Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Hall of Merit > Discussion
Hall of Merit
— A Look at Baseball's All-Time Best

Tuesday, August 17, 2004

Ben Taylor

The best Negro League first baseman before Buck Leonard. How does he compare with his white contemporaries?

John (You Can Call Me Grandma) Murphy Posted: August 17, 2004 at 09:35 PM | 81 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. Chris Cobb Posted: August 18, 2004 at 02:50 AM (#803452)
Here's the Ben Taylor data I've gathered

Expert Assessments
Bill James -- #3 First Baseman
CPPD -- 80% of experts voted to elect to HoF
Holway -- 8 all-star selections; 3rd team first-baseman on all-time Negro-League all-star team (Mule Suttles & Buck Leonard were selected for the first and second team)

From Riley

.334 lifetime batting average
1909 22-3 with Birmingham Giants
1911 30-1 with St. Louis Giants (vs. all levels of competion)
1912 .379 with St. Louis Giants and NY Lincoln Giants (vs. all levels?)
1914 .333 with Ind ABCs
1915 .308 with ABCs
.500 in Cuban play
1916 .335 with ABCs
1920-22, .323, .407, .358 with ABCs
1923 player-manager for Washington Potomacs, not of ML caliber
1924 .314 and p-manager of Potomacs
1925 .328 with Harrisburg, p-manager
1926-28 .242, .307, .336 with Baltimore, p-manager
1929 .322 with Bacharachs, p-manager

From Holway
1910 0-3, .125 ba for West Baden Sprudels
1911 2-0, ..667 ba for St. Louis Giants
2-0 for West Baden Sprudels; Holway all-star at pitcher
1912 .375 for West Baden, 0-1 as pitcher
1913 .400 for West Baden
.347 for Chi Am Giants; Holway all-star
1914 .323/.318 for Ind ABCs; 5-3 as pitcher; Holway all-star, MVP
7 HR (22/550 ab), 24 2b, 6 3b, 19 sb (all lead league). Appx. 175 ab
Totals 175 ab, 56 hits, slg % of .646
1915 .356/.280 for ABCs; 19 2b, 0-1 as pitcher; Holway all-star
1-9 vs. major-league competition
1916 .327 for Ind ABCs; Holway all-star
4-19 in playoff vs. Chi Am Giants
0-3 vs. major-league pitching
7-18 in Florida Winter League
1917 .234 for ABCs, 0-1 as pitcher
3-3 vs. major-league competition
1918 .229 for ABCs
1919 .200 for AC Bacharach Giants
3-13 vs. major-league competition
1920 .359 for Ind ABCs; Holway all-star
1921 .374 for Ind ABCs, 22 2b leads league; Holway all-star
1922 .364 for Ind ABCs, 30 2b leads league
1923 no data
1924 .319 for Was Potomacs
1925 .333 for Har Giants
1926 .181 for Bal Black Sox
1927 .282 for Bal Black Sox; Holway all-star
1928 .322 for Bal Black Sox
0-4 vs. major-league competition
1929 .306 for AC Bacharach Giants

Holway Career totals
1909-34 937-2920, .321
16-50, .320 vs. major-league competition
.335 mean avg., 1911-1922, 1924-29

i9s career totals
1910-1926
8257 ab, 2531 hits, 541 2b, 81 3b, 117 hr, 834 bb, .307 ba, .370 obp, .434 slg, 804 ops, 1307 rc

Assessment:

I haven’t done a season-by-season WS projection for Taylor yet, but I have worked through the following rough estimate of Taylor’s career value.

Closest offensive match (and surprisingly close – almost identical career span and length and pro-rated hits) is Max Carey. Carey has 254 bws: I boost Taylor by 3% over Carey to 262. Give Taylor an A at 1b for defense and 3 seasons fewer defensive innings than Carey leaves him with 35 fws, as against Carey’s 94.8. Pro-rate that by seasons and it rises to 40. Taylor has about 30 pitching ws, I think. So 262 + 40 + 30 = 332. Season-adjusted to 162 games, that becomes 349. Peak and peak rate will not be great, most likely, though he did have a series of excellent seasons, 1912-1916.
   2. KJOK Posted: August 19, 2004 at 01:09 AM (#805300)
Taylor's 1928 Batting Stats:

G-67
AB-256
H-78
D-16
T-3
HR-2
R-30
BB-17
Sac-7
SB-4
AVE-.305
OBP-.348
SLG-.414
   3. Michael Bass Posted: August 19, 2004 at 06:28 PM (#806691)
I'm definately interested in your Win Shares breakdown on Taylor. He has no shot at a top 6 berth on that ballot, but I want to give him full consideration before he goes way deep into the backlog.

-----------------------------

re: His pitching, of what calibre are we talking? You've given him credit for 30 pitching WS here; unless told otherwise, I'm thinking those are not win shares he would have accrued in the major leagues.
   4. Chris Cobb Posted: August 19, 2004 at 08:38 PM (#807327)
of what calibre are we talking? You've given him credit for 30 pitching WS here; unless told otherwise, I'm thinking those are not win shares he would have accrued in the major leagues.

Well, back to the problem of Negro-League pitching. . .

Here's what I think about Taylor's pitching, interpret it how you will.

1) His pitching vs. position-player career is shaped like that of George Van Haltren or Bobby Wallace -- starting as primarily a pitcher, then switching to first base, so we're not looking at occasional fill-in pitching, we're looking at Ben Taylor as a full-time pitcher from 1909 through 1911, with occasional pitching work for the next five seasons.

2) His pitching was good enough to get him onto the top Negro League teams, because that's how he came up. Both Riley and Holway think he was a very good pitcher in 1911 -- Riley has him 30-1 vs. all competition; Holway has him 4-0 vs. Negro-League competition. Holway names him as his western pitching all star for that year also. I see that as good enough to have gotten a shot at the major-league level; others may differ.

3) If we give credit, how much? His 1909 season was for the Birmingham Giants, which was a lesser team, so that probably ought to be treated as minor-league play. He enters the northern game in 1910. It's been my policy to count breaking into the top available competition as the point to start giving MLE credit. Not much info to go on about how good he was that year, so I give him just a bit of credit -- say 6 ws total. He pitches very well as a full-time starter in 1911. For being the top pitcher in the west that year I'd give him credit as a major-league average pitcher. Given usage patterns of that time, that's 14 ws. For the rest of his pitching career, maybe he should get some credit, maybe not. He did a fair amount of pitching in 1914, so maybe 2 win shares there and nothing for the rest.

So, on thinking this through again, I'd give him 22 pws, I think, rather than 30. I may have forgotten that his 1909 season was definitely minor-league.

If you want another view of his pitching, i9s has done pitching projections for him as well as hitting projections. I'll be looking at those more closely when I try to do season-by-season MLEs for Taylor.
   5. Thane of Bagarth Posted: August 19, 2004 at 09:09 PM (#807430)
I thought Taylor was highly regarded as a pitcher before seeing the breakdown of his stats. Maybe it's because he always seemed like a bargain as a good two-way player in Bill James' Classic Baseball.

In real life it doesn't look like he was all that great on the mound outside of those first couple of seasons against undefined competition. If Riley specifically cites one year as against "all competition," what does it mean for 1909 when there is no comment about the competition? That certainly doesn't boost my impression of his opponents' quality.

I don't think I will give him much, if any, bonus for his pitching. He's still one of the top 3 or 4 NeL 1B in my book, well behind Leonard--whether he's 2, 3, or 4 will depend upon a closer review of the reputation and statistics of Luke Easter and Mule Suttles.

Riley mention's Taylor playing for a team managed by his brother, the West Baden Sprudels, in 1910. He ended up winning "a disputed team batting title," but there is no description of his pitching performance. Holway only attributes 3 decisions (all loses) to Taylor that year (BTW, thank you, Chris, for posting all that).

