Baseball for the Thinking Fan

Login | Register | Feedback

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

Sunday, November 07, 2004

Nip Winters

Here’s one guy that a vigorous discussion of his credentials is needed.

John (You Can Call Me Grandma) Murphy Posted: November 07, 2004 at 02:20 AM | 18 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: November 08, 2004 at 03:27 PM (#957784)
Nip Winters

Expert Assessments

Received 20% of expert vote for HoF in _Cool Papas_
6-time Holway all-star
5.5 times listed as best pitcher in his league by James or Holway
Ranked by Holway as #3 left-handed Negro-League pitcher of all time (I think -- don't have the book with me at work and forgot to double-check this one)

Here's the data I've gathered on Nip Winters from Holway.

Nip Winters

Negro-League Record
1920 no data, pitched for Baltimore Black Sox
1921 3-3 for AC Bach Giants, team 34-28; 0-1, 4.50 TRA in playoff vs. Hilldale; 1-0-1, 0.50 TRA in World Series vs. Chicago Am Giants
1922 7-10 for AC Bacharach Giants, team 20-24; 1-0 in World Series vs. Chi Am Giants
1923 10-3, TRA 3.03 for Phi Hilldales, team 32-17; GSA, All-Star
1924 27-4 for Phi Hilldales, team 58-23; GSA, All-star; 3-1, 1.16 ERA in WS vs. Monarchs
1925 21-13, TRA 3.88 for Phi Hilldales, team 65-26, all-star; 1-0, 3.00 TRA? In WS vs. Monarchs
1926 23-4, 2.23 TRA for Phi Hilldales, team 58-34; GSA, all-star
1927 18-16, 2.96 TRA for Phi Hilldales, team 47-70; all-star
1928 10-11 for Phi Hilldales & NY Lincoln Giants, teams 26-27, 17-21
1929 2-4 for NY Lincoln Giants, team 38-18
1930 No data, may not have pitched for a top team
1931 1-2 for Phi/Hilldale Stars, team 44-16
1-1 for Nwk Browns, team 3-9
0-2 for NY Bacharach Giants, team 0-4
pitched for Was Potomacs, no record
1932 1-2 for Was Pilots, team 16-35
1933 no data, pitched for AC Bacharach Giants, who were not in league

124-75, .623
Holway lists 126-74, .630

Teams 405-309, .567
Teams without Winters, 281-234, .546
Record adj. to .500 teams behind him 115-84, .577

Vs. Major-League competition
1923 2-1 vs. Phil Athletics
1926 2-0 vs. All-Star team (a good one), lp Lefty Grove, Rube Walberg

Brief assessment--Winters was undoubtedly the best pitcher in the Eastern Colored League during his 5-year peak, 1923-27, and either he or Joe Rogan was the best pitcher in Negro baseball during this time. Outside this peak, however, he has little value. Is the peak enough to make him a serious candidate, or no?
   2. John (You Can Call Me Grandma) Murphy Posted: November 08, 2004 at 04:34 PM (#957893)
Chris, in your judgment, where would Winters fall among the best major league pitchers between 1920 and 1933? Anyone who Nip resembles? How does his peak matchup with a Vance, Luque or any of the other high peak guys? Thanks!
   3. Dr. Chaleeko Posted: November 08, 2004 at 05:17 PM (#957953)
Chris, to tag onto Murph's question, how do you see Winters in comparison to Mendez and Redding?
   4. Michael Bass Posted: November 08, 2004 at 05:41 PM (#958001)
Any chance of Win Shares estimates on him and Johnson? Sounds to me like he goes below Redding and Mendez, but I'd like to see more before deciding for sure. I have no clue whatsoever on Johnson.
   5. Chris Cobb Posted: November 08, 2004 at 05:44 PM (#958009)
Preliminary comparisons only -- I want other people to have a go at the numbers before I say too much.

I suspect that Nip's closest comp is a picher who peaked just after 1932 -- Dizzy Dean.

His 1924 and 1926 seasons are pretty close to Vance's 1924 and 1928 seasons and Luque's 1923 season, but he doesn't have as many very good seasons to back up his peak seasons as Vance and Luque do.

