Not Just Friends of Frisch – Part 5

HHS contributor “no statistician but” continues his look at Hall of Famers who maybe shouldn’t be. Specifically, he’s examining those HOFers with a Hall of Stats rating under 100. Earlier posts looked at position players. This post concludes the series with a look at pitchers. More after the jump.

To me, pitchers present a special problem of interpretation with reference to both the Hall of Fame and the Hall of Stats. The HOF includes 83 pitchers out of its 261 Major and Negro League player inductees, 32% of the total, a huge body of talent to assess.

And the HOS, to me, shows its weakest side in the rankings it gives pitchers, since it favors accumulators to a tremendous degree. By the HOS ranking Early Wynn (108) is ‘better’ than Billy Pierce (101), a premise that to my insane mind no sane person would subscribe to. Further, anyone who ranks Bret Saberhagen (121) over Juan Marichal or Don Drysdale (both 115) will like the HOS approach, but I wonder if many such people exist.

Thus, to me the following listing of pitchers in the HOF but not the HOS cuts both ways in terms of its arguability. For starting pitchers, in any event, I have included the HOS rating, innings pitched, ERA+, pitching WAR (From Baseball-Reference), and Win Shares (WS) and Career Assessment Win Shares (CAWS), both from Baseball-Gauge (note that CAWS+ is based on 100 = 220 CAWS for starters and 160 CAWS for relievers). For relievers, it’s those stats plus game appearances and saves. Previously, I’ve used the 7th ranked player by JAWS as a reference, but I’ll depart from that approach here to instead select reference pitchers with workloads more like the pitchers under discussion.

LH Starters
Rube Marquard 65—3306.2 IP; 104 ERA+; 36.6 pWAR; 210 WS; 188 CAWS; 86 CAWS+
Herb Pennock 82—3571.2 IP; 106 ERA+; 43.9 pWAR; 239 WS; 201 CAWS; 91 CAWS+ (WW I: 1 year)
Lefty Gomez 77—2503.0 IP; 125 ERA+; 43.1 pWAR; 191 WS; 188 CAWS; 85 CAWS+
Reference: Carl Hubbell 136—3590.1 IP; 130ERA+; 68.6 pWAR; 305 WS; 262 CAWS; 119 CAWS+
Last man in HOS: Jerry Koosman 100—3839.1 IP; 110ERA+; 57.0 pWAR; 246 WS; 200 CAWS; 91 CAWS+

RH Starters
Jack Chesbro 76—2896.2 IP; 111 ERA+; 41.9 pWAR; 210 WS; 210 CAWS; 95 CAWS+
Chief Bender 91—3017.0 IP; 112 ERA+; 42.9 pWAR; 228 WS; 195 CAWS; 89 CAWS+
Waite Hoyt 95—3752.1 IP; 112 ERA+; 53.3 pWAR; 267 WS; 209 CAWS; 95 CAWS+
Burleigh Grimes 98—4180.0 IP; 108 ERA+; 46.9 pWAR; 279 WS; 238 CAWS; 108 CAWS+
Bob Lemon 76—2850.0 IP; 119 ERA+; 37.8 p WAR (10.8 oWAR); 230 WS; 223 CAWS; 101 CAWS+ (WW II: 3 years)
Catfish Hunter 65—3449.1 IP; 104 ERA+; 36.6 pWAR; 202 WS; 186 CAWS; 84 CAWS+
Jack Morris 77—3824.0 IP; 105 ERA+; 44.0 pWAR; 222 WS; 185 CAWS; 84 CAWS+
Reference: Jim Palmer 126—3948.1 IP; 125 ERA+; 67.5 pWAR; 307 WS; 264 CAWS; 120 CAWS+
Last man in HOS: Orel Hershiser 101—3130.1 IP; 112 ERA+; 51.3 pWAR; 206 WS; 175 CAWS; 80 CAWS+

