Baseball for the Thinking Fan

You are here > Home > Baseball Newsstand > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

## Monday, November 30, 2020

#### Bill James: The Biggest Problem With WAR

The player’s WAR can vary by 150%, based on 3 errors of 3% each.  THAT is the basic problem with WAR.  I mean, we argue about all kinds of things.  We argue about whether clutch data should be included in runs created estimates; we argue about whether the fielding estimates are consistent with external evidence, etc.  But the REAL problem is that:

1)    Estimates are never exactly right; they are always just estimates, and

2)    WAR uses an analytical system to process those estimates which has the potential to enormously magnify whatever inaccuracies are included.

In a WAR estimate, there are dozens and dozens of internal estimates—estimates of runs created, estimates of runs saved by fielding, estimates of the run value of a single, a double, a triple or a double play, estimates of the park effect, etc.

The problem is more serious than that.  First of all, as I said, the replacement level is not really an estimate.  It’s just a made-up number.  It could be off by 20 or 25%—by itself, before it is magnified.

But that understates the problem, too.  WAR assumes that the Replacement Level is a constant.  It is NOT a constant; it’s a variable.  Some teams, an outfielder gets hurt, it doesn’t really matter because they’ve got a fourth outfielder who is about as good as the starters.  Other teams, it matters a lot because their fourth outfielder is a pair of stuffed pajamas.  The actual replacement level is specific to the locale.  Rather than trying to estimate what the replacement level actually is in this case, WAR simply assumes that it is always the same.  To return to the analogy of the wheat farmer, this is like assuming that all trucks weigh the same.  It leads to large inaccuracies.

RoyalsRetro (AG#1F) Posted: November 30, 2020 at 01:29 PM | 197 comment(s) Login to Bookmark
Tags: war

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.

Page 2 of 2 pages  < 1 2
101. sunday silence (again) Posted: December 02, 2020 at 08:27 AM (#5991986)

It looks like about half his errors in his career-high 24-error 2000 season were on throws, but I don't know what the typical figure is.

SoSH: Inspired by your work. I decided to investigate Geo Brett's legendary 26 error 1976 season.

Total throwing errors: 3! 3 effin errors and James said off handedly "at least 13." This is why James can be infuriating at times.

OK all the others, except for one, are the standard: "boot a ball at 3b, no one advances that shouldnt, -0.7 def. runs."

The exception: June 19 BUddy Bell scored on a basehit, I guess he should have been out at the plate. That's the only runner advance so this error cant possibly be worse that 0.7 runs if that.

The three throwing errors resulted in 3 runners advancing 5 bases total. Scoring it -1.45, -1 and -1 def runs.

Did any errors "cost them the game." Probably two. The throwing error on May 27 was costly as several runners advanced, and Fregosi followed with HR two batters later when they should have been out of the inn. Another game in May on a standard error led to a couple unearned runs and probably lost the game for them. So two games where his errors were huge factors.

I think what makes this exercise even more interesting is that Bill James is a lifelong KCR fan and presumably watched tons of their games. At least thats the impression I got from reading him. I wonder if here he is watching the Royals early in the season and remembers this costly throwing error on May 27 costing them the game and another one in the month. Then forming some impression of young player (I think this is Brett's second season) and just taking that with him.

So to summarize, 26 errors, worth about 19.5 defensive runs. Really not bad at all.

A fun excercise. Im pretty sure if you did the same thing for BObby BOnilla, or Leo GOmez, or Dale Berra it would be a different story. I'd love to work this for Arky Vaughan.
102. Rally Posted: December 02, 2020 at 08:42 AM (#5991990)
The exception: June 19 BUddy Bell scored on a basehit, I guess he should have been out at the plate. That's the only runner advance so this error cant possibly be worse that 0.7 runs if that.

So Bell is on 2b, single to left field. The guy who hit the single stays at first. I'm trying to imagine how this became an E-5. Maybe Bell was held at 3B, the throw came in from left for Brett to cut off, but he muffed it, the ball rolls away and Bell scores. Maybe Bell went a little too far past third, but the throw to the plate would have beat him. He turns around to go back to third, the catcher throws it to Brett in plenty of time, but George drops it while trying to make a tag.
103. sunday silence (again) Posted: December 02, 2020 at 08:45 AM (#5991991)
yeah I was wondering the same thing. I think your first suggestion is most likely. If the ball rolled towards the infield then the runner on first probably holds. One the second suggestion, one would think the runner goes to instinctively if they're in the middle of a run down. That's a no brainer I think.

If he never really had a chance to get Bell (who I think was slow) then the play would only be worth something like -.25 def runs. interesting.
104. Ron J Posted: December 02, 2020 at 08:55 AM (#5991992)
#101 Confirmation bias in action. Keyed to pay attention to a you player James sees a couple of key errors and they make an outsized impression.