Re Mendez and Redding. Better peak than Redding, but not nearly as much career. Peak very similar to Mendez -- maybe slightly higher, maybe a year shorter, but he lacks Mendez's outside-peak achievements.
   6. Dr. Chaleeko Posted: November 08, 2004 at 06:56 PM (#958142)
Chris (or anyone!), are timelining issues more significant in the Negro Leagues than in the majors? In the 1920s NgLs, was the quality of play--relative to the previous decades--rising more quickly than it was in MLB during the same period?

Also, re Winters himself, Riley repeatedly mentions that Winters had control problems and describes him as "wild" pretty often. In another thread (don't recall which, might have been the 1937 discussion thread), someone posted league walk rates, and they were much lower for the NgLs than for MLB. So if Winters was indeed as wild as Riley suggests, could Nip have pitched successfully in a league which took more walks than the NNL and ECL probably did?
   7. Chris Cobb Posted: November 08, 2004 at 09:30 PM (#958325)
OK. Here's a season-by-season projection of innings, support-neutral ERA+, support-neutral w-l, and win shares, for Nip Winters. All of this is derived from Holway's data by my methods, since there are no i9s numbers to rely on for help. I believe I have been more conservative in my IP projections than they have been, but less conservative in ERA projections. The main features of the method are explained below.

Any suggestions on improving the methods are welcome. This is my first try at a complete career projection for a Negro-League pitcher, so I have no other data points to compare it to yet. As we look at Joe Rogan, William Bell, Andy Cooper, Bill Foster, and others, I'll be preparing similar projections, and we'll see over time if they look too high or too low, but your initial criticisms will be valuable as well.

USE. Was Winters a #1 starter, #2 starter, etc. or a non-starting pitcher, by major-league quevalence? This designation key to estimating his IP. Each year about 30 pitchers in Negro-Leagues would be at the same level of quality as the 80 starters in the majors. By proportion, there would be about 6 #1, 6 #2, 6 #3, 6 #4, and 6 #5 starters. I estimate a Negro-League pitcher’s innings from the pitcher's usage group and his place within that group, comparing the Negro-League #1 pitchers to the Major-League #1 pitchers for that season, and so on. I found the high, low, and mean ip for each of these usage positions on major-league staffs for each relevant season. So, to make these usage estimates, I’ve picked out, roughly, the top 30 pitchers in the Negro Leagues during each year of Winters’ career and put them in rough rank order.

ERA+. This is derived from Winters’ wins above team. I calculated that for each of his seasons, adjusted his support to .500, calculated his support-neutral winning percentage, and found the ERA+ that would lead to such a record, according to the Pythagorean method. If the team’s fielding seemed especially good or bad, I adjusted Winters’ Negro-League ERA+ downward or upward by 10 points accordingly. I then prorated Winters’ NeL ERA+ by .87 to get an MLE ERA+, which is listed below. From this ERA+, I derived a support-neutral MLE wining percentage, which, multiplied by innings and divided by 9, produced the projected won-lost record. Note that I used IP/9 to get decisions, which is conservative. A real pitcher collecting real decisions would acquire them at a rate of 1 per 8.5 – 9 innings, so check the real pitchers IP/dec when comparing Winters’ projected support-neutral w-l record to real w-l records.

WS. These are calculated directly from Winters' W-L record by the basic WS formulas. I start with the number of wins the pitcher would have if his record were .500. W * 3 * .52 * .67 yields an average pitcher's share of these wins. The pitcher gains 3 win shares for each win above average and loses 3 win shares for each win below average. This system projects higher highs and lower lows than the fully nuanced system, but they balance out over a career.