How could someone who went 22-3 for the Birmingham Giants the year before and 30-1 for the St. Louis Giants the year after only get 3 pitching decisions for the West Baden Sprudels? Did he play for some other unrecorded team? Anyhow, after hitting well in 1912 he appears to have been used primarily a 1st baseman.

Anybody know what a Sprudel is?
   6. Thane of Bagarth Posted: August 19, 2004 at 09:15 PM (#807438)
Chris, thanks for the clarification re:quality of opponents. I'm more inclined now to give him a slight boost in value for the pitching, but I'm still somewhat skeptical.
   7. Chris Cobb Posted: August 19, 2004 at 09:42 PM (#807497)
How could someone who went 22-3 for the Birmingham Giants the year before and 30-1 for the St. Louis Giants the year after only get 3 pitching decisions for the West Baden Sprudels? Did he play for some other unrecorded team?

The data for 1910 and the data for 1909 & 1911 are showing different things. Holway's data shows only players' records against other black teams -- competition at more or less their own level.

The numbers from Riley show records against all competition. At this point, black teams mostly played white semipro teams, sometimes participating in regional leagues, sometimes barnstorming. They generally cleaned up against this competition.

So all of the stats that Holway provides from this era record play in only a handful of games per season.

If stats exist for Taylor's play vs. "all competition" in 1910 with the Sprudels, Riley doesn't provide them. We are left to infer that the Sprudels played a bunch of games against lesser competition, since otherwise they couldn't have paid the bills, but we have no idea what they did, though we can assume they won pretty regularly. They did not have that kind of success against Rube Foster's Leland Giants . . .
   8. Thane of Bagarth Posted: August 19, 2004 at 09:59 PM (#807524)
I came upon this while searching for the meaning of “Sprudel.”

It is sort-of a summary of Paul Debono’s book The Indianapolis ABCs, which “is a labor of diligent research and civic pride.” Most of the info is about Ben’s older brother C.I. Overall, it is an interesting article that goes into detail about the rivalry between our buddy Rube and C.I. Taylor.

"Debono's extensive account of black baseball in his hometown is a welcome contribution to literature of black baseball and a requisite addition to the library of researchers and curious students alike."...

"C.I. was the patriarch of African American baseball's most distinguished family. His younger brothers, all of whom attended college, were, in
birth order, pitcher 'Steel Arm' Johnny, third baseman 'Candy' Jim, and first baseman 'Old Reliable' Ben. Johnny was the only pitching specialist of the four, and, as Debono points out, 'the only one not to
manager the ABCs at some point.'"


The boldface is mine. I'm not sure there can be much to make of that comment, but I thought it was interesting that there was no mention of Ben's pitching.

Here’s most of what it says about the Sprudels:

"By 1909 the West Baden Sprudels--one of black baseball's most curious of nicknames--and French Lick Plutos played almost daily for the entertainment of hotel guests, either between each other or against visiting professional and semipro teams. They even formed a two-team association called the Springs Valley League, and took their rivalry seriously. In 1910 C.I. Taylor moved to West Baden to take over the Sprudels and in the following year he signed three players who would play key roles in bolstering the ABCs of the future: Dizzy Dismukes, George Shively, and Morty Clark. The West Baden Sprudels played well against, and occassionally defeated, Rube Foster's Leland Giants, while the French Lick Plutos kept pace by signing skilled players, including, in 1913, Bingo DeMoss."
   9. Thane of Bagarth Posted: August 19, 2004 at 10:04 PM (#807540)
Oh, and to answer my own question, “what is a Sprudel?” This site has a picture and description:
"Sprudel, a little gnome-like creature, was used in promoting the hotel."...

"The name 'Sprudel' comes from Germany. Sprudel is the German translation for 'spring.' On the grounds of the hotel there once stood springs that generated a large portion of the revenue for the hotel."

Knowing is half the battle.
   10. Chris Cobb Posted: August 25, 2004 at 05:15 PM (#818188)
Here are season-by-season MLE win share projections for Ben Taylor.

Basis

Batting -- The i9s career data with a 5% discount matches my translations of Taylor's actual statistics, so I've built these win shares based on their projections with a 5% discount, except for 1927-29. i9s sees Taylor's MLE career ending then, but he played three more years. i9s are right to see his final seasons as possibly not of major-league quality, though he was pretty good in 1928. Folks can make their own decisions about including these seasons in their ranking of Taylor. I've created the win shares by finding the closest equivalent major-league player in terms of OPS and pro-rating the total by plate appearances. I alternated NL and AL seasons (even, NL; odd, AL), so those who make competition adjustments might tweak the totals for even or odd years.

Pitching -- based on i9s projections for years in which there are actual records of Taylor pitching significant amounts.

Fielding -- Treated Taylor as an A- first baseman (2.10 ws/1000 innings) for his career, about equal to Ed Konetchy. Then I varied his rate for each season to reflect a career curve and estimated defensive innings from plate appearances

Year – bws – fws – pws = total
1910 – 5.0 + 0 + 7.0 = 12.0
1911 – 7.2 + 1.0 + 10.0 = 18.2
1912 – 9.7 + 0.8 + 11.0 = 21.5
1913 – 21.5 + 2.6 = 24.1
1914 – 24.4 + 3.4 + 2.0 = 29.8
1915 – 12.0 + 3.1 = 15.1
1916 – 21.6 + 3.1 = 24.7
1917 – 16.8 + 2.9 = 19.7
1918 – 22.0 + 3.2 = 25.2*
1919 – 14.5 + 2.4 = 16.9*
1920 – 19.0 + 2.5 = 21.5
1921 – 20.9 + 2.7 = 23.6
1922 – 16.1 + 2.3 = 18.4
1923 – 12.3 + 2.4 = 14.7
1924 – 8.0 + 2.1 = 10.1
1925 – 11.5 + 1.6 = 13.1
1926 – 4.4 + 1.2 = 5.6
1927 – 1.2 + 0.3 = 1.5
1928 – 5.0 + 1.0 = 6.0
1929 – 2.4 + 0.5 = 3.9

total 325.6 career win shares

*projected to 154 game season, not the number of games actually played in the majors in these seasons.
   11. Michael Bass Posted: August 25, 2004 at 05:27 PM (#818213)
Thanks, Chris!

My feel on Taylor is a Hooper/Beckley/Konetchy-esque career. Average or slightly above for a long time, very little in the way of peak. None of those guys are on my ballot, so I doubt Taylor will either.
   12. John (You Can Call Me Grandma) Murphy Posted: October 13, 2004 at 02:41 PM (#913839)
Chris, do you know how many games and games pitched Taylor played for each season (or where I can obtain this information)? I'm trying to compute his WS per 162 Games for each season. Thanks!
   13. John (You Can Call Me Grandma) Murphy Posted: October 13, 2004 at 02:45 PM (#913844)
BTW Chris, I assume those numbers would be projections, since Taylor wasn't playing a regular major league schedule.
   14. Chris Cobb Posted: October 13, 2004 at 03:05 PM (#913872)
Chris, do you know how many games and games pitched Taylor played for each season (or where I can obtain this information)? I'm trying to compute his WS per 162 Games for each season. Thanks!

The information of the exact number of _actual_ games Taylor played is not available. I calculated projected games played based on the i9s projections as part of my calculations, but didn't save those numbers. If you want to recreate that projection, just use the i9s PAs for Taylor and divide by about 4.2 to get an estimate of games. You could calibrate the divisor more exactly to seasonal conditions if you wish.

BTW Chris, I assume those numbers would be projections, since Taylor wasn't playing a regular major league schedule.

I direct your attention to the first line of the post, quoted below:

Here are season-by-season MLE win share projections for Ben Taylor.