And these kind of things would have been moderately tough to run a tracer on back then. I think people forget how recent retrosheet data is. It's always been there for some people, but I can recall having to figure out how to parse retrosheet data to write one of the first articles using it. And to bring it back to topic, it was about Brett's hot streak in his run for .400
105. sunday silence (again) Posted: December 02, 2020 at 09:07 AM (#5991995)
speaking of weird stuff. I was checking the AL in 1966 and and CLE has their top 5 WAR guys all pitchers. Has that ever happened?
106. Rally Posted: December 02, 2020 at 09:34 AM (#5992001)
One of the teams I thought of was the 1991 Angels. They were a .500 team with a trio of lefties having great years, plus Bryan Harvey saving 46 games with a 1.60 ERA. 5 of their top 6 are pitchers, but #4 is Wally Joyner who hit .301 with 21 homers.

Seems like a very rare thing.
107. David Nieporent (now, with children) Posted: December 02, 2020 at 11:45 AM (#5992040)
That production isn't available in the marketplace and doesn't have any independent tangible meaning or significance. Yeah, if you could actually go into the marketplace and get assured replacement level production it would make sense. But you can't. You can't even approximate the concept.
If this were true, and of course FLTB presents no evidence that it is, then it would mean that replacement level were being set too high, not that replacement level doesn't have significance as a concept. There clearly exists some level of play at which you can get assured production in the marketplace.

There are better critiques that FLTB isn't smart enough to bring, such as the fact that the level of freely available talent is surely different during a season than it is in the offseason, which could mean that players who are less durable from day to day should be treated differently than players whose careers are just short. Two 81-game seasons shouldn't necessarily be deemed to have the same value as one 162 game season.
108. Rally Posted: December 02, 2020 at 01:41 PM (#5992067)
Two 81-game seasons shouldn't necessarily be deemed to have the same value as one 162 game season.

That might be a fun study. Easy to find examples of star players getting hurt and playing 60-100 games. Look at what kind of production teams got from the backups.

A little bit harder to find a good comparison group. A star player leaving for free agency doesn't work because that player made a lot of money, and now you have payroll space to spend on an above-replacement alternative. Maybe someone like Dave Winfield in 1989, injured and out for the season. But some teams do have payroll room to spend big in that situation, like when Corey Seager got hurt in April so the team trades for Manny Machado.

Looking back, the Winfield example doesn't work. They replaced him by trading for Jesse Barfield. He was past his peak but still far from replacement level talent. He made decent money (for 1989 baseball) and the Yankees also gave up Al Leiter to get him.
109. Hank Gillette Posted: December 02, 2020 at 08:42 PM (#5992159)
I don't think that the part about him pushing back against other people's work fully holds up. He praised DIPS and accepted it as true, mentioning it in his 2001 book and crediting the guy who came up with it.
If I remember correctly, it was kind of a back-handed compliment, because he said something like, “Oh, yeah, I should have come up with that.” Or, maybe he truly meant that a compliment.

KInd of off the main topic, but another thing he does that I really dislike is in his “Hey Bill” column, he will publish someone’s question and then respond with something like, “I have no idea what you are talking about.” That seems excessively mean-spirited to me. He couldn’t email the person and tell them their question wasn’t clear?

And, yes, I felt like I got burned a couple of times asking him questions, so I have a bias.
110. Jeff R. Posted: December 02, 2020 at 11:48 PM (#5992202)
KInd of off the main topic, but another thing he does that I really dislike is in his “Hey Bill” column, he will publish someone’s question and then respond with something like, “I have no idea what you are talking about.” That seems excessively mean-spirited to me. He couldn’t email the person and tell them their question wasn’t clear?

Yeah, that bugs me too--like, why does Bill even print the question if he's not going to put any effort into an answer?
111. caspian88 Posted: December 03, 2020 at 12:25 AM (#5992206)

Yeah, that bugs me too--like, why does Bill even print the question if he's not going to put any effort into an answer?

Pass.
112. Rally Posted: December 03, 2020 at 09:33 AM (#5992226)
Bill can certainly be a crank but at least he's talented, and for that I'll keep reading him. Unfortunately this attitude seems to have rubbed off to another writer on his site. Yesterday some dude called Dave Fleming (I doubt its the same guy who won 17 games for the 1992 Mariners) did his own anti-WAR screed.

The site recently had comparisons of Vlad Guerrero and Bobby Abreu, two contemporaries who had about the same value but did it in different ways. He brings Gary Sheffield into this, making a good observation that Sheffield combined the patience of Abreu with the power of Vladimir. All 3 players had about the same WAR total, so he bashes WAR for not giving credit for Sheffield's obviously superior offense.

A reader points out the obvious, Sheffield is way ahead in offensive WAR or batting runs, but loses ground on the defensive side. Then Fleming goes off in the comments about how he knew all that and tells everyone that WAR is a broken car and it doesn't matter exactly why its broken, just that it doesn't run. Why is it broken? Well, apparently because Sheffield does not rate the way Dave Fleming thinks he should. What's his evidence that Sheffield's defense was not that bad? He's got none. Summary opinion without evidence, or as some would say, pure BS.

Thing is, Dave Fleming is nowhere near talented enough to pull off that kind of attitude. If he weren't somehow writing to fill space between Bill James posts, he'd have a blog somewhere read by a single digit audience.
113. Hank Gillette Posted: December 03, 2020 at 02:15 PM (#5992277)
A reader points out the obvious, Sheffield is way ahead in offensive WAR or batting runs, but loses ground on the defensive side. Then Fleming goes off in the comments about how he knew all that and tells everyone that WAR is a broken car and it doesn't matter exactly why its broken, just that it doesn't run. Why is it broken? Well, apparently because Sheffield does not rate the way Dave Fleming thinks he should. What's his evidence that Sheffield's defense was not that bad? He's got none. Summary opinion without evidence, or as some would say, pure BS.