Relievers
Rollie Fingers 53—944 G; 1701.1 IP; 341 SV: 120 ERA+; 25.1 pWAR; 187 WS; 157 CAWS; 98 CAWS+
Rich Gossage 90—1002 G; 1809.1 IP; 310 SV; 126 ERA+; 41.9 pWAR; 219 WS; 182 CAWS; 114 CAWS+
Bruce Sutter 55—661 G; 1042 IP; 300 SV; 136 ERA+; 24.6 pWAR; 161 WS; 157 CAWS; 98 CAWS+
Lee Smith 62—1022 G; 1289 IP; 478 SV; 132 ERA+; 29.4 pWAR; 193 WS; 159 CAWS; 99 CAWS+
Trevor Hoffman 63—1035 G; 1089.1 IP; 601 SV; 141 ERA+; 28.1 pWAR; 185 WS; 148 CAWS; 92 CAWS+
Reference: Hoyt Wilhelm 108—1070 G; 2254.1 IP; 228 SV; 147 ERA+; 49.8 pWAR; 253 WS; 188 CAWS; 117 CAWS+
Last man in HOS: Hoyt Wilhelm

For the final time: The challenge here is to argue for (or against) the presence in the Hall of Fame of any or all of the fifteen HOFers listed with a HOS rating of below 100. One argument is disqualified: Saying that Richard Roe doesn’t belong because John Doe, who was better, has been passed over for inclusion fails on every count to meet the terms of the challenge. The merits or demerits of the listed players, comparisons to other HOFers, the use of more detailed statistics, historical and biographical information — these and similar bases for argumentation are all welcome. Later on I’ll weigh in with some observations about what I think might be at issue regarding some of these pitchers.

0 0 votes
Article Rating
Subscribe
Notify of
guest

54 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
no statistician but
no statistician but
5 years ago

To contradict my boilerplate remark above, I’m going to weigh in now. I think the long shadow of Mike H’s series on win shares and his CAWS system of evaluation, plus the discussion thereof, might have the effect of obscuring the different nature of what the NJFOF series has been about. The players here aren’t discussable as HOF heavyweights. Mike is fishing in a different pond. Further, the almost universally negative evaluation of position players in the comments on the earlier parts of this series makes me wonder if there isn’t a generally held notion that to say anything at… Read more »

Voomo Zanzibar
5 years ago

Amazing how with all the Yankees offensive juggernauts, their starting pitching staffs were just-good-not-great.
The Yanx career leader in WAR for pitchers is a reliever.

Voomo Zanzibar
5 years ago

I see Gossage as HOFer.

3rd all-time in WAR for relievers.
The two ahead of him are on another planet, but 4th place is a galaxy away from Goose:

56.3 … Rivera
49.8 … Wilhelm
41.7 … Gossage
31.1 … Hiller
29.3 … Lee Smith
…..

In an era of great Closers with great facial hair, Gossage was at his peak arguably the most intimidating and most effective, and his moustache rivaled that of Lyle, Fingers, and Quisenberry (which would be a great name for a mortuary).

Voomo Zanzibar
5 years ago
Reply to  Voomo Zanzibar

Highest single season WAR, relievers:

8.2 … Gossage
7.9 … Hiller ………. sometimes a great moustache
7.3 … Eichhorn …. terrible moustache
6.6 … Sutter ……… terrible beard
6.2 … Abernathy .. clean shaven, earlier fashion era
6.1 … Radatz …….. see abernathy
6.1 … Kern ………… a beard so bad I say it was good.
6.0 … Gossage

Voomo Zanzibar
5 years ago
Reply to  Voomo Zanzibar

Highest WAR in a season of fewer than 50 IP
(Gossage makes this list because it was 1981):

2.7 … Victor Cruz
2.6 … Mike Myers (126.9 park factor! Denver in 2000)
2.5 … Sean Doolittle
2.4 … Ray Searage
2.3 … Gossage
2.3 … Jay Howell
2.3 … Josh Hader (active…)

Victor Cruz was a rookie.
I cannot find any info about why his career was so short.

https://www.baseball-reference.com/players/c/cruzvi01.shtml

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Voomo Zanzibar

Voomo, Your point about Victor Cruz’s premature disappearance is interesting. He did have an impressive MLB record. His ERA+ was fine in ’81, but his WHIP of 1.412 may have been the reason he was demoted. Although his brief stint with Texas in ’83 was successful, he was playing the role of set-up man for only the latter portion of the season, so it’s not too surprising he didn’t make the roster the next spring (though the Rangers could have used a good reliever). If you look at his Minor League record in ’84-’85 I think the reason his career… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago

I want to comment on relievers first because while I had a list of relevant stats ready to discuss starters a month or two ago, the discussions of Mike’s CAWS system have changed a lot of what I was thinking of saying about them. I’ll have to rework that. But relievers — firemen, swingmen, and closers — seem to me really outside the bounds of CAWS. Mike sets up an entirely different metric to deal with them and I’ve already discussed that in an earlier post. Because relievers seem to me to be best judged on statistics other than WS… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Doug

Great work, Doug, and presented very clearly. The case for Lemon seems strong. I suspect that he’s being undermined in WAR by several factors: — Pitching in front of a very strong defense during most of his high-IP seasons (RA9def averaged 0.24 in the period 1947-54); — a generally favorable park factor (96); — facing weak overall opposition because Cleveland was frequently one of a minority of strong teams. The last two combined limit his RA9avg – RA9 to 0.47 (I randomly chose Justin Verlander for comparison: 1.21 in comparable IP). That is, the way B-R calculates, Lemon is not… Read more »

Doug
Doug
5 years ago
Reply to  Bob Eno (epm)

I was thinking about your point about the imbalance in the AL in those days, with only 2 or 3 strong teams most seasons. So, it’s surprising how little difference there was in the games and IP for these four pitchers against winning or losing teams. That must mean that these four did most of the pitching against the better teams, and the other spot starters were used mostly against the more numerous weak sisters.

no statistician but
no statistician but
5 years ago
Reply to  Doug

Doug: Lemon’s BB rate was consistent with the era, I’m thinking. He never led the AL in walks, and although he finished in the top ten every year, he generally was first or second in innings pitched. Nowadays pitchers don’t mind giving up homers. Back in the 1946-1955 decade, when the Eddies and others were working the walk, pitchers didn’t mind THAT. One of the (several) things that modern statistical evaluations do badly is failing to to adjust enough for trends. Walks were high, strikeouts were low in Lemon’s era. Double plays, as a result, were high. That players reflect… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago

nsb, This theme seems to me such an overriding issue in Hall decisions that it really deserves at least an outline study. How do the parameters of excellence change over time, and exactly how should it affect the way we view stats from different eras differently? When the BBWAA has voted on recently retired players, we can be pretty sure that there has been at least rough comparability between the standards the voters had in mind and those that governed play during a candidate’s career. The logic of the Old Timers Committee was that you need “scholars” of baseball during… Read more »

Doug
Doug
5 years ago

Best 9 Consecutive Win Seasons, selected live ball era pitchers.
200 – Grove (1927-35)
189 – Feller (1938-50)*
187 – Spahn (1953-61)
186 – Lemon (1948-56)
185 – Marichal (1961-69)
184 – Roberts (1950-58)
183 – Jenkins (1967-75)
182 – Hubbell (1929-37)
178 – Feller (1938-49)
176 – Palmer (1970-78)
175 – Ferrell (1929-37)
174 – Perry (1966-74)
173 – Grimes (1920-28), Gibson (1964-72)
172 – Carlton (1972-80)
171 – Hunter (1968-76), Seaver (1969-77)
170 – Wynn (1951-59)
*omitting Feller’s partial 1945 season after returning from military service

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Doug

Doug, Here are the guys on your list, adding the percentage of team wins that they owned. I omitted Feller ’38-’49 because of the partial ’45 season, and I fudged on some mid-season trade situation by just averaging the wins of the two teams. 200 – Grove (1927-35) 24.1% 189 – Feller (1938-50)* 24.8% 187 – Spahn (1953-61) 23.3% 186 – Lemon (1948-56) 21.9% 185 – Marichal (1961-69) 22.5% 184 – Roberts (1950-58) 26.2% 183 – Jenkins (1967-75) 24.2% 182 – Hubbell (1929-37) 23.0% 176 – Palmer (1970-78) 20.9% 175 – Ferrell (1929-37) 24.6% 174 – Perry (1966-74) 23.0% 173… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago

When it comes to starting pitchers, one of the key questions to ask is to what degree we want to measure their success solely on pitching and to what degree batting counts. So far as I can tell from the Hall of Stats site, although Adam Darowski doesn’t appear to specify, it seems that the ratings are based on pitching only. WS does consider pitcher batting, but since WS has no negatives that means most pitchers rack up 0.0 offensive WS each season, and batting only counts for good hitters, and only for years when those good hitters hit good.… Read more »

no statistician but
no statistician but
5 years ago
Reply to  Bob Eno (epm)

Re Lefty Gomez: In two seasons he was the best pitcher in the AL, 1934 and 1937, Grove’s higher WAR in ’37 notwithstanding. His career was short, ten qualifying seasons, but with an ERA+ of 123 or better in 7 of them. His 6-0 record in World Series play is unmatched. Ruffing, with whom he is often associated, wasn’t nearly as good, just effective longer. The latter’s 2.067 WAR/162 pales compared to Gomez’s 2.8. His subsequent career as a self-deprecating raconteur, like Dizzy Dean’s long stint on the Game of the Week as a professional hick, should have nothing to… Read more »

mosc
mosc
5 years ago

The only pure relievers I think are worthy are Rivera and Gossage. I’d take Eckersley, Smoltz, and Wilhelm for their total work as well. Other than that, I don’t think any of these guys are particularly close.

no statistician but
no statistician but
5 years ago

Herb Pennock and Waite Hoyt are linked in much the way Gomez and Ruffing were a decade later. Pennock, the lefty, was the better pitcher most seasons, but Hoyt, the righty, six years younger, timed his career better, escaped the dark hole that the Red Sox were becoming two years sooner, and had his career year in ’27. Pennock debuted prematurely for Connie Mack and after a decent season in 1914 floundered for the next three and spent the fourth (1918) in the Navy. He resumed in 1919 for the Red Sox a mature pitcher and led the staff in… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago

nsb, Perhaps it’s frailty of character, weakness of intellect, or the fading memory of sexual prowess, but I really don’t think any of these candidates, other than Lemon, warranted election to the Hall. Every one of them had outstanding careers or they wouldn’t be in this discussion, and several are attractive for the way they handled themselves and their careers, especially, for me, Bender, Gomez, Hunter, and Morris. But I see the essence of this discussion to be about how we set consistent criteria for the Hall beyond what numbers can capture, and to make that discussion meaningful, I think… Read more »

Doug
Doug
5 years ago

Pennock probably benefited from the advocacy of the Babe in getting his Hall selection. Call it Friends of the Babe, if you will. Several other second-tier players from those Yankee teams made it to Hall, but Pennock did so 20+ years before the others and while Ruth was still alive.

Bob Eno (epm)
Bob Eno (epm)
5 years ago

I want to comment on part of nsb’s initial post that caught my eye: By the HOS ranking Early Wynn (108) is ‘better’ than Billy Pierce (101), a premise that to my insane mind no sane person would subscribe to. Further, anyone who ranks Bret Saberhagen (121) over Juan Marichal or Don Drysdale (both 115) will like the HOS approach, but I wonder if many such people exist. I had the same reaction that nsb does, but after looking into this more closely, I think that the story is complicated. Let’s look at Saberhagen and the ‘60s stars first: bWAR…..pWAR…..bWAR/162..pWAR/162..CAWS+…CbWAR+..CpWAR+…ERA+… Read more »

no statistician but
no statistician but
5 years ago
Reply to  Bob Eno (epm)

The point about Saberhagen vs the others is not that his career stats are deficient but that he disappeared from view, at first roughly every other year, then for years at a time. In fact, he had only seven qualifying seasons, and in two of those he pitched under 180 innings. I’m not a fan of either Drysdale or Marichal, but they were there making a positive difference that could be depended upon from season’s end to season’s end year after year. Saberhagen to me is just a statistical outlier whose record has limited substance, exaggerated by the foibles of… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago

Well, nsb, you’re making sort of the reverse of a Hall narrative argument: the stats are there for Saberhagen, but the narrative reduces their significance. I think it’s a good point. Saberhagen did win to CYAs, and his scattered high points were very high. But his career lacked the predictable shape that adds extra values for contending teams (a point Doom has made several times), while Drysdale and Marichal’s careers had classic shapes. I really don’t see any problem with advanced stats here, but stats, whether advanced or traditional, only get us part of the way, and on reflection I… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Bob Eno (epm)

Hmmm: “. . . to CYAs . . .” I’m helpless against armies of typos, but that one’s more illiterate than most.