Always read the fine print :-) .
   15. John (You Can Call Me Grandma) Murphy Posted: October 13, 2004 at 03:19 PM (#913894)
I direct your attention to the first line of the post, quoted below:

Here are season-by-season MLE win share projections for Ben Taylor.

Always read the fine print :-) .


I was referring to the games played and games pitched for Taylor for each season if you had them, not for your MLE Win Shares projections.
   16. Chris Cobb Posted: October 13, 2004 at 04:36 PM (#914043)
Ah, I see. Sorry to have misinterpreted!
   17. John (You Can Call Me Grandma) Murphy Posted: October 13, 2004 at 05:17 PM (#914133)
Ah, I see. Sorry to have misinterpreted!

No problem, Chris. I've been known to have done the same thing a few times myself. :-)
   18. John (You Can Call Me Grandma) Murphy Posted: October 13, 2004 at 10:26 PM (#914715)
Going over Ben Taylor's credentials again, I still have him as slightly behind Konetchy, though I think he was the best first baseman (black or white) for 1913, 1914, 1916 and 1918. He could make my ballot sometime in the future.
   19. KJOK Posted: November 26, 2004 at 12:50 AM (#982570)
Given that we really seem to have a 'dearth' of quality WHITE 1st baseman in the 1900-1920 time period, and therefore given that Taylor may be the best 1st baseman, black OR white, during that period, he'll probably bump up to being on my re-evaluated ballot this time.
   20. Gary A Posted: December 18, 2004 at 03:20 PM (#1028973)
1921 Ben Taylor
NNL Indianapolis ABCs

G-72
AB-272
H-109
D-13
T-6
HR-2
R-56
W-23
HP-3
SH-7
SB-9
AVE-.401 (3rd in league; NeL .263)
OBA-.453 (3rd in league; NeL .324)
SLG-.515 (NeL .361)

My impression of the ABCs home field is that it was a pitcher's park.
   21. Dr. Chaleeko Posted: January 14, 2005 at 07:38 PM (#1080122)
Chris Cobb,

Would Gary's additional information on Taylor or the park factor information on other threads alter your WS projections for Taylor?
   22. Gary A Posted: January 14, 2005 at 10:56 PM (#1080638)
My initial impression of the ABCs' park wasn't entirely accurate:

Washington Park (Indy)park factors:
1920: 102
1921: 106
1922: 107
1923: 85
   23. Chris Cobb Posted: January 15, 2005 at 03:05 AM (#1081006)
Dr. Chaleeko,

My MLEs for Taylor are based off of i9s. Since i9s provides projections rather than direct major-league estimates, their season-by-season projections don't always match directly to the shape of the players' careers, although their career totals work as MLEs. Therefore, I can't say exactly what the effect of Gary's info would be, because I don't know the precise assumptions behind their data.

Sometime it would be nice to do an independent set of MLEs for Taylor, but I don't think it would differ all that much from my modification of the i9s projections.

FWIW, I have Taylor slightly ahead of Lundy in my own rankings.
   24. KJOK Posted: February 23, 2005 at 12:42 AM (#1160274)
1916 Indiannapolis ABC's:

G-46 (Lead Team, Team-46)
AB-164 (Lead Team, Tm Ave-151)
H-56 (Tm-44)
2B-4 (Tm-4)
3B-2 (Tm-3)
HR-0 (Tm-0)
Sac-9 (Tm-11)
Ave-.341 (Tm-.291)
Slg-.390 (Tm-.373)

PITCHING:
IP-17
HA-15
SO-10
BB-4
W-1
L-0

Per:
Paul Debono
McFarland Publishing Inc.
THE INDIANAPOLIS ABCS HISTORY OF A PREMIER TEAM IN THE NEGRO LEAGUES
   25. KJOK Posted: March 29, 2005 at 11:25 PM (#1223072)
Suttles has quite a few supports, but Taylor seems to be off the radar screen, even though they roughly project as:

Name PAs OPS+
Taylor 9091 138
Suttles 8010 137

and Taylor was likely the better fielder....
   26. karlmagnus Posted: March 29, 2005 at 11:40 PM (#1223092)
Sometime when poor Chris has a moment, between 2 and 4 am, it would be great if he could produce some MLEs for Taylor. I agree he should be in the discussion with the Wilson-Beckwith-Suttles-Bell group, and may well be 2nd in that group behind Wilson. Would be a pity to elect the wrong guy. I don't trust WS, but Chris's MLEs appear to me at least internally compatible, so give you a good ranking between players.
   27. John (You Can Call Me Grandma) Murphy Posted: March 29, 2005 at 11:43 PM (#1223095)
Suttles has quite a few supports, but Taylor seems to be off the radar screen, even though they roughly project as:

Name PAs OPS+
Taylor 9091 138
Suttles 8010 137

and Taylor was likely the better fielder....


Though I don't have both of them on my ballot, I have Taylor ranked higher than Suttles.
   28. Chris Cobb Posted: March 30, 2005 at 01:59 AM (#1223263)
Sometime when poor Chris has a moment, between 2 and 4 am, it would be great if he could produce some MLEs for Taylor.

A problem for Taylor is that statistics for the first half of his career are _much_ more fragmentary than for the second half of his career. I don't have the data to do my usual work-up. I can do it for his post-1920 play and make some rougher estimates for his play in the teens, but I can't produce truly comparable data for his full career.

I'll try to do what I can (and also revisit Dobie Moore) before the voting begins for 1949, since Suttles might be in line for election then. I'm hopeful that the Hubbell/Mackey combination will give us at least another year to think about Suttles' placement, but I don't know exactly how Mackey's going to look yet.
   29. Gary A Posted: March 30, 2005 at 04:37 AM (#1223440)
I can contribute these stats from 1916. Compare what KJ has posted above from Paul Debono; I'm not sure if Debono's stats are against top black competition alone, or if it includes games against white semipros. I'm missing seven box scores that I'm aware of.

1916 Ben Taylor
Indianapolis ABCs

Batting
G-33 (team 33)
AB-117
H-38
D-2
T-0
HR-0
R-19
W-10
HP-2
SH-2
SB-4
AVE-.325 (NeL west .250)
OBA-.388 (NeL west .326)
SLG-.342 (NeL west .319)

For what it's worth, the ABC's home/away runs ratio (or raw pf) for the season was 116 (based on 29 home games, 11 road games). The contrast with Schorling was largely responsible for this.

Taylor didn't hit much for power in these games; his SLG is 8th among the ABCs regulars.

Fielding-1b
G-33
DI-265.3
PO-346
A-17
E-11
DP-15
FAVE-.971

I don't have positional averages for this year yet.

He also pitched in two games, one as a starter and one in relief, without a decision:
IP-7
H-8
HR-0
W-2
K-3
HB-1
TRA-6.43
   30. sunnyday2 Posted: March 30, 2005 at 01:09 PM (#1223666)
Excellent point, Kevin, there is reason to believe that we have overlooked Taylor. And I would think there are others whom we have to some degree overlooked for the same reason.

The reason being, with Taylor as an e.g. As Chris says in another thread, he (and before him Kevin, I think) developed MLEs for all seasons in which the NeLers played. It is up to each of us to decide how to interpret the data, especially in terms of longevity and in-season durability that they might have experienced among faster company.

I have always discounted these MLEs somewhat, but like everyone here I am becoming more and more comfortable with the MLEs. And so I probably discounted Taylor more steeply than the more recent players like Suttles, Wilson and Bell.

The other problem Taylor has is specific to Taylor and that is his almost complete lack of power as a 1B. On top of that is the relative dearth of data for his early years--but really my point is that we also have a dearth of data relative to context. How much power "should" he have as a deadball 1B in the NeLs.

All of that probably conspired to understate Taylor's value. But recognizing as much and OTOH doing something about it are two different things. While I am comfortable saying that Suttles goes to the head of the pack of corner OFs, I have it pretty firmly planted in my head that Taylor was no George Sisler nor Jake Beckley.