He is not nearly as critical of WAR in the reader forums. I thought his article was exceedingly dumb, when it obvious why Sheffield gets knocked down to the Vlad Guerrero and Bobby Abreu level of WAR. It would be one thing if he was criticizing WAR’s defensive calculations, or saying that WAR over-emphasizes defense (it does seem to weigh defense more heavily than Win Shares).

There are people who seem to feel obliged to suck up to Bill on his web site. One of the worst is someone that Bill obviously can barely tolerate, so it is not doing him any good.

I have no idea how Fleming got the writing gig on Bill James Online. I think he is okay, but nothing exceptional. Of the non-James writers, I think the best is Steven Goldleaf, although some of his articles are behind the paywall (I don’t currently have a subscription).
114. Hank Gillette Posted: December 03, 2020 at 03:13 PM (#5992296)
I don’t mean to pile on, but here is another example of James being cranky. In a tweet yesterday, he said:

For general information. . .there is no such thing as a "Hall of Very Good". It's just a really stupid expression.

Why is it a stupid expression? At least around here, people understand what it means: someone who had a really good career but doesn’t rise to one’s personal line for the Hall of Fame.
115. Srul Itza Posted: December 03, 2020 at 05:16 PM (#5992322)
If I remember correctly, it was kind of a back-handed compliment, because he said something like, “Oh, yeah, I should have come up with that.”

I did not read it that way. I thought it was more along the lines of of "D'oh, why didn't I see that?"
116. It's regretful that PASTE was able to get out Posted: December 03, 2020 at 09:59 PM (#5992358)
If I remember correctly, it was kind of a back-handed compliment, because he said something like, “Oh, yeah, I should have come up with that.” Or, maybe he truly meant that a compliment.

After describing Voros' Revelation in the New Historical Baseball Abstract, James listed four comments on it:

1. Like anything else, it's possible to take this TOO literally; the pitcher does have SOME limited influence on the results of balls in play.
2. With that qualification, Voros is clearly correct.
3. This is a very significant discovery, very useful.
4. I feel stupid for not realizing this 20 years ago.

The article concluded with "Thank you, Voros, for the edge." It was clearly complimentary.

Some years ago, on Bill James Online, amid a discussion of James' long held (and as far as I can tell, correct) belief that extra days of rest don't help prevent pitchers from hurting their arms, a reader pointed out that that may be true, but doesn't it make them more effective? James said "I doubt it," but then studied it and later published the results and concluded, yes, extra rest clearly DOES make pitchers more effective, and enough so that 5-man rotations generally are better than 4-man rotations (a reversal from two decades of advocating for a return to the 4-man rotation).

Bill James is no narcissist. What he is, has always been, will be until he dies (besides a tremendous writer), is an irrepressible contrarian. Whatever the accepted dogma is on any issue at any time, you can bet your last dollar he will be found doing his damnedest to advocate the opposite. He became famous advocating for statistical analysis when it was regarded as heresy; it's no surprise that once it became orthodox he turned on it.

But he has always been honest. If he studies something and the data demonstrates he was wrong, he publicly says "I was wrong, here is why." He's done it on his own site many times, the past 15 years.
117. It's regretful that PASTE was able to get out Posted: December 03, 2020 at 10:10 PM (#5992359)
One of my favorite little asides in the New Historical Baseball Abstract was his introduction to "The Greatest Team What Ever Was," which he explained was the working title of a book he was writing at the time, which book's basis was throwing 16 of the greatest single-season teams in history into a simulated season and describing the simulated games as they were played, culminating in the crowning of an all time champion.

By way of explaining why he ultimately shelved the project, he wrote that he did so because:

(a) You can't write a book saying the 1927 Yankees were the greatest team ever, because it's already been said too many times; and

(b) You can't write a book saying the 1927 Yankees weren't the greatest team ever, because they were.

That's who Bill James is: if he starts doing the legwork and realizes the conventional wisdom is probably correct after all, he loses interest, and moves on to something else. He's spent his life tirelessly looking for angles from which to attack Conventional Wisdom. On any issue.
118. Hank Gillette Posted: December 04, 2020 at 12:29 PM (#5992433)
The article concluded with "Thank you, Voros, for the edge." It was clearly complimentary.
I stand corrected. I either misremembered or misinterpreted it when I read it originally.
119. McCoy Posted: December 04, 2020 at 12:35 PM (#5992435)
Yes and no. See his 30+ year war with replacement level and linear weights.

And Pete Rose.