Doug
Doug
5 years ago

I’m going to make a case for Rollie Fingers. We tend to forget or minimize how different and how much more difficult (in my view) it was to be a primary reliever in Fingers’ time. It’s worth remembering that there used to be an award for the best reliever called the “Fireman of the Year” award, as in the pitcher who was most adept at putting out fires, coming into games in difficult situations and getting out of those jambs. And, oh, by the way, since you’ve done that, why not stick around and pitch another inning or two and… Read more »

no statistician but
no statistician but
5 years ago
Reply to  Doug

Here we come to the same type of problem I raised concerning Bob Lemon’s walks. As I’ve tried to argue several times over the past few years, retrospective statistical formulas like WAR, whether intentionally or not, end up promulgating a one-size-fits-all solution to evaluation. Fingers’ pathetic HOS score of 53, given your penetrating breakdown of his outings, appears almost a slight approaching libel of character. I’ve haven’t written this for a while, but here it is, taken out of mothballs: the devil is in the details. Obscuring differences leads to unintended results that obscure rather than illuminate. And the solution… Read more »

no statistician but
no statistician but
5 years ago

Seems like I’m talking in circles here. Let’s try “Failing to account for pertinent differences leads to unintended results that obscure rather than illuminate.” Also, how about “whatever mathematical correlations the mathematician decides are correct.”

Bob Eno: I’m up on you two to one.

Bob Eno (epm)
Bob Eno (epm)
5 years ago

Sorry, nsb, I’ve been tracking your slips and corrections for years. When it comes to blockheaded typos and grammatical solecisms, you’re simply not in my calss.

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Doug

nsb’s right: Fingers is much undervalued by advanced stats because the context in which he pitched was ephemeral, and the distortions were never understood or corrected. (At least the distortions that pertain to closers are now being recognized, though no mathematical solutions have been found — Adam Darowski has a great reply to nsb’s complaint about Fingers’ rating, “I’m also not happy yet with how the Hall of Stats handles relief pitchers.”) But I don’t think Doug’s description of Fingers’ performance changes the equation. I think he has given us a narrative about Fingers that is less impressive when contextualized,… Read more »

Doug
Doug
5 years ago
Reply to  Bob Eno (epm)

Wonderfully reasoned, as always, Bob. I might part company with you when you conclude, I think, that relievers probably shouldn’t really be celebrated (be it HOF, COG or whatever) since they are relievers because they weren’t good enough to be starters. They probably also weren’t good enough to be outfielders or third basemen, either, but I don’t think that’s what’s germane. Rather, these pitchers are on a ball club because they fill a role that the club values. And, the best relievers are entrusted with arguably one of the most important roles – preserving wins. How that role is fulfilled… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Doug

Well, Doug, I’ve probably fired all the ammo I’ve got, and your position is still standing. I guess I’ll just have to dig some works and camp the troops in place, awaiting a future call to battle on somewhat different ground. Apart from the DH, which has its own problems, relievers are probably the toughest cases for assessment. As you say, they play a valuable role. But what does “role” mean? Every position player plays a unique role in the game and we judge them differently according to role. But those differences are peripheral, because the unique role pertains only… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago

I’d like to celebrate Independence Day by returning to an issue discussed on the last thread: the third in Mike Hoban’s CAWS series. The issue concerns the design of an ideal one-number metric for a Hall monitor. (It has just struck me for the first time how silly it sounds to talk about “Hall monitors” — after high school, anyway.) A couple of days ago, mosc left a very (a href=”http://www.highheatstats.com/2019/06/caws-career-gauge-part-3/#comment-142724″>interesting comment on that issue in response to one of mine, and lying as it is in the middle of a string that effectively closed last month, I think his… Read more »

Doug
Doug
5 years ago
Reply to  Bob Eno (epm)

That idea of mosc’s and Doom’s seems like an interesting approach.