So in reconsidering Taylor, some comparison to ML 1B along with Suttles, Beckwith and Bell, would be helpful.

And again, who else have we overlooked from the early days when we were probably very uncomfortable with the MLEs? Carr, Rile, Hudspeth?
   31. Mark Shirk (jsch) Posted: March 30, 2005 at 02:03 PM (#1223708)
Is Taylor's OPS+, as stated above, derived from his I9's? If so then wouldn't a discount of 5-10% be in the offing? While I came into the project towwards the end of the i9's generation of NeL players, I remember conventional wisdom being that we needed to discount offensive numbers and upgrade pitching numbers. Am I correct?
   32. Michael Bass Posted: March 30, 2005 at 04:13 PM (#1223935)
And again, who else have we overlooked from the early days when we were probably very uncomfortable with the MLEs? Carr, Rile, Hudspeth?

Overlooked is the wrong word for the guy I'm about to suggest, but I believe a fresh look at Spot Poles would be a good idea at some point. I'd also suggest Bill Monroe, but I think he's more in the Frank Grant territory of most of his career being almost hopeless to translate.
   33. Gary A Posted: March 30, 2005 at 04:28 PM (#1223960)
Of Carr, Rile, and Hudspeth, I'd say Carr might be the one who is most underrated. Rile was good but had a very short career as a 1B (he was a pretty good pitcher), and Hudspeth seems okay but I've never seen any very impressive numbers for him. Edgar Wesley would be another 1B, but he probably played a good deal of his career for undocumented, pre-league clubs in Texas. What we've got for him in the 1920s NNL is good, though.

Of players who were mostly pre-league, deadball era, Candy Jim Taylor (3b) might be the most overlooked. He often played alongside his brother Ben and was up there with him as a hitter (in 1916 I've got him at 317/359/425). He had a very long career and hit .372 with 20 home runs (co-league lead with Heavy Johnson) in 1923 at the age of 37 (I think)--albeit for the St. Louis Stars (he hit 19 of 20 hrs in St. Louis!).
   34. Dr. Chaleeko Posted: March 30, 2005 at 05:47 PM (#1224065)
I agree, Poles is overlooked or underconsidered. I see him as right in the thick of it with Averill: a little better than Leach and Roush, not quite a nice as Duffy and GVH. All four are superior candidates to low-peak, long-career guys like Carey and Bell (IMO).

My mental image of Poles's value is a fusion of Roy Thomas's peak/prime value with Rich Ashburn's extended prime and career value. I'm not saying that Poles was that type of player, but those two just happened to together comprise the sort of value pattern I imagine Spots has.
   35. KJOK Posted: March 31, 2005 at 12:59 AM (#1224832)
Is Taylor's OPS+, as stated above, derived from his I9's? If so then wouldn't a discount of 5-10% be in the offing?

Good catch. I had already started doing my MLE's when Chris started applying the extra I9 discount, so to be consistent with the players I had already done I neve applied it to my subsequent MLE's, including Taylor, which means I'm comparing apples to oranges somewhat. I've stopped doing MLE's since Chris does them much better, so we just need to encourage Chris to attempt MLE's for Taylor so we can make a better comparison.
   36. KJOK Posted: January 17, 2006 at 11:18 PM (#1825925)
Some updated 1921 data from Gary A.'s research:

G-83
AB-316
H-125
2B-19
3B-6
HR-2
R-65
BB-25
HBP-4
SH-9
SB-9
Ave - .396 (League .270)
OBP - .446 (League .330)
SLG - .513 (League .368)

BATTING AVERAGE LEADERS 1921 NNL:
1. Charleston, SLG - .436
2. Blackwell, SLG - .430
3. Santop, HDC - .400
4. Taylor, ABC - .396
5. Beckwith, COG - .378
6. Lloyd, COB - .357
   37. KJOK Posted: January 17, 2006 at 11:27 PM (#1825945)
On Base Average Leaders 1921 NNL:
1. Charleston, SLG - .519
2. Blackwell, SLG - .504
3. Taylor, ABC - .446
4. Santop, HDC - .441
5. Bennette, ABC/COB - .428
6. Beckwith, COB -.425
7. Hill, DTS - .424

Slugging Average Leaders 1921 NNL:
1. Charleston, SLG - .728
2. Blackwell, SLG - .708
3. Santop, HDB - .636
4. Torriente, CAG - .582
5. Beckwith, COG - .576
6. Moore, KCM - .569
7. Wesley, DTS - .539
8. Lundy, Bag - .523
9. Carr, KCM - .522
10. McNair, KCM - .521
11. Taylor, ABC - .513
12. Hill, DTS - .507
   38. rawagman Posted: May 04, 2006 at 03:57 PM (#2003735)
Taylor is close to balloting for me. Anyone want to discuss? Advocate? Think I'm off-base?
   39. Mark Shirk (jsch) Posted: May 04, 2006 at 05:09 PM (#2003799)
I think you are off-base but that is just my opinion. Looking at the MLE's above Taylor looks a little like Jake Beckley. He has a little better peak but not as man average yeras tacked onto the end of his career. I have Taylor slightly ahead of Beckely but neither in my top 50. I think the HOF whiffed on this one unless they have some data that we don't.
   40. John (You Can Call Me Grandma) Murphy Posted: May 04, 2006 at 05:26 PM (#2003825)
I think Taylor may be overrated by some due to his pitching WS. Beckley, Chance, and Konetchy are superior, IMO.
   41. rawagman Posted: May 05, 2006 at 07:07 AM (#2004995)
Is defensive 1B work underrated? I can't remember which article, but in the recent Hardball Times Annual, it was pointed out that Albert Pujols may be the best defensive 1B in baseball as his numbers of bad-throw scoops (which largely goes unrecorded in the traditional fielding stats) was far and away higher than any of his peers. He had saved something like 20 extra throwing errors from the rest of his infield last year. You decide for yourselves what 20 saved errors is worth in runs and wins.

Now Ben Taylor was a noted specialist in digging out the dirtballs.

I have Ben Taylor as best 1B glove man in America prior to Frank McCormick. He had minimal power (think Mark Grace light) but superb BA (Sisler). His career was long. Very long. How long? Open to debate. At least 20 years, although blackbaseball.com credits his career as running from 1910-1940 (?!?)
I'll stick with 20 years.


Taylor 9091 138

That's a projection earlier provided by KJOK.
PA/OPS+
What did Beckley do?
Beckley 9526 125

I see Beckley as an 'A' defender. Taylor was an 'A+'.
Neither man really peaked. Just primed forever.

I think Taylor has convinced me to join my ballot as my #2 1B.
   42. favre Posted: May 05, 2006 at 09:41 AM (#2005044)
Based on Chris'old projections from i9s, the image of I have of Taylor (forgive me if someone has already made the comparison) is Keith Hernandez: outstanding glovework, very good line drive hitter, not a high peak.
   43. rawagman Posted: May 05, 2006 at 10:03 AM (#2005045)
He's Jake Beckley, only better
   44. Joey Numbaz (Scruff) Posted: May 05, 2006 at 10:15 AM (#2005046)
I think Taylor was the Grace/Hernandez to Beckley's Palmeiro. I'll take Raffy over either of the light power guys. And I'll take Beckley over Taylor, though they are similar in the long-career, only very good peak, never great.
   45. rawagman Posted: May 05, 2006 at 10:36 AM (#2005051)
from what I've been reading, Taylor was the greatest 1B glove in NeL history - where would he rank among ML 1B? That might be the key.
The ISO wasn't great for Taylor, but much of his prime years were in the deadball era.
Does anyone have a way to translate his numbers to some other context?
I don't have available park information. It seems he played everywhere.
   46. Thane of Bagarth Posted: May 05, 2006 at 03:09 PM (#2005213)
I think Taylor was the Grace/Hernandez to Beckley's Palmeiro. I'll take Raffy over either of the light power guys. And I'll take Beckley over Taylor, though they are similar in the long-career, only very good peak, never great.