Not sure what his current stance is in regards to steroids.
120. Howie Menckel Posted: December 04, 2020 at 02:26 PM (#5992470)
my favorite fun fact about the '27 Yankees: they used 25 players, total, in that season.

the 2020 Dodgers used 38 players - in 60 games ! that included 21 pitchers.

the '27 Yanks by PA:
700+ PA - 2 Combs, Gehrig
600-699 - 2 Ruth, Lazzeri
500-599 - 2 BMeusel, Koenig
400-499 - 1 Dugan
300-399 - 1 Collins
200-299 - 2 Morehart, Grabowski
100-199 - 2 Gazella, Durst
0-99 PA - 3 Bengough, Paschal, Wera

all 15 hitters were age 22-31. the least-used player, Julie Wera (44 PA), appeared in at least one game in all six months.

the '27 Yanks by IP
200+ IP - 4 Hoyt, Moore, Pennock, Shocker
100-199 - 2 Ruether, Pipgras
50-99 IP - 1 MThomas
0-49 IP - 3 Shawkey Giard Beall

the 10 pitchers all were age 27-36. the least-used pitcher, Walter Beall, tossed an inning on May 30. everyone else made at least 16 appearances.

arrivals in 1928 - missing out by that much - included Dickey, Durocher, George 1B Burns, Coveleski, and Shocker. NYY used 30 players that year, with 4 of those being 'cups of coffee' to just 1 in 1927.
121. Rally Posted: December 04, 2020 at 02:46 PM (#5992474)
They were great but I'm not as impressed with the 27 Yankees as Bill was.

In 1987 I had 7 fictional teams in Microleague. Instead of having them only play each other, they a part of their schedule against MLB teams, and also a small part on a time travel road trip against some of the all-time great teams that came with the Microleague disk.

The championship team from my league, the LA Shockers, played against the 27 Yankees and dominated them. The game I remember is the Autobot pitcher Sandstorm striking out 18 against the Yankees. Ruth, Gehrig, and co. had no idea what to do with a slider from Cybertron.
122. snapper (history's 42nd greatest monster) Posted: December 04, 2020 at 02:52 PM (#5992475)
my favorite fun fact about the '27 Yankees: they used 25 players, total, in that season.

The Yankees played the 1977 World Series with a 24 man roster b/c they had traded someone in Sept.
123.  Posted: December 04, 2020 at 03:25 PM (#5992484)
That's who Bill James is: if he starts doing the legwork and realizes the conventional wisdom is probably correct after all, he loses interest, and moves on to something else. He's spent his life tirelessly looking for angles from which to attack Conventional Wisdom. On any issue.

You're certainly going in the right direction with all this, but the "any issue" isn't precisely right. Being a bit of a contrarian type myself, we're sort of hot-wired to be able to find the low-hanging fruit and I'm pretty sure James would say the same thing. The higher hanging fruit is the things the contrarian doesn't really touch, because conventional wisdom is pretty clearly right. So if you're an internet participant, you sort of read a bunch of things that you kind of nod your head "yep" at and don't write about, and then aim your sights at the clearly wrong stuff. You're still a contrarian at that point, but it reads much more contrarian than it actually is. Conventional wisdom is wrong sometimes, perhaps more than it should be, but it's nothing close to always wrong or even predominantly wrong. The secret is knowing which is which.

James became an obsessive weirdo about Rose, but big picture he's more wrong than right; and he's clearly right about WAR. He's been the "crank" he's correctly deemed to be now, pretty much since at least the time he wrote the first Abstract. If he wasn't a crank, he never would have devoted his life to debunking baseball bromides so aggressively. Those early Abstracts are pretty much a mountaintop of crankery.
124.  Posted: December 04, 2020 at 03:28 PM (#5992487)
They were great but I'm not as impressed with the 27 Yankees as Bill was.

Depends how you adjust for the fact that they played in a pure white boy league and not really an open, "major" league. It's impossible to seriously conclude that they could have competed in an integrated open league so well as to have been the best team ever, with whatever cross-era assumptions and normalizations one wants to make. If James never mentioned that obvious fact, he whiffed badly on the issue.
125. snapper (history's 42nd greatest monster) Posted: December 04, 2020 at 04:49 PM (#5992500)
Depends how you adjust for the fact that they played in a pure white boy league and not really an open, "major" league. It's impossible to seriously conclude that they could have competed in an integrated open league so well as to have been the best team ever, with whatever cross-era assumptions and normalizations one wants to make. If James never mentioned that obvious fact, he whiffed badly on the issue.

I would guess the impact would largely be on the marginal white players, not on the stars. Did the established white stars suffer a noticeable decline when MLB integrated? Just eyeballing some NL stars that were still in their prime (Musial, Kiner, Reese, Slaughter, Kiner) I don't see any.
126. Mefisto Posted: December 04, 2020 at 04:55 PM (#5992501)
Integration was a pretty slow process. By a reasonable measure, it took 20 years in the NL and nearly 30 in the AL. That would make it hard to find any effect, especially since aging would be a confounding variable.
127. Howie Menckel Posted: December 04, 2020 at 05:53 PM (#5992506)
The Yankees played the 1977 World Series with a 24 man roster b/c they had traded someone in Sept.

