and run-scoring relative to other parks used by black teams. Even in 1916, there were fewer runs scored there than in the other two main venues of black baseball in the midwest (the parks in Indianapolis & St. Louis). This, IMO, is the single biggest reason the American Giants' batting averages don't, at first glance, match the players' reputations.

Also, another fact about Hill we didn't know until recently is that he did walk quite a bit, as much as (or more than) any other contemporary Negro Leaguer. So there is some additional value hidden there.

With Poles, what is the basis, in the first place, of the notion that he was a switch-hitter? I'm assuming somebody said he was--a teammate, somebody who saw him play? Anybody know?

I seriously doubt I’ve collected more Spot boxes than the 50-man army of researchers.

Is the "50-man army" compiling pre-league data? I thought the HoF project was limited to the post-1920 organized league period.

Any rumors when the long-promised Negro league encyclopedia might (finally) be published?

With Poles, what is the basis, in the first place, of the notion that he was a switch-hitter? I'm assuming somebody said he was--a teammate, somebody who saw him play? Anybody know?
I'm assuming somebody here must know how to contact James A. Riley. It would be nice to ask him that question.

and regarding the Strato-o-matic team:
Frankly I would have estimated that the simulation baseball gaming industry supports an army of 0 researchers and compilers for hundred-year old baseball game data. Nothing more than things like spot-checking some 1886 and 1906 games to see where Frank Grant and Grant Johnson probably batted in the lineup --something that a more rather than less interested employee gets to spend a few days on during the course of a year.
(I know that another simulation game relies heavily on enthusiastic customers to donate data they compile for their own replays of old seasons. A player compiles all of the starting lineups for the 1890 Players League during the course of simulating the season, later makes a data file available to the community, both other players and the game developer, for no $ payment. By "relies heavily" I mean that so much data originates that way, not that boosts the revenues of the company or its stock price enough that anyone could justify data-gathering/organizing as a business.)

Are you SS or the strat-o-matic business also collecting newspaper coverage of Sol White and Frank Grant careers?

You can contact Holway through his website:

http://baseballguru.com/jholway/

Are you SS or the strat-o-matic business also collecting newspaper coverage of Sol White and Frank Grant careers?
I personally have a few White and Grant box scores, but not enough to "card" these players by Strat-O-Matic's standards. Couple dozen games doesn't cut it for their product...Can't really speak to the methods of other simulation games companies, other than to say what you describe sounds like garbage.

After a discussion with Chris Cobb on the Major League Equivalancies thread, I've decided to switch the conversion factor for walks from .81 to .95, consistent with Chris's recommendation. This new conversion factor leads to some modest changes to the Hill MLEs I posted here:

Career:
BA/OBP/SLG––BA+/OBP+/SLG+/OPS+
.300/.393/.414––114/120/119/139

1903-13:
BA/OBP/SLG––BA+/OBP+/SLG+/OPS+
.313/.403/.433––119/123/126/148

1914-21:
BA/OBP/SLG––BA+/OBP+/SLG+/OPS+
.280/.377/.386––105/116/109/125

Another interesting implication of these revised MLEs is that they show Hill retiring with the career record for walks--my estimate is 1,265 BB, which would have surpassed Billy Hamilton's record of 1,187. Less than 3 years later, however, Hill's record would have been broken by Eddie Collins. (Hill was also frequently hit by pitches; my estimate is that his career total would have been 181 HBP, which would have ranked him # 5 at the time he retired.)

I just noticed a mistake in my spreadsheet, affecting the 1914-21 and Career lines. Here's a corrected version:

Career:
BA/OBP/SLG––BA+/OBP+/SLG+/OPS+
.300/.397/.415––114/122/119/140

1903-13:
BA/OBP/SLG––BA+/OBP+/SLG+/OPS+
.313/.403/.433––119/123/126/148

1914-21:
BA/OBP/SLG––BA+/OBP+/SLG+/OPS+
.280/.388/.386––105/120/109/129

Do we have eight "Negro Leaguers" in centerfield with Jim O'Rourke?
Hill, Torriente, Dihigo, Oms, Charleston, Bell, Stearnes, Irvin

Do we have nine "Negro Leaguers" in centerfield with Jim O'Rourke?
Hill, Torriente, Dihigo, Oms, Charleston, Bell, Stearnes, Irvin, Brown

Dihigo is pitching, but there's Willard Brown and Larry Doby to boot, so in alphabetical order:

Bell, Brown, Charleston, Doby, Hill, Irvin, Oms, Stearnes, Torriente.