I like the idea of Hernandez as a comp for Taylor, except I would argue Taylor had a few more productive seasons at the end of his career (Keith sort of fell off a cliff around age 35). If people are opposed to Taylor's election, Grace is the player who you'd want to use as a proxy--let's not underrate Keith Hernandez our attempts get a handle on the situation.

I know WARP3 isn't everybody's favorite stat, but just quickly looking at Palmeiro's and Hernandez's BP cards:

Top 5/7 WARP3 seasons (non-consecutive)
KH 50.2/67.6
RP 49.4/66.1

Career WARP3
KH 108.1
RP 132.9

WARP3 after age 35
KH 00.1
RP 26.3

If Taylor to Hernandez is a decent comp, considering longevity, I don't think Taylor to Palmeiro is too bad a comp either.
   47. Chris Cobb Posted: May 05, 2006 at 04:26 PM (#2005277)
Just to give everybody an eyeball look at Ben Taylor's offense, I did a quick back-of-the envelope MLE calculation for his 1921 season, for which Gary A. has provided very complete data, via KJOK, in post 36 above. I also used a park factor of 103, based on the park factors provided farther up the thread by Gary A.

Without having done all the nitty gritty calculations necessary to get the estimate of plate discipline exactly right and without doing any regression (not possible since I don't have surrounding seasons to create a baseline), Taylor's line, translated into the 1921 American League, would be around this:

BA/OBP/SLG -- OPS+
.384/.435/.469 -- 130

Given that this is an OBP-heavy OPS+, he probably has more value that a hitter with a similar OPS+ but less OBP.

This is an excellent offensive season. If Taylor's offense was no better than this, however, then he didn't hit enough to distinguish himself from the Jake Dauberts and Joe Judges of the first base world, who are solid members of the Hall of Very Good.

If he has a number of seasons up in the 140-150 range, then he might be comparable to Keith Hernandez. He might also be comparable to the nearly forgotten Ed Konetchy, who has totally dropped out of our discussions but has (I mention in passing) 10 more career WARP1 than George Sisler and was universally regarded as a great defensive first baseman, an assessment backed up by both WARP and win shares.

The data I have on Taylor from the Macmillan encyclopedia indicate that 1921 was one of Ben Taylor's best offensive seasons. Certainly he did not have another season of this calibre after 1921. He was, however, 33 that year, so we should expect this to be towards the end of his offensive prime, which would have run from circa 1915 to 1921. Unfortunately, there isn't much data available from those seasons. He doesn't look great in the 1916 data Gary A has provided above. I'll try to go over the data in Holway from the later teens and see what it suggests.

But if we are going to treat Ben Taylor as a serious candidate as a very good hitting, great-defense first baseman with a long career, then Ed Konetchy absolutely deserves another look, also.
   48. Chris Cobb Posted: May 05, 2006 at 04:28 PM (#2005279)
Clarification: The data on Taylor from Macmillan really begin in 1920, so it doesn't offer any view at all of Taylor's earlier career. 1921 was his best season from 1920 onward.
   49. Dr. Chaleeko Posted: May 05, 2006 at 08:29 PM (#2005501)
The trouble with Taylor for me is that we'll likely not get a great handle on his offense because the most important growth/peak years of his career likely occured in a loosely or un-documented era. Unless we can cull together enough credible 1920s/1930s information to create a backward-looking guess at what his age 25-33 seasons looked like, we'll always have this problem.
   50. Gary A Posted: May 06, 2006 at 06:56 PM (#2007658)
Updated Ben Taylor, 1916
Indianapolis ABCs
(I've got box scores for 40 of 43 known games)

G-40 (team 40)
AB-160
H-46
D-3
T-1
HR-0
R-24
W-9
HP-1
SH-3
SB-6
AVE-.313 (NeL west .247)
OBA-.357 (NeL west .320)
SLG-.347 (NeL west .322)

Home/road runs scored ratio for the ABCs' home park (Federal League Park) was 118, for 31 games at home, 12 away. The road parks they played in were: Schorling Park (6 games), Redland Field, Cincinnati (3), a park in Terre Haute IN (2 games), and a park in Peru IN (1 game).

ABCs games I counted, btw, include 16 games vs. the Cuban Stars; 10 vs. the Chicago American Giants; 5 vs. the Lincoln Stars; 4 vs. the All-Nations Club; 2 vs. Chicago Leland Giants; 2 vs. Chicago Union Giants; and 1 vs. St. Louis Giants.

The best hitter in the west that season was probably Cristobal Torriente, who hit .346/.431/.447 for the Cuban Stars and All-Nations, though Lloyd (.325/.388/.403) and Hill (.304/.400/.393) were good, too, considering the Schorling effect.

George Shively had the best hitting stats on the ABCs (.336/.394/.397). He also led the team with 32 runs scored, and led in the west with 19 stolen bases.
   51. Chris Cobb Posted: May 06, 2006 at 07:06 PM (#2007677)
Gary A.,

Thank you for the expanded 1916 data set for Ben Taylor! I note, however, two inconsistencies with the 1916 data you posted earlier.

In post 29, you list Taylor as having 10 walks in 117 at bats for 1916, but in post 50 you list him with 9 in 160. His HP also go down from 2 in post 29 to 1 in post 50.

Are you correcting earlier errors here, or have a couple of errors entered into your expanded counts?
   52. Gary A Posted: May 06, 2006 at 07:33 PM (#2007758)
I think it could be because I decided not to count three games against the Montgomery Gray Sox that were included in the earlier count. It's always hard to figure out where to draw the line, especially in the pre-league era. The Gray Sox just didn't have enough well-known players--only McKinley "Bunny" Downs had much of a career elsewhere, and more than half the players don't have known first names. It was a mistake to include them in the first place.

The ABCs won all three games, of course. In those three games Taylor went 4 for 9 with 2 walks and a hit by pitch.
   53. Chris Cobb Posted: May 06, 2006 at 10:06 PM (#2007941)
Thanks Gary!

One other question -- for the league BA/OBP/SLG rates: do those include or exclude pitchers? I don't remember what your standard practice is.

(In case it isn't obvious, I'm gearing up to do MLEs for Taylor for 1916 and for 1914, for which there's pretty complete data in Holway. Not having done MLEs for and pre-1920 seasons before, I have to start from scratch in the spreadsheet . . . )
   54. Gary A Posted: May 07, 2006 at 02:10 AM (#2008209)
No, I haven't usually in the past subtracted pitchers from league totals. But I can easily; I just did it for the 1916 western teams. So here are the "league" totals, minus pitchers (with totals including pitchers in parentheses):

AB-6119 (6915)
H-1576 (1709)
D-231 (258)
T-79 (84)
HR-29 (30)
R-903 (970)
W-610 (681)
HP-57 (63)
SH-143 (166)
SB-241 (248)
AVE-.258 (.247)
OBA-.331 (.320)
SLG-.335 (.322)

(Pitchers defined as anyone whose total games pitched amount to half or more of total games played.)

Actually, within a few days I'm going to try to post the complete stats I have so far for 1916 in Excel files at my blog.
   55. Gary A Posted: May 07, 2006 at 02:16 AM (#2008216)
Also, if you need this: I have a total of 107 western games with box scores. 970 runs were scored in these games, for an average of 4.53 per team-game. Counting games without box scores, I have 132 western games, in which a total of 1142 runs were scored (4.33 runs per team-game).

For games with box scores, the runs per nine innings is 4.77 (970 runs in 1829 innings).