Nip Winters’ projected MLE record, including innings pitched, support-neutral won-lost record, and ERA+

Year Team      USE  IP   sn W-L   MLE ERA+   WS (incl. 9 bws)
1921 Atl City  nSP  104  5-8       79        1
1922 Atl City  #3   196  10-13     90        9
1923 Hilldale  #1   260  21-9     150        30
1924 Hilldale  #1   323  25-11    148        41
1925 Hilldale  #1   270  15-15    100        17
1926 Hilldale  #1   284  24-7     173        42
1927 Hilldale  #1   306  22-12    138        35
1928NY        #2   234  14-12    109        19
1929NY/Bal    #5    90  3-7       62         0
1930not in NeL     -- no MLE credit --
1931 NY/Hil/Nwk nSP  63  3-4       86         2
1932Was, DC       -- no MLE credit --
1933 Was, DC        -- no MLE credit --
career             2140 142-98    121       196

Conclusion: Very similar to Dizzy Dean, but without the folk-hero status and the broadcasting career.

Because usage patterns were different for NeL and ML pitchers, these projections may underestimate value of Negro-League pitchers. At their peak, they threw many _more_ innings than they would have thrown in the majors, which may have shortened their peaks. By translating their innings to major-league usage patterns, I have reduced their in-season values without giving them a corresponding boost to career length as a result of escaping extreme overuse. I don't know whether such a career boost would be appropriate or not. Discussion of that subject would be very welcome!
   8. Dr. Chaleeko Posted: November 08, 2004 at 10:17 PM (#958389)
Chris, that's a wicked cool system you've got there. I think there's a range of pitchers you could compare Winters to, based on your numbers. Dean is one, though his ERA+ was substantially higher (10 points higher, though in fewer innings), so he's probably the top end of possible comparisons, along with Koufax (if you want a more modern take).

Lefty Gomez strikes me as an apt comparison: 125 ERA+, 2500 innings.

Another 1930s lefty: Thorton Lee was a 119 ERA+ in 2300 innings.

One more modern guy. Ron Guidry: 120 ERA+ in 2400 innings.

So, you have a lefty whose career fits somewhere among Lee, Guidry, Gomez, Koufax with a value pattern like Sandy and Dizzy, but ERA+ like the Lee-Guidry-Gomez trio. Is it enough?
   9. Brent Posted: November 13, 2004 at 06:06 PM (#964777)
The WS estimates that Chris Cobb came up with for Winters seemed a bit high to me, as I think he may have implied when he said:

This system projects higher highs and lower lows than the fully nuanced system, but they balance out over a career.

As an alternative, I tried taking the formula "short-form" WS from p. 15 of the Win Shares book, which for pitching WS is:

"6. For pitchers as pitchers, multiply the league ERA by 1.50 and subtract 1.00.

7. Figure how many earned runs the pitcher would have allowed had that been his ERA.

8. Subtract his actual earned runs allowed.

9. Add his Saves.

10. Divide by three. The result is his pitching Win Shares (again, not less than zero)."

For Winters, since Chris is working with estimated ERA+ rather than actual ERA, I assume a league average ERA of 4.50. Then, I believe the formula for short-form PWS (ignoring saves) works out to be:

PWS = IP * [5.75/9 - (450/ERA+)/9] / 3.

After adding 9 batting WS over his career, I get the following year-by-year WS estimates for Winters:

1921 1
1922 6
1923 28
1924 34
1925 14
1926 34
1927 30
1928 15
1929 0
1931 1

career 163

These values seem to match up better with the WS values for ML pitchers having comparable seasons. For example, Winters' 1926 projected season, with 284 innings and an ERA+ of 173, is similar to Vance 1924, with 308.3 IP and 173 ERA+; Vance was credited with 36 WS for that season. Winters' 1924 projection, 323 IP and 148 ERA+, looks similar to Faber's 1922 season, 352 IP and 145 ERA+; Faber had 31 WS.

I also note that Chris's hypothesis that the highs and lows balance out over the career doesn't seem to hold up in this case.
   10. Howie Menckel Posted: November 13, 2004 at 06:57 PM (#964802)
accordin' to

"Winters was one of the Negro Leagues' best lefthanded pitchers. The ace curveballer starred in the first modern Black World Series in 1924 as a member of the Hilldale club. Backed by Hall of Famers Judy Johnson and Pop Lloyd and stalwarts Frank Warfield and Biz Mackey, Winters went 3-1 (1.16) versus the Kansas City Monarchs. He led the Eastern Colored League with marks of 26-4 in 1924, 21-10 in 1925, and 18-6 in 1926. In barnstorming games against Babe Ruth's All Stars, Winters split two decisions with Lefty Grove.