fwiw the 27 Yankees used 11 hitters and 4 pitchers in 4 games. the 77 Yankees used 13 hitters and 7 pitchers in 6 games.
128. It's regretful that PASTE was able to get out Posted: December 04, 2020 at 06:14 PM (#5992509)
James did mention it, and said about 40 times in that book that segregation was a gigantic reason why it's appropriate to make timeline adjustments in favor of more recent players when comparing them to long-ago players. He also mentioned in passing that the 1998 Yankees may have been the best team ever (he wrote the book in 1999) and has since reaffirmed that he believes they were. (Which is true; they were.)
129. Gonfalon Bubble Posted: December 04, 2020 at 07:19 PM (#5992513)
The 2003 Seattle Mariners used only five starting pitchers. Three of them started 32 times, the other two started 33 times.
130.  Posted: December 04, 2020 at 08:22 PM (#5992517)
Stating the obvious (my specialty :), every "greatest team" question is really two: greatest absolutely, or greatest relative to league? and the latter always involves acknowledging the strength of the league. The 1869 Red Stockings were unbeatable but played in a "league" of one and drew their players from white guys in the northeast and midwest of a nation the population of present-day Poland.
131. base ball chick Posted: December 05, 2020 at 12:04 AM (#5992540)
bill james is, to put it without swear words, a total jerk who thinks he knows it all and has contempt for anyone who is a non syncophant - reminds me of donnie poo, except he can speak in complete sentences. he is not capable of thinking things over once he has said something is SO, and he'll defend to his death his stupidity. i absolutely detest this guy. i don't get why so many of youse guys like him or even read him. yeh i know he introduced you to thinkig about baseball stats, um, differently. but cmon dudes. this is like a female sticking with a guy who beats her but she loves him anyhow in spite of all the broke bones because he brings her flowers just like he did when they first met. time to move on

win shares is not a good way to rate a player next to any other player not on his own team. in spite of what the stories all say, no one of 25 "puts the team on his back" and wins games all by his lonesome

not that WAR is some kind of perfect unquestionable number. i don't get where all those Adjustment Factors" come from and why 2 different groups of mathly gifted folks have such different conclusions.
132. Hank Gillette Posted: December 05, 2020 at 02:10 AM (#5992546)
bill james is, to put it without swear words, a total jerk who thinks he knows it all and has contempt for anyone who is a non syncophant - reminds me of donnie poo, except he can speak in complete sentences. he is not capable of thinking things over once he has said something is SO, and he'll defend to his death his stupidity. i absolutely detest this guy.

Come on, BBC, tell us what you really think!

I think it was easier to like Bill when he was just a writer that I never expected to meet (and of course, this was before the internet). He actually looked at things baseball people had been saying for years and questioned them, like “Pitching is 75% of baseball.” He pointed out that while people might say that, they didn’t actually act like it. Pitchers did not receive 75% of baseball salaries, and were not the most sought after players.

James questioned a lot of the conventional wisdom and did it in an engaging way. He pointed out that OBP and SLG were much more important than AVG, and questioned the value of stolen bases, and the existence of clutch hitting (as a repeatable skill).

He wrote in a very conversational tone that made you feel like you were in a (one-sided) conversation with a guy who lived and breathed baseball and had obviously thought a lot about it. He said uncomplementary things about a lot of other people, but back then, it was people who were part of the establishment who had a lot more power than he did. He still does that, but now that he has become part of the establishment, that trait is a lot less appealing.

Sometimes you shouldn’t actually meet people you admire, even online.
133. Mefisto Posted: December 05, 2020 at 09:19 AM (#5992556)
I haven't read James in probably 20 years, but his Abstracts were a revelation to me about how to think about the game. I met him in person at a SABR Convention in the mid-80s and he was friendly and generous in discussing anything I brought up.

From stuff that gets posted here, it seems as if he's become much more cranky and intolerant since I last read him.
134. McCoy Posted: December 05, 2020 at 09:24 AM (#5992558)
Or the people reading him have.
135. Ron J Posted: December 05, 2020 at 01:15 PM (#5992574)
#133/134 Worth reading "The Mind of Bill James" to get a handle on this. He had the reputation of being difficult to work with going back to the early 80s. He had absolutely no use for an editor and spelled this out plainly to his new copy editor (the author of the book in question -- so over time they built up a relationship). And from time to time this shows in his work.

But he's obviously not cranky and intolerant all the time. I've only met him once (briefly) and he sure came off as an introvert. From what I've read he doesn't readily extend trust and that prickly exterior is a defense mechanism.
136. McCoy Posted: December 05, 2020 at 01:56 PM (#5992589)
Sure, my point is that his audience has aged and became more confident in their new baseball knowledge. The pupil believes he's become a master and the relationship changes.
137. Srul Itza Posted: December 05, 2020 at 02:02 PM (#5992593)
my favorite fun fact about the '27 Yankees: they used 25 players, total, in that season.

Even that understates it -- one of the 25 was a pitcher who pitched one inning, in one game, and never came to bat.

So it was really a 24 man roster for the whole season, for all practical purposes.
138. Howie Menckel Posted: December 05, 2020 at 02:13 PM (#5992596)
the least-used pitcher, Walter Beall, tossed an inning on May 30.
139. sunday silence (again) Posted: December 05, 2020 at 02:19 PM (#5992598)

Sure, my point is that his audience has aged and became more confident in their new baseball knowledge. The pupil believes he's become a master and the relationship changes.

OK but its not some delusion on the part of the students. THeir master definitely has his weak spots. I could never figure out why he started cited MVP voting when he was doing his rankings of the all time players by position. The whole pt. of the book is to get away from perceptions and use objective means to rank these guys. Bringing in the MVP voting is just like making a horse by committee you get the worst of both worlds. You lose the perfectly objective standard that you were using and now you've fudged it with subjective voting.