Good luck with pre-1920 MLEs!
   56. Gary A Posted: May 10, 2006 at 03:56 AM (#2012379)
I've posted 1916 standings and batting files here; tomorrow I'll put up fielding and pitching stats.
   57. Joey Numbaz (Scruff) Posted: October 16, 2006 at 08:09 PM (#2214176)
I think we should maybe take another look at Taylor.

He was 32 in 1920, meaning a good chunk of his career was in the deadball era, where there's a case that 1B defense was more important than in the post-Ruth game.

Also, he played in an era where batting average was important, and he hit for a high batting average.

Finally, the Negro League Committee looked everything over, and concluded he belongs - to me that counts for something as well.
   58. Willie Mays Hayes Posted: October 16, 2006 at 08:29 PM (#2214211)
Having re-read this thread, I'm in full agreement with Joe. Ben (and Beckley, as a consequence) will get a much closer look for the next ballot by me.
   59. rawagman Posted: October 16, 2006 at 09:00 PM (#2214276)
I already have Ben Taylor in my PHOM and ranked as my #1 eligible 1B. 2nd on this year's ballot.
Here's the why:
Defense - With Negro League players, we have to go by word of mouth. Word of mouth on Taylor is that he was the Negro League's all time greatest defending first baseman. Every source I've read on him comments on that. He also played at a time when 1B defense was seemingly more important that it is today. More bunt-heavy (at least in white ball) than in Beckley's time, and until age 32, he played with dead balls.
Hitting - I have his career MLE OPS+ as 138 in 9091 PA's according to KJOK. I want to recheck this, but my spread sheets also point out he had 7 seasons of OPS+ above 120. Which doesn't say much of anything. How does he compare to near contemporary 1B, George Sisler?
Career - Some of us have said that the HOF's Negro League commision may have missed out on Redding cause their search didn;t cover his career. Was that not the same time as Taylor's? They sure got his case, though.Most accounts point to Taylor having played until 1940 - 30 years! That's a long career. What did he do during WWI? Some pitching credit is also accountable.
Seems like he collected some ink as well.
I see Taylor as being everything Jake Beckley is purported to have been. How far apart do you have them? Taylor seemed to stick out more among his peers, with better defense, a better bat and almost as much career (only behind Beckley and Vernon among eligible and interesting 1B at this point). What's not to like?
   60. DL from MN Posted: July 10, 2007 at 05:42 PM (#2435786)
We've elected Keith Hernandez who was a OBP heavy 129 OPS+ 1B with 8550 plate appearances regarded as the best fielding 1B of his era. Ben Taylor is projected as an OBP heavy 130 OPS+ 1B with 9000 plate appearances regarded as the best fielding 1B of his era. Taylor is not close to election and he has some pitching credit and some upside in his projection.
   61. Paul Wendt Posted: July 10, 2007 at 05:59 PM (#2435813)
Hitting - I have his career MLE OPS+ as 138 in 9091 PA's according to KJOK. I want to recheck this, but my spread sheets also point out he had 7 seasons of OPS+ above 120. Which doesn't say much of anything.

Offhand, it says to me that 9000 PA at 138+ is "impossible".
   62. Paul Wendt Posted: November 06, 2010 at 09:57 PM (#3685109)
Regarding Ben Taylor see "First Basemen on the opening ballot" #143

Joe or John:
I am writing here because comment is not permitted at the Negro Leagues directory or at /files/hall_of_merit which now displays the August 29, 2005 edition of "... Important Links" at the top.
> Monday, August 29, 2005
> The Baseball Hall of Merit’s Important Links
I think that is where I am accustomed to find the list of important links.

In that list of important links, the Negro Leagues directory has this unhelpful linkname.
> "Links to discussions of ...".
   63. Paul Wendt Posted: November 06, 2010 at 09:59 PM (#3685110)
> /files/hall_of_merit
Now my browser displays this address.
http://www.baseballthinkfactory.org/files/hall_of_merit/newsblog/
   64. Joey Numbaz (Scruff) Posted: November 08, 2010 at 05:47 AM (#3685710)
Paul, if you click 'read more' below that, you get the whole page. The links to discussion of . . . is just where the page hits the front page character limit.

I realize that could be better.
   65. KJOK Posted: September 17, 2011 at 07:32 AM (#3927875)
   66. KJOK Posted: September 17, 2011 at 08:29 PM (#3928204)
CORRECTED LINK:

Ben Taylor's Real Stats
   67. KJOK Posted: January 14, 2012 at 03:33 PM (#4037108)
1914 and 1915 Blackball stats have been added to the Negro Leagues database at seamheads.com.

These are two prime years for Taylor - I think he wins the "MVP Trophy" for 1914....

   68. KJOK Posted: October 17, 2012 at 09:11 PM (#4274898)
A free issue of the Outsider Baseball Bulletin, focusing on 'pre-league' Negro League players, can be downloaded at:

Seamheads
   69. theorioleway Posted: October 18, 2012 at 07:40 PM (#4275989)
KJOK: Great article! If it's not too late, you might want to change the 2B chart, as it shows their OPS+ for WAR/Sn. Also, besides Taylor and Moran, anyone else from this article you would rep for HOM? I was interested that you didn't include Castillo on your chart/wrapup at the end.
   70. KJOK Posted: October 22, 2012 at 02:12 AM (#4278702)
Thanks - It is too late - that escaped the editing process somehow.
Castillo wasn't included mainly due to being a Cuban League player instead of a Negro Leagues player, so he's not going to Cooperstown.

As I indicated in the article, there are a few other players that would merit consideration besides Taylor and Moran (I think most of the others are already in the HOM).
   71. bjhanke Posted: October 22, 2012 at 03:00 AM (#4278706)
Evaluating dead-ball era defense at first base is 1) very important, and 2) ridiculously hard. The reason it's important is because DBE hitters bunted a LOT, so defense at 1B was probably at an all-time value high. The essential problem that makes it hard is analogous to what I call the "Johnny Bench Problem" in the New Historical Abstract. The JBP is, essentially, this: Bill's system doesn't give Bench an A+ ranking at catcher because he didn't throw out that many baserunners. This is not because his percentage of runners thrown out is bad, it's because no one would run on him after they'd seen his arm, except for people like Brock and Wills, who could run on anybody (and the occasional "attempt" by someone like Willie Stargell, who was almost certainly not trying to steal a base, but was on the front end of a hit and run where the batter swung and missed). That is, Bench is caught being evaluated by a stat that the OFFENSE has a lot of control over. At the extreme, the offense can ALWAYS reduce the value of a catcher's arm to ZERO by simply never trying to steal at all. Even Bench didn't get that treatment, but his stats are clearly queered by the reaction of runners to his arm.

The problem for DBE 1B is not throwing runners out; it's the bunts. Like attempting a SB, a hitter can choose whether to bunt or not. Therefore, it is very likely that the best DBE 1B don't have huge numbers of assists and putouts because no one would bunt on them. This may be why Sisler and Chase rank so low in Bill's system, although there are caveats for both men. Chase may have reduced his defensive stats in the process of selling ballgames, and Sisler was not the same player after the 1922 injury that he had been before. I've never seen a breakout of yearly Defensive Win Shares for Sisler, and would very much like to.

And the problem is even worse than this. If you're a DBE hitter, and want to bunt, you can always choose to bunt at the third baseman instead of the 1B. This may be why George Kelly ranks so much higher than Sisler and Chase, although Chase had a godlike defensive rep, and Sisler's rep is second only to Hal's. Both Chase and Sisler played, mostly, for bad teams, with poor 3B. Kelly played mostly on very good teams with good 3B. Bunters against Sisler or Chase were more likely to bunt at the 3B than they would be against Kelly, for reasons that have nothing to do with any Kelly superiority at pouncing on bunts.

This also affects their hitting. A DBE 1B was expected to be able to pounce on bunts. "Pouncing on bunts" and "big strong guy who hits for power" are usually not descriptions of the same player, although there is always the occasional exception (Konetchy).