Any thoughts on what a “substantive” WAA level should be?

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Doug

Well, first, I see I botched hyperlink (nsb, are you counting?). You can still cut-and-paste, though, so no harm done, I hope. I may have misrepresented mosc. The phrase he used was not “solid” or “substantive season,” which is how I understood him. The phrase was, “the longest relevant timeframe.” His example (Willie Mays) suggests he just meant, “till you exhaust all WAA+ seasons.” But since there’s an issue of contiguity in “timeframe,” I wonder whether, had Mays tacked on a few successful at bats in 1974, after a WAA- season, they would have figured in mosc’s or Doom’s calculations.… Read more »

no statistician but
no statistician but
5 years ago
Reply to  Doug

My problem is this: I don’t think WAA is any better—truer, less likely to be based on assumptions that are not accurate in every circumstance— than is WAR or conflations of it with other formula-based valuative systems. In the notorious 1996 campaign, Roger Clemens went 10-13, and the Red Sox went 14-20 (.411) in his appearances. Yet he was awarded 5.4 WAA based on a waaWL% of .659. That is, according to the formula, an average team would have played at a .659 winning percentage in the games he started. Were the Red Sox a substandard team that year to… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago

The BoSox provided Clemens an average of 4.29 R/G. An average team in the 1996 AL scored 5.39 R/G. That’s why an average team with Clemens pitching would have had a .659 winning percentage. It’s not that the BoSox were below average (they scored 5.72 R/G); it’s that they were below average when Clemens was pitching. Given Clemens’ opponents and park, the average pitcher’s RA/9 was 6.07, Clemens’ rate was 3.93. That’s why advanced stats thinks he was pretty good, while his W-L suggests he was mediocre. I don’t see the disconnect. I think you’re looking at stats that tell… Read more »

no statistician but
no statistician but
5 years ago
Reply to  Bob Eno (epm)

Here we come down to the problem that plagues me about the use of the term Wins. “Wins” above replacement, “Wins” above average. The only wins that count are the ones that show up on the scoreboard at the end of the game. I don’t doubt that Clemens was occasionally run-deprived overall during the season compared to others (although not particularly game by game), but considering his performance in a similar season, 1992, I see what to me can’t be explained away by the magic of numbers: 1992: 246.2 IP; 18-11 (team 21-11); 2.41 ERA; 203 H; 80 R; 66… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago

nsb, In 1996, Clemens was pitching in an environment where teams were scoring 5.59 R/G; in 1992, they were scoring 4.32 R/G. The ’96 BoSox upped their game and scored 0.20 more R/G for Clemens than they did in ’92, but the league upped its average 1.37 R/G. In ’92, Clemens allowed only 2.92 R/9IP while the teams he faced averaged 4.32 (4.86 when adjusted for park factors). That’s great. In ’96 he allowed 3.93 R/9IP while the teams he faced averaged 5.42 (6.07 adjusted). (Let’s ignore for now the fact that the Sox of ’96 have a much worse… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Bob Eno (epm)

Looking again, I over-simplified mosc’s description of his method. I think the formula should look like this, with S1 meaning best season, second best S2, etc. S1/1+(S1+S2/2)+(S1+S2+S3/3) . . . . (S1 . . . SN/N) I realized when watching my cat cower at the neighbors’ fireworks that the way I’d (mis)represented it wouldn’t work. I’m not convinced that this will give adequate weight to players with short careers. (Doom and I once had an argument over who contributes more, a player who’s retired or one who’s resisting retirement and playing poorly, and Doom chose the latter, so he wouldn’t… Read more »

Doug
Doug
5 years ago
Reply to  Bob Eno (epm)

Thanks Bob,

I’ll try some spreadsheet magic and see if I can compute it reliably.

When it’s written (s1+s2+ … sn / n), do you suppose it’s really meaning (s1+s2+ … sn) / n ?