This would be perfectly obvious without advanced metrics

Or his insistence that KCR was really the better team in the 1985 world series. I think that was in one of the yearly abstracts and it was a childish argument it was based on stats from the 7 game series. The obvious response to that is Small Sample Size. But another way is to see that all those stats were fueled by the 11-0 game 7 which STL was already down to a 10% win expectancy by third inn. so most of these runs were scored in garbage time. Everyone had tuned out. That's not really much of a data pt. THe whole debate of which team is better is kind of absurd so why is an intelligent guy like James trying to take sides in a question that seems unanswereable. You dont need advanced metrics to see that.

He classifies BIll Mazeroski as a "class C" hall of famer. Which I guess he was trying to do what we all do nowadays and rank guys as "inner circle" or "compiler" or "great peak" or whatever. I guess that's a fair thing. One of the criterion was "was he the major star on the team?" And in a sense I guess Maz wasn't. But OTOH until you get a handle on how valuable his fieldling was you really cannot say how much value he has. Moreover the whole HIstorical Abstract seems to be put together without any attempt to make some sort of guess as to fielding values. You know, even if you cant quantify it did they keep running the guy out there at SS? Did the team have success? etc. YOu could have made the attempt. There are things that can't be counted that still matter (Einstein).

It was doubly silly cause James had been out there in the forefront explaining that while fielding range has issues, you can't just wave it away on the idea that some guys are getting more balls hit to them. if that was the case than you'd simply rearrange your positioning. He made a great argument that if Sid Bream's value is based on his fielding, well why is Bonilla still making thirty errors over there at third base? All of this was perfectly objective. Then he gets all weirdo and decides that he's gonna measure all the SS and 3b by how they hit...

Anyhow you don't need to be old and cranky to see that sort of stuff. I don't buy the idea that we as readers have aged. His faults were there from the get go.
140. Ron J Posted: December 05, 2020 at 03:11 PM (#5992602)
#139 And in the "Breaking the Wand" essay he discusses how he found himself writing an increasing number of "Dear Jackass" letters. And at various points in the Abstracts acknowledges that he could get testy.

It's no big thing. I think that since there's less groundbreaking stuff our eyes are simply more drawn to the other stuff.
141.  Posted: December 05, 2020 at 03:25 PM (#5992603)
I could never figure out why he started cited MVP voting when he was doing his rankings of the all time players by position.

Because contemporary opinion is an important data point for people we didn't first-hand observe. Or even for those we did.
142. sunday silence (again) Posted: December 05, 2020 at 03:31 PM (#5992604)
thats classic, you utter a statement thats true on its face and yet you ignore entire point Im making there.
143. Rally Posted: December 05, 2020 at 03:59 PM (#5992607)
Looking at the Historical Abstract (original) right now, and I don’t see it as ranking players by batting with no consideration for defense. Brooks Robinson is #3 for 3B in career value, ahead of George Brett. Aparicio and Maranville rank 5 and 7 at short, ahead of Reese and Boudreau. I don’t think you can get there without some consideration of defense.
144. epoc Posted: December 05, 2020 at 04:13 PM (#5992613)
Re: Earl Wilson v. Frank Robinson for 1966, I wonder if someone could explain something to me.

According to BB-Ref, Robinson created 74 runs above replacement (total for hitting, fielding, position, etc.), which translates to 7.7 wins. Wilson created 63 runs above replacement, which translates to 7.8 wins (or 7.7 if you look at the 1966 WAR leaderboard instead of at Wilson's player page). How do Robinson's 74 RAR earn the same number of wins as Wilson's 63?
145. alilisd Posted: December 05, 2020 at 04:39 PM (#5992625)
The championship team from my league, the LA Shockers, played against the 27 Yankees and dominated them. The game I remember is the Autobot pitcher Sandstorm striking out 18 against the Yankees. Ruth, Gehrig, and co. had no idea what to do with a slider from Cybertron.

Clearly a small sample size blip ;-)
146. alilisd Posted: December 05, 2020 at 04:42 PM (#5992626)
How do Robinson's 74 RAR earn the same number of wins as Wilson's 63?

Wilson had 3 Rbat, 1 Rbaser, and 8 Rpos, which added to his pitching runs of 63 gives a total of 75. Edit: Actually that's only his batting line with Detroit. Total was 5 Rbat and 12 Rpos for 17 runs. Ah, and I see what you're saying now. Under Pitcher value he shows 46 runs, so 63 in total. Perhaps the difference is in playing time credit as a pitcher? his 35 runs above replacement for Detroit are quantified as 4.5 WAR, and his 11 for Boston to 1.4 WAR. All told that's 5.9 WAR as a pitcher and 1.9 as a hitter, which does come out to 7.8 altogether.
147. sunday silence (again) Posted: December 05, 2020 at 04:48 PM (#5992632)
Brooks Robinson is #3 for 3B in career value, ahead of George Brett. Aparicio and Maranville rank 5 and 7 at short, ahead of Reese and Boudreau. I don’t think you can get there without some consideration of defense.