This is bad enough in the major leagues, where there are 154-game schedules and significant stats, but no record of how many "ground balls" were actually bunts, much less how many of them went towards first and how many towards third. Trying to evaluate Negro League DBE 1B defense is, very probably, impossible statistically. You're better off just going by the reputation, although you still have to keep in mind that the offense can always reduce the effect of a great defensive 1B by either not bunting, or only bunting towards third. Like I said, it's ridiculously hard. I, personally, would be very reluctant to evaluate Ben Taylor, or any of his contemporaries, by their defense, either statistically or even by reputation. - Brock Hanke
   72. theorioleway Posted: October 22, 2012 at 08:10 PM (#4279431)
Brock - FWIW, both TZ and DRA rate Sisler as good pre-injury and then indicate a major drop-off in defensive production. You can see his defensive win shares by season at thebaseballgauge.com. They go down slightly after the injury, but not dramatically. TZ and DRA are much less kind on Chase, although DRA is kinder. Assuming you mean Highpockets Kelly, he does rate well by all systems (although it doesn't seem that much higher by defensive win shares, but maybe I don't appreciate the scale), but he didn't really play in the dead-ball era. His main years as a starter were 1920-1929. Your point about the difficulty of judging defense from this era is still true, but it doesn't mean we shouldn't give it our best effort. I think the Ben Taylor supporters would also argue that Taylor was far from being all-glove. Seamheads/The Baseball Gauge have a record of 2,979 PA for Taylor at a .338/.400/.468 BA/OBP/SLG slashline, which they calculate equals a 161 OPS+. Now, that obviously doesn't mean he was the equivalent of Jimmie Foxx (career 163 OPS+), but I think it's fair to say the man could hit. As mentioned earlier in this thread, if you take 30 points of OPS+ off, you're still in Hernandez/Palmeiro territory.
   73. bjhanke Posted: October 23, 2012 at 01:20 AM (#4280353)
Oh, yes. I wasn't saying anything about Taylor as an overall candidate. I was only talking about defense, because what I really had to offer was the odd "Johnny Bench Problem" that can cause all sorts of difficulties with certain defensive stats, especially assist and putout stats at 1B and 3B. Taylor hasn't made a ballot of mine yet, but he's been close. I just use "he was good" as a defense estimator, because I'm pretty sure I can document that, and not at all sure that I can document anything more.

A lot depends on whether you're trying to use a "quality" system, like most WAR systems are, or a "value" system like Win Shares. If you're using a value system, then there's no way around the Johnny Bench Problem. Johnny Bench ends up as an A- catcher instead of an A+ because the runners that he did not throw out because they would not try to steal were real baserunners, some of whom came around to score without needing any stolen bases, resulting in real runs on the scoreboard working to help the Reds' opponents win real ballgames from them. And some of the people who did not bunt on Chase or Sisler reached base anyway, and some of those eventually scored, resulting in real runs, too. In a value system, there is absolutely no way around that, because everything is anchored to real wins. That's why the Johnny Bench problem frustrated Bill. Check out the Historical Abstract essay on Bench; it's pretty clear that Bill was frustrated because there was no way his system could give Bench the defensive quality grade he knew Bench deserved, because in a value system, Bench does not deserve that grade. He's too good at throwing out base stealers to have great value at it. Sounds wrong, but that is one of the holes in value systems. And Sisler and Chase may have been too good at pouncing on bunts to have great value at it.

In a quality system, you can try to adjust for things like this, but you have to be very careful. You can try to give Johnny Bench credit for his arm by adjusting the stolen base attempts against him to a league-normal number and then applying his real caught stealing percentage against that. But even that is dangerous, because of selection bias. The guys who really DID try to steal on Bench were, in general, the best base stealers around. Don Malcolm was just in town visiting, and we took a look at game logs from 1968, when Bench was very young. Stolen base attempts against Bench were dominated by people like Lou Brock, Maury Wills and Tony Taylor, which is the A list. Obviously, Bench would have had a better caught stealing percentage against other people's attempts. So, if you just adjust the number of attempts to league-normal, you are still not giving Bench's arm enough credit. EXCEPT that some of the "stolen base attempts" against Bench were nothing of the sort. They were hit-and-run plays where the batter swung at the ball and missed. Hence, the occasional "stolen base attempt" by guys like Willie Stargell. So Johnny Bench's caught stealing percentage is queered by those guys. This is a lot of why CS percentages in the 1950s were high; there were so few actual SB attempts that the data is seriously queered by the number of hit-and-run results. How many of what to count? Who can tell, with all that noise in the data? The same thing happens with DBE bunts. How to adjust for opportunity, when every bunt attempt can be made at the first baseman or the third baseman or not made at all? You can spend a lot of time trying to see what the assist data for 3B were like with and without the given 1B, and what the assist and putout data were for a given 1B with different 3B. But you are trying those adjustments against a lot of noise.

So, that's even more of the point I was trying to make. I didn't include all of that in the first post because it was long enough already. And none of it has anything to do with Ben Taylor's offense, or with anyone else's offense, for that matter. It's a defense problem that is important in the dead ball era simply because the era has a boatload of bunt attempts in it, and they were not separated from the non-bunt grounders by the scorers of the time, and there's no film to go back and check out.

Now, I gotta go check thebaseballgauge.com. THANKS for the reference! - Brock
   74. DL from MN Posted: October 23, 2012 at 10:20 AM (#4280451)
Can you give credit to Bench for "stolen bases not attempted"? There is value in keeping runners at 1B instead of 2B. Take the league average SB attempt rate and credit Bench for the attempts not made. Obviously this method doesn't work well for all players but it might help with the outliers.

On Ben Taylor - there is deadball and there is Negro League deadball which I imagine are not exactly the same thing. The Negro Leagues played a more deadball style of baseball (bunts, stolen bases, etc) much longer than the American League. I rate Taylor as "very good" on defense and am still bothered by the inconsistency of electing Beckley, Hernandez and Palmeiro but not Ben Taylor.
   75. Mike Emeigh Posted: October 23, 2012 at 12:41 PM (#4280607)
And Sisler and Chase may have been too good at pouncing on bunts to have great value at it.


That's not quite the same thing. The bunt during the Deadball Era wasn't an elective play in the sense that the stolen base was when Bench was catching, because most teams were carrying three or four offensive black holes in the lineup. Teams that bunted less frequently against a Sisler or Chase just because of their fielding prowess would have been shooting themselves in the foot.

-- MWE
   76. Mike Emeigh Posted: October 23, 2012 at 02:54 PM (#4280738)
So I ran some numbers for Bench, and it's very true that he was preventing stolen base attempts.

Bench caught over 14,000 innings for the Reds between 1968 and 1983 (I excluded his partial 1967 season). There were 37 catchers during this period who caught at least 5000 innings. Among them, Bench ranks fourth in CS% on steals of second (including catcher pickoffs but excluding pitcher pickoffs) behind Thurman Munson, Lance Parrish, and Rick Dempsey, but is first by a good margin in the number of runners on first base per steal attempt. Reds' opponents attempted one steal for every 20.9 PA with a runner on first while Bench was catching. The #2 man, Manny Sanguillen, was at one for every 19.65 during this same time frame - the distance between Bench and Sangy is larger than the distance between Manny and the #5 guy (Jerry Grote).

Odd note: Jim Sundberg, who had a very good CS percentage (40.5%) was near the bottom of this list in terms of PA per steal attempt. So was Gary Carter (40.8%). That might be due to extenuating circumstances (Rickey in Sundberg's case, the Lumber and Lightning boys in Carter's).

-- MWE
   77. DL from MN Posted: October 23, 2012 at 03:58 PM (#4280808)
So - what is that worth assuming league average SB success rates? How many runs did he save versus an average catcher due to the other team avoiding the running game?
   78. Mike Emeigh Posted: October 23, 2012 at 05:26 PM (#4280976)
So - what is that worth assuming league average SB success rates?


