Hot topics bound . . .

Here are my notes on John Donaldson, from various sources:
The top negro league lefty, possibly in Jose Mendez territory. Pop Lloyd called him the “toughest pitcher he ever faced.” The Glasgow, Missouri, native was a good hitter. From 1924-1930, against semi-pro teams, he hit .400 with a 83-35 W/L record.

I have him ranked just below Addie Joss at the moment.

In response to James Newburg's query (welcome back, btw!), I've done an analysis of John Donaldson in the same fashion as the analysis of Rube Foster.

This analysis relies more heavily on i9s than the Foster analysis did. I have no independent data (of any usefulness) on Donaldson's first three seasons, when he was, by reputation, at his most brilliant. And I have only incomplete w-l records for the rest of his negro-league career. However, the data I do have suggest to me that i9s underrates Donaldson's later career by about 5%, and underrates his last big season (1916 -- first one for which there is data) by more than that. So I've generally adjusted their numbers upwards by 5%, except for 1916.

Here's the straight i9s career reconstruction of Donaldson's pitching career (no credit for military service)
1913 – 1921, 1915 IP, 2.82 ERA, 104.3 ERA+, .521 support-neutral wp, 117-108 est. record, 134 career win shares

Here, for a different view, are my year-by-year estimates

(no batting ws included)

Season – IP – ERA+ -- SN wp – SN W/L -- WS (comments)
1913 – 291 – 115.8 -- .573 – 19-15 – 24 (no raw data, 5% boost to i9s)
1914 – 338 – 139.7 -- .661 – 26-14 – 38 (no raw data, 5% boost to i9s)
1915 – 284 – 138.0 -- .656 – 22-11 – 32 (no raw data, 5% boost to i9s)
1916 – 318 – 113.4 -- .563 – 21-17 – 24 (8% boost to i9s* based on available data)
1917 – 103 – 106.8 -- .533 – 6.5-5.5 – 18 (Military service, 7.2 raw WS * 2.5)
1918 – 157 – 93.2 -- .465 – 8.5-10 – 11 (Military service, 7.1 raw WS * 1.5)
1919 – 182 – 97.0 -- .485 – 11-11 – 10 (3-2 for a Detroit Stars, strong offense)
1920 – 145 – 80.3 -- .392 – 7-10 – 3 (5-4 for strong KC Monarchs, “dead arm”)
1921 – 97 – 88.1 -- .437 – 5-7 – 3 (played mostly outfield)
career – 1915 – 112 – .556 -- 126-100 – 163

*Went 5-1 for an apparently weak-hitting All-Nations team vs. negro-league competition, 0-2 in Florida winter ball. This converts to a .563 SN wp, as I see it, where i9s, as I figure it, has him at a .522 SN wp

Comments – while I haven’t dealt systematically with Donaldson’s batting value, he was a good enough hitter to play as a regular outfielder for two years in 1921-22, but not good enough to sustain a career there. So he was a very good hitter for a pitcher, I think, even at the major-league level, but he was not a good enough hitter to have ever been a starter as a position player at the major-league level. I don't think it would give him enough value to make him a viable candidate, but peak-oriented voters might see the matter differently.

Awesome stuff, Chris! This moves Donaldson up some in my rankings, but he's nowhere near my ballot still.

At the risk of sounding greedy, could you run something similar for Dave Brown? i9s projections have him as what seems to me like nearly a non-candidate (only 5 full seasons, ERAs not nearly in Donaldson's range). I have him basically completely outside my consideration set right now, but I'd like to make sure before I do it completely. (He does seem to have a tiny bit of expert opinion on his side)

Looks a lot like Dave Stewart to me Chris. Funny, I just wrote the same thing about Dave Brown . . .

Chris, What are the sources for your stats on John Donaldson?

Dave,

The stats I present in the post above are derived from the major-league _projections_ developed by the group running the Integrated 9s project and website. You can see their work at http://www.i9s.org.

Their projections are based on their work with whatever available actual statistics on Donaldson they have found.

In my post, I modified their projections in ways that seemed reasonable to me and used them in conjunction with actual national-league and american-league statistics from his career to calculate projected ERA+ and support-neutral won-lost records.

1921 John Donaldson Batting
NNL Kansas City Monarchs

G-90 (team 90)
AB-340
H-96
D-10
T-5
HR-0
R-57
W-23
HP-1
SH-18
SB-11
AVE-.282 (NeL .263)
OBA-.330 (NeL .324)
SLG-.341 (NeL .361)

He was the Monarchs' regular centerfielder and usually batted third.

1921 John Donaldson Pitching
NNL Kansas City Monarchs

W-0
L-2
SV-1
TRA-5.97 (NeL 5.20)
G-8
GS-2
CG-1
SHO-0
IP-34.7
H-33
HR-0
R-23
BB-10
K-13
HB-1
SH-2
SB-11
DP-3
OOAVE-.250 (NeL .263)
OOBA-.308 (NeL .324)
OSLG-.333 (NeL .361)

Obviously he was past his prime as a pitcher. The Monarchs generally sent him to the mound only in emergencies.

I think we might be underrating John Donaldson if majority of these teams he barnstormed against were of minor leauge quality

It's my understanding that he was generally pitching against semipro competition, at least after 1920. Although the HoF bio doesn't make it clear, Donaldson was pitching in integrated semipro leagues in the upper midwest through much of the 20s: that's where a lot of these tremendous numbers come from. It's unfortunate that the HoF bio is not more precise in establishing the contexts for the stats it presents.

The data Gary A. has posted above show that Donaldson was done as an NeL-quality pitcher by 1920.

His records from the teens are much more suggestive of greatness, but they are fragmentary, and most of the eye-popping numbers were accumulated against competition of very mixed quality.

If those who are more knowledgeable about Donaldson's history than I am see errors in this account, please make corrections! I am drawing here on what I've read in Riley's and Holway's works.

John Donaldson's Real Stats

Hey, gang,

Please find my latest MLE for John Donaldson here. As always, when new data arrives, these will be updated. Thanks!