That's a fair pt. But how did he get there? Does he explain the reasoning? I guess I can pull out the book.
148. John DiFool2 Posted: December 05, 2020 at 05:00 PM (#5992639)
I could never figure out why he started cited MVP voting when he was doing his rankings of the all time players by position. The whole pt. of the book is to get away from perceptions and use objective means to rank these guys. Bringing in the MVP voting is just like making a horse by committee you get the worst of both worlds. You lose the perfectly objective standard that you were using and now you've fudged it with subjective voting.

His position there was that something may be lost by looking purely at the numbers, that those who were in position to actually see the players in question day in and day out, while certainly not paragons of Absolute Objectivity, might have seen things that an observer from decades later may miss. IOW he acknowledged that no system is going to be ideal, and that the subjective side must be acknowledged, since it will still be there even if (especially if in some cases) one completely discounts it. It was a position of humbleness, vs. the way that you're implying it is.

[This BTW is often what bugs me about the HoM--everyone seems to have their own Perfect System, and often {broad brush warning} seem unable to step out of their system to look at the big picture. I found this out one day when I dared suggest that the error bars on mid-80's defensive stats were sufficiently large that I simply couldn't conclude that Cal Ripken Jr. was 100% certain to be better that year {1984} than Ryne Sandberg, and got flamed to death for my temerity.]

But yeah his entire recap of the Royal's championship run was pure fanboyism. But still enjoyable reading on its own merits.
149. Rally Posted: December 05, 2020 at 05:04 PM (#5992640)
My understanding is Bill used his intuition to balance the relevant factors and come up with a ranking. And usually his intuition is pretty good. I don't think he quantified what Brooks Robinson's defense was worth, certainly didn't commit to any number, but somehow he factored it in.

The 2nd edition of the Historical Abstract came out about the same time as win shares, so he had that to go on for the updated book.
150. Rally Posted: December 05, 2020 at 05:16 PM (#5992647)
Bill's actual ratings are at the end of each position chapter, and relatively brief. You won't find a logical argument for why he ranked Brooks above George.

In the Schmidt comment, he does make the case why he ranks Mike #1. In the Brooks Robinson comment, he talks about Brooks ranking too low, only #108 for non-pitchers, and goes on to inadvertently make a strong case for WAR.
151. Howie Menckel Posted: December 05, 2020 at 05:18 PM (#5992648)
[This BTW is often what bugs me about the HoM--everyone seems to have their own Perfect System, and often {broad brush warning} seem unable to step out of their system to look at the big picture.

if I had a newsletter, I have a feeling I'd have another subscriber!
:)

24. Howie Menckel Posted: December 20, 2019 at 10:01 PM (#5910237)

The annual fine print:

Overall, I think there is a bit too much slavish devotion in some quarters re WAR, WAA and an ever-increasing number of acronyms, which are intriguing tools but which still may not yet be sufficiently mature (though they continue to improve).

So my fondness (but not blind allegiance by any means, especially where durability is an issue) for ERA+ and OPS+ helps, I think, as a reality check even as fielding issues are quite significant for pitchers/team defenses - and especially for hitters, of course.

I tend to be mostly prime-oriented with hitters, and prime and career with pitchers. But a huge peak sometimes catches my eye, and a remarkably long, effective hitting career also works for me.
152. What did Billy Ripken have against ElRoy Face? Posted: December 05, 2020 at 05:42 PM (#5992659)
the least-used pitcher, Walter Beall, tossed an inning on May 30.
His nickname had to be “End All,” right?
153. baxter Posted: December 06, 2020 at 01:21 AM (#5992702)
On Robinson v Brett, Robinson played more games at 3b than Brett played in his entire career. Robinson still the career leader (Beltre couldn't catch him). Brett played 1692 at 3b, so, about 11 seasons worth at 150 games per year (1981 a strike season; also, Brett missed games). Robinson arguably greatest fielding 3b ever; that has to count in favor against Brett who to move to 1st, although he DH's slightly more than he played 1st

Did James have Matthews at #2, called him "Mike Schmidt senior," fine played underrated at time. Matthews did not go in first ballot; took multiple ballots.
154. sunday silence (again) Posted: December 06, 2020 at 01:43 AM (#5992705)

But yeah his entire recap of the Royal's championship run was pure fanboyism. But still enjoyable reading on its own merits.

You know I pulled out "The Politics of Glory" after posting the above, because I couldnt find HIstorical Baseball Abstract in hopes of finding more guffaws. But I really couldnt. I was reading the part about Gerry Priddy and Phil Rizzuto and once James is locked into a subject, he's got all the anecdotes and the statistics, he's not making wild ass pronouncements, his writing is focused and its great. He was focused on telling the story of Priddy and how he got moved and then traded. ANd he sticks to the facts and the anecdotes and its great. I enjoyed every sentence of it.
155. sunday silence (again) Posted: December 06, 2020 at 01:52 AM (#5992706)
You won't find a logical argument for why he ranked Brooks above George.

See to me I always felt like that was illogical based on his premise. Like his starting point is to tell you that you cant just believe what people tell you, you have to dig into the numbers, you have to use logic. He would point out how this whole "pitching is 90% of baseball" is nonsense, cause you know off/def is a zero sum game. Or why it can't be that Buddy Bell is stealing ground balls from Frank Duffy or whomever. He would go through all this point by point.