Based on runners on 1B against the Reds (which was a below-average number - the Reds did have pretty good pitching despite the public perception), I estimate about 431 more SB, 32 more CS at average rates of attempts and CS, so that comes out to be something like, what, 75 runs over the course of Bench's career, or about 4-5 runs a year. Does that seem about right?

-- MWE

   79. bjhanke Posted: October 23, 2012 at 07:11 PM (#4281071)
Mike - Um, Mike, I think something has gone horribly wrong in your post above. 431 SB against only 32 CS would be a historically unique rate of success, several standard deviations above norm. Just to check, here is the entire NL SB effort for 1975: 1176 SB, 558 CS. That's a ratio of about 2-1, which 1) is normal, most seasons have league SB% at about 2-1, and 2) is right about at the break-even point, where the negative value of the CS is offset by the positive gain on the SBs. It actually looks like baseball players and managers, just working off of what seems to provide extra runs, keep attempting SBs until they reach the break-even point of CS, and then they stop. Ballplayers aren't dumb. They know about when to stop using a strategy. So, in answer to DL's question, yes, IF you're using a quality system rather than a value system, you can try to evaluate prevented SBs, but you're going to find that you get almost no advantage out of it, if you assume league average caught stealing rates, which are about 2-1. Mike's 431/32 is about 13.5 to 1. There has never been a season where people stole bases at that rate. Nothing even close. Since Mike can do the math, I just assume a typo here. Maybe the caught stealing number is actually 232 or something. But I repeat, the actual SB%, over the last 60 or so years, has been right about at the break-even point, so you get no real gain for adding volume at that percentage.

If you use a value system, you cannot do that at all. It would amount to adding runs prevented to Johnny Bench's record without deducting them from someone else, which you have to do to get the actual contributions to total up to the team effort.

Also for Mike Emeigh (#75) - I wrote a lot of words, so I can't blame you for missing this, but I did make mention that you can always bunt at the 3B, if your offensive game is based largely on bunts, like Roy Thomas or someone. So refusing to bunt at Chase or Sisler would not be suicidal. It would just switch the direction of the bunt. This would make Sisler and Chase look worse, and their third basemen look better, but it would not crash the entire concept of bunting.

I hasten to add that Mike is, in general, very good about getting things right, so I want to apologize to him for not being clear. That is, I'm not picking on Mike, he doesn't deserve it. But he missed the 3B bunt thing, and his estimate of SB/CS ratios is wildly off. Mike just had to be having a bad day. His work always commands respect, which is why I immediately double-checked. - Brock
   80. Brent Posted: December 14, 2014 at 11:29 PM (#4861849)
At this point, Seamheads provides nearly complete coverage of Ben Taylor’s baseball career. I believe that only his 1925, 1927, and 1929 seasons aren’t available yet in the Seamheads dataset. Thus, we have a much more reliable database to evaluate his career than was available several years ago, when most of the analysis on this thread was done. Despite the more comprehensive data, many HoM voters appear to be confused about how to interpret the data.

The statistics currently shown on Seamheads show 3,450 plate appearances against NeLg and CuLg competition and a batting line of .335/.397/.461. His OPS+ is 158. How do we convert that record to an MLE against major league-level competition?

I’ll demonstrate a few easy calculations that you can apply to any NeLg player in the Seamheads database. These formulas are designed to convert a career OPS+ against NeLg competition to a major league equivalent OPS+ against MLB-level competition. These formulas, however, do NOT provide a full set of MLEs. To calculate MLEs, you need to adjust for both quality of competition and for differences in run environment. If you’re calculating seasonal MLEs, you also have to decide how much to regress the seasonal statistics. But if you just want to convert a player’s OPS+ against NeLg competition to an MLB-equivalent level of competition, it’s possible to skip all the detailed calculations required for seasonal MLEs. The following calculations are based on the methods I posted, for example, on the John Henry Lloyd thread a few years ago, and follow from the methods first developed a number of years ago by Chris Cobb.

In addition to the BA/OBP/SLG and OPS+ statistics, I’m going to need to calculate and adjust a few other statistics. In particular:

H* = H/(AB+BB+HBP) (= .304 for Ben Taylor)

W* = (BB+HBP)/(AB+BB+HBP) (= .094 for Taylor)

EB* = (2B+2*3B+3*HR)/(AB+BB+HBP) (=.114 for Taylor)

To adjust for quality of play, I’ll multiply H* by 0.9 and multiply W* and EB* by 0.9^2 = 0.81, based on some research done by Chris Cobb. These give

Adj-H* = .273

Adj-W* = .076

Adj-EB* = .092

It’s now easy to calculate Taylor’s adjusted OBP as .273+.076 = .349. To calculate his adjusted SLG, we need to change the denominator of his hit and extra-base rates to AB’s instead of (AB+BB+HBP), so I divide Adj-H* and Adj-EB* by (1-Adj-W*)=(1-.076). That calculation results in an adjusted batting average of .296 and an adjusted isolated power of .100. Adding those two numbers, we get an adjusted slugging percentage of .396. These values represent the on-base percentage and slugging percentage we would expect Ben Taylor to have had against MLB-level competition, but playing in the same run environment that he faced in the NeLgs.

The final step is to compare these numbers against a “league average” baseline. Remember, the relevant league average here is for the NeLgs, since we haven’t made any conversions for run environment. I’m going to cheat a little bit here—we know that the sum of his actual OBP relative to the league and his actual SLG relative to the league was 58 percent better than the league (because that’s what’s meant by an OPS+ of 158). I’m just going to assume that each (OBP and SLG) was 29% better than the league, implying that the league average would have been .308 for OBP and .357 for SLG. Now I know that these numbers aren’t quite right, but I also know that it’s going to take me quite a bit of time to do those calculations correctly. But I also tried doing the calculations with several alternative assumptions that also add up to 58 (for example, assuming his OBP was 16% better than the league and his SLG was 42% better than the league). The interesting thing is that I got almost the exact same result no matter what I assumed. I think the reason for that is that the adjustment being applied to OBP is almost the same as the adjustment applied to SLG, so the exact baseline for each doesn’t make much difference. It seems safe to just use an arbitrary assumption that splits it 50-50.

Anyway, to calculate Taylor’s MLE OPS+, I just compare his adjusted OBP (.349) to the assumed league value (.308) and his adjusted SLG (.396) to the assumed league value. Using the standard formula of OPS+ = (OBP/lgOBP + SLG/lgSLG -1)*100, I get an MLE OPS+ of 124. (As mentioned, I get the same value of 124 with different assumptions about lgOBP and lgSLG that are consistent with his career OPS+ of 158.) That MLE value of 124 suggests that Ben Taylor’s offensive value may have been roughly in the same range as Ed Konetchy.

Now, I’m going to post something on the Cristobal Torriente thread suggesting that maybe the adjustment factor should be a little higher. For example, 0.92 for hits and 0.92^2 = 0.85 for walks and extra bases may be a plausible adjustment factor. Using those values, and applying the same calculations, I get an MLE OPS+ of 131, which is better, but still below other first base candidates like Fred McGriff.
   81. DL from MN Posted: December 15, 2014 at 11:04 AM (#4862026)
That OPS+ calculation does not seem to be at odds with the assessment from 2007:

"Ben Taylor is projected as an OBP heavy 130 OPS+ 1B with 9000 plate appearances regarded as the best fielding 1B of his era."

It does look like that might be on the optimistic side but still plausible.

I think Ed Konetchy is a great pick for a "floor" for Ben Taylor's level of production although Konetchy had a shorter career so Ben Taylor would still be better.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
rr
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

Syndicate

Page rendered in 0.8108 seconds
43 querie(s) executed