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Doug

If you check post, Doug, I think he means it the way I wrote the formula. I could be wrong . . .

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Bob Eno (epm)

Oh. Yes, you’re right.

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Doug

Wow! Your computer is bigger than mine. (Well, actually, your understanding of how to do this efficiently is bigger than mine. Way bigger.) The Top 25 list looks reasonable, though the inclusion of Spalding is an alert that true “Pioneer” players are going to have skewed standing in the list. The Hall of Stats doesn’t even include National Association players. But if you go down further, things get pretty strange. Let’s say the cutoff for consideration is 20.0 WAA_MD (not sure what “MD” means). That would give us 886 qualifying players (Vic Wertz is #886). Why go down so low… Read more »

Doug
Doug
5 years ago
Reply to  Bob Eno (epm)

MD is mosc-Doom.

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Doug

Of course.

Doug
Doug
5 years ago
Reply to  Doug

I’ve updated the link with some more information.

The multiplier (ratio of gWAA_MD to gWAA_Pos) has quite a range, from infinitesimally more than one (generally for players with 2 positive WAA seasons, with the second barely so) to over 2 for a few players. Among notable players (10 or more positive WAA seasons), the highest multiplier is 2.07 for Red Faber and the lowest is 1.20 for Pie Traynor.

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Doug

This is going to take a while to study, Doug. The ratios you point to concern the impact of the WD system, which is one issue. For example, Faber’s high ratio seems to be due to his career profile: he has exactly two top seasons, according to gWAA (7.6, 5.7), then his ratings fall off a cliff (1.8, 1.7 . . .). Those two top years are components of 16 seasons in WAA+ territory, increasing their weight and accounting (I think) for the top-heavy ratio. (In bWAA, the series goes: 8.1, 6.0, 1.5, 1.5, 1.5 . . . .) Another… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Bob Eno (epm)

I’ve looked through the list of Gauge WAA+ (MD) to identify borderline Hall cases, using the standard that the top 232 qualify. For the purposes of our interests, I’ve removed all players who played the majority of their careers before 1893, when the mound position was reconfigured and Major League baseball came to resemble the modern game — there were 23 of these players, ranging from consensus greats like Cap Anson to more obscure figures like Jim Whitney. Whatever their fame, the stats of the pre-1893 era simply are not comparable to later stats, particularly for pitchers. (Adam Darowski subjects… Read more »

Doug
Doug
5 years ago
Reply to  Bob Eno (epm)

Seems like the MD method seems rather prone to signifcant distortions, at least using the Gauge WAA.

I’ll see if I can get Doom or mosc to join the discussion.

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Doug

I ran the formulas for Tinker and Aparicio using B-R WAA and things look much more reasonable. Tinker’s still well ahead, but by only 44.2 to 37.3. The HoS gives Tinker a smaller 103 to 94 advantage (which makes sense, since its basic formula is WAR+1.8*WAA+). The real impact in this case comes from the switch from WAR to WAA. Aparicio gets three WAR+ seasons knocked off of his 18 (plus one WAR-), while Tinker loses only one WAR- season out of his 15. Tinker is actually just higher than Aparicio in WAR for the 14 seasons each registered a… Read more »

Bob Eno (epm)
Bob Eno (epm)
5 years ago
Reply to  Bob Eno (epm)

Although this string seems to be effectively closed, I want to append a generalized issue I see with the mosc-Doom method, one that grows out of reflecting on the Tinker/Aparicio contrast and the exceptionally poor showing of Lou Brock under the system. The issue concern the toggle between WAA+ and WAA-neutral/-. Like all arbitrary thresholds, the toggle can introduce distorting results. Here is example: Player A’s career WAA seasons: 9.0, 8.0, 7.0, 6.0, 5.0, 4.0, 3.0, 2.0, 1.0, 0,1, 0.1, 0.1, 0,1, 0.1, 0.1. Player B’s career WAA seasons: 9.0, 8.0, 7.0, 6.0, 5.0, 4.0, 3.0, 2.0, 1.0, 0.0, 0.0,… Read more »

Doug
Doug
5 years ago
Reply to  Bob Eno (epm)

Bob, you have hit the nail squarely on its head.