Dihigo is pitching, but there's Willard Brown and Larry Doby to boot, so in alphabetical order:

Bell, Brown, Charleston, Doby, Hill, Irvin, Oms, Stearnes, Torriente.

I'm sure I've made this point before, but in a league that isn't at the VERY TOP of the pyramid, you're going to have the best players cluster around SS, C & CF, and many players would move to the right of the defensive spectrum. If they had been able to play in MLB, a lot of these guys would be RF or LF or maybe 1B, etc.

"If they had been able to play in MLB, a lot of these guys would be RF or LF or maybe 1B, etc."

I would have moved Brown to an outfield corner, definitely. Perhaps Bell to left too, because of his weak arm. But then there's the Ashburn precedent, etc. And Irvin wasn't a centerfielder at all. The rest are legit CFers though.

Dihigo, Irvin and Torriente played 1B, but their talents were not well-used there. Much better examples of guys who would have been moved to 1B are substandard defensive infielders like Beckwith, Scales and Clarkson.

Thanks.
By the way that was my approximation to chronological order (as usual) without looking it up.

I'm sure I've made this point before
[I think so but it's worth repeating]
, but in a league that isn't at the VERY TOP of the pyramid, you're going to have the best players cluster around SS, C & CF, and many players would move to the right of the defensive spectrum. If they had been able to play in MLB, a lot of these guys would be RF or LF or maybe 1B, etc.

Mule Suttles started as the fringe of that cluster, in LF, and moved to 1B. I think of him as the Greg Luzinski of the Negro Leagues. Some of Riley's description fits (btw he lists '1b, lf' both in bold). The nicknames Mule and Bull help, but I suppose they were conferred partly for similar reasons, if not wholly.

I like that they are all classified together here. It will enforce some more comparison of the NeL outfielders among themselves, and this project won't rank them 'as centerfielders' but 'among the centerfielders'.

"I think of him as the Greg Luzinski of the Negro Leagues."

I realize this belongs in the Suttles thread, but that's too far down the food chain, Paul. Suttles was a monster hitter, no matter how one-dimensional he might have been. Kiner or Killebrew would be closer to it, or even a right-handed McCovey. If you needed to step it down a little, perhaps Frank Howard. But I think Suttles was far scarier than Howard.

Fred Worth has found a copy of JOHN Preston Hill's death certificate, and the birth date on the certifcate says October 12, 1882, which confirms Patrick Rock's finding, and means his HOF plaque is incorrect.

Oh, I have seen reports of John Hill and now I wonder whether John Hill around 1901 is Pete Hill.

Paul - that would be a good find. His real name is apparently John Preston Hill, not Joseph or Pete...

There was also an infielder named John Hill who was about Pete Hill's age & who played on some of the same teams. In fact, that might be part of the reason Hill started going by his middle name Preston or the nickname Pete.

I hate to pimp for my blog, but <url=http://agatetype.typepad.com/agate_type/pete-hill>go here</url> to see Pete Hill's death certificate and some of the other records found by Patrick, Fred, & the rest.

Sorry, I can never remember how to do links properly here. Try this.

Or just this:

http://agatetype.typepad.com/agate_type/pete-hill

This article seems to be related to what Kevin, Paul, and Gary are talking about in posts 80-85. Any comments on the linked article?

"Hill of Pittsburg will play for the Unions this year"? (No, I haven't read that anywhere.)

Daily newspapers and Sporting Life published many baseball notes such as "Olean has signed Jimmy Barrett of Brattleboro" or "Jimmy Barrett of Brattleboro will play for Olean this season". The meaning may be that Barrett grew up in Brattleboro or lives there or last played for the Brattleboro ballclub.

I suppose that such notes, and slightly longer but no less ambiguous ones, are the sources for much birth city and state data where birth month and day are missing.

For a city so small as Brattleboro, I believe, even "of Brattleboro, Vermont" may mean that he last played for the Brattleboro ballclub.

Link to post about finding his grave.

Pete Hill's Real Stats

Hi, everyone,

Please find my MLEs for Pete Hill. They will be updated as new info comes to light.