» October 20, 1924: Kansas City Monarchs manager Jose Mendez takes the mound to spin a 3-hit, 5–0 shutout over the Hilldales to win the final game of the first Negro League World Series. Nip Winters had pitched the first three Hilldale wins.
   11. Chris Cobb Posted: November 13, 2004 at 08:02 PM (#964861)
The application of the Win Shares short form to my ERA+ and win share projections for win shares is interesting and may provide a more accurate projection of what the win shares system would find than my straight win-based analysis does.

I note, however, that this form, in relying on ERA, has the same problem as real ERA/ERA+ in that it does not separate out the fielders' contribution to the pitcher's earned runs allowed.

I tried to make Winters' ERA+ a support-neutral figure by adjusting it for fielding support. In several seasons, his actual ERA+ would have been better than the number I provided.
   12. Dr. Chaleeko Posted: May 22, 2005 at 10:52 PM (#1355023)

WINS 8th with 127 wins.

LOSSES 7th with 79 losses.

DECISIONS 8th with 208 decisions.

(50 Decisions Minimum) 29th
(25 Decisions Minimum) 48th
(10 Decisions Minimum) 81st

(50 Decisions Minimum) 8th
(25 Decisions Minimum) 12th
(10 Decisions Minimum) 24th

WAT 6th with 18.1 WAT

(50 Decisions Minimum) 23rd
(25 Decisions Minimum) 39rd
(10 Decisions Minimum) 66th


1922 2nd in East with 6 wins.

1923 Led ECL with 10 wins (tied), 5th in NgLs.

1924 Led ECL and NgLs with 27 wins.

1925 Led ECL and NgLs with 21 wins.

1926 2nd in ECL and NgLs with 23 wins.

1927 t-2nd in ECL with 18 wins, t-3rd in NgLs.

1928 t-3rd in ECL with 10 wins, t-7th in NgLs.
   13. Dr. Chaleeko Posted: May 22, 2005 at 10:57 PM (#1355025)
Winters is a really prime example of the distortions of schedule on how Negro League pitchers look in comparison to one another. Winters's career took place in an era when the number of league games in the NgLs was historically high and when the reporting of them may also have been at an historical high. Therefore, he had much greater opportunity to amass the counting stats you see above and the positions on the leaderboards that you see.

The effect is to make Winters look like a great career/bulk candidate AND a great peak candidate. In fact he is not a great career candidate and his career is shaped more like Dizzy Dean's or Sandy Koufax's than the rankings and numbers say.

The flip side is someone like Dave Barnhill who may or may not be both a good peak and career candidate, but whose career has a variety of interruptions, has somewhat shorter schedules, and whose teams may or may not have received equal coverage to the 1920s pitchers.
   14. KJOK Posted: September 17, 2011 at 07:58 AM (#3927884)
   15. Qufini Posted: September 17, 2011 at 02:41 PM (#3928004)
Thanks for posting these, KJOK. However, the Nip Winters link led me to Steel Arm Johnny Taylor.

Edit: The Jud Wilson and Joe Williams links are incorrect as well.
   16. KJOK Posted: September 17, 2011 at 04:52 PM (#3928070)
Thanks Chris. That worries me a bit, if our IDs are changing with new updates possible. Hopefully I just screwed up!

Nip Winter's Real Stats
   17. KJOK Posted: September 17, 2011 at 08:18 PM (#3928170)
Ok, REALLY corrected now:

Nip Winters' Real Stats
   18. Qufini Posted: September 17, 2011 at 08:23 PM (#3928180)
Worked that time. Thanks, KJOK. I was a little worried too.

You must be Registered and Logged In to post comments.



<< Back to main

BBTF Partner

Dynasty League Baseball

Support BBTF


Thanks to
Adam M
for his generous support.


You must be logged in to view your Bookmarks.


Page rendered in 0.4072 seconds
41 querie(s) executed