And then at the end it just seemed like when in doubt place the guy who was the better hitter up there. Even if you dont know the defensive numbers, why not take a stab at them? Give the players a rating like a, b, c and how many runs can we expect to save with Brooks Robinson? How many with Eddie Mathews?

He just seemed to punt that whole issue. And that issue of defensive value is just towering over the whole thing.

To me Mathews was great for the first ten years, then he sort of tailed off the second half. Depending on how you account for defense its quite possible that Boyer was better in the second half of their careers. It seemed he just put Mathews at no. 2 cause HRs.
156. sunday silence (again) Posted: December 06, 2020 at 02:02 AM (#5992707)
The Brooks Robinson part had another weird thing now that I think about it. He starts to explain basically why we need something like WAR because you have to give a guy credit for holding down a position. So that average play does not equal Zero. I sort of get that. And then he mentions RObinson at the end of his career playing at far less than his prime. ANd he doesnt think its fair that that tail end should detract from the great part of his career. So he doesnt want to assign that negative value. I forget what Brooks had at the end something like 0.5 or 0.3 WAR seasons...

But why the hell not? What's the logic to that anyhow? If someone is playing at less than average value then surely its hurting your team if you could find an average player who would have more value there. You could just as easily come up with a system that counts those tail end seasons as -1.5 or -1.7 and then what would you have? Would BRobinson lose a few places?

So James is all completely objective and logical and every by the numbers, right up until the time he doesnt want to be. He doesnt want Brooks to have negative value so he makes it so a less than average season is still positive value.

That's what sort of rubbed me the wrong me. It's not logical. What's the point of being so logical when you just toss it out at the end and just make up a number so Brooks Robinson doesnt lose some places on the historical panthenon?
157. McCoy Posted: December 06, 2020 at 08:37 AM (#5992716)
Bill's ratings in the new historical come down to the unknowable intangible factor that he includes and it appears he screwed up his database. I think he admitted the date base screwup but I don't know if he admitted it was as widespread as it appears to be. I actually tried to replicate his work back in the day and discovered those two things about the ratings. All of his win shares steps were largely pointless.
158. sunday silence (again) Posted: December 06, 2020 at 09:14 AM (#5992720)
I thought the win shares thing was proprietary? or not?

Speaking of those. I can never understand them. Is it better to be 6-3 or say 5-2? or is 11-7 better than all of them? That's what always has me confused I dunno if Im supposed to look at the overall bulk or the ratio or what? Obviously WAR seems a lot more intuitive.
159. snapper (history's 42nd greatest monster) Posted: December 06, 2020 at 10:32 AM (#5992721)
Obviously WAR seems a lot more intuitive.

Yeah. How many ferwer games would we win if we had to put some random AAAA schlub at the position. That's super intuitive, and a real situation teams face.
160. FrankM Posted: December 06, 2020 at 10:32 AM (#5992722)
But why the hell not? What's the logic to that anyhow? If someone is playing at less than average value then surely its hurting your team if you could find an average player who would have more value there. You could just as easily come up with a system that counts those tail end seasons as -1.5 or -1.7 and then what would you have? Would BRobinson lose a few places?

I disagree. If Brooks retires a couple of years earlier than he did, that doesn't make him a better player. He's the same guy.
161. snapper (history's 42nd greatest monster) Posted: December 06, 2020 at 10:59 AM (#5992725)
I disagree. If Brooks retires a couple of years earlier than he did, that doesn't make him a better player. He's the same guy.

But it affects his counting stats, which benefit a HoF case. Albert Pujols only gets to be a ~675 HR guy because he's gonna play 5 seasons after he should have been benched. If you get those 100 HRs, you get the -7 WAA too.

If a player has those bad seasons at the front of the career we count them. They should count at the back too.
162. baxter Posted: December 06, 2020 at 11:30 AM (#5992728)
155 re boyer/matthews, what is the basis for saying 2nd half of career favors boyer?
Their careers overlapped, w/matthews starting earlier (he has a couple seasons more games than boyer).
Matthews was a great offensive player and decent fielder; he didn't have to get moved off 3rd like Brett did. Also, the difference is "just the HR's," that's 230 of them, while playing half the time in County stadium milwaukee; that counts for quite a bit, however one calculates it.

Matthews' offensive comps (aside from #1 Schmidt, appropriately enough) and Mantle, are mainly 1b's and corner OF's (Banks also in there; Killebrew is a combination).

163. Eric J can SABER all he wants to Posted: December 06, 2020 at 11:35 AM (#5992730)
But it affects his counting stats, which benefit a HoF case. Albert Pujols only gets to be a ~675 HR guy because he's gonna play 5 seasons after he should have been benched. If you get those 100 HRs, you get the -7 WAA too.

Looking at Pujols specifically, he was going to be a Hall of Famer whether he ever played a game for the Angels or not.

Taking a more borderline case - say, Craig Biggio, who might not have gotten in (and almost certainly wouldn't have gotten in on the second ballot) unless he reached 3000 hits. From 2002-07, he had -6.7 WAA (4.6 WAR). Is he a Hall of Famer without those years? He probably deserves to be, at least; I'm reluctant to fault him for hanging around long enough to reach a number that matched how good he was in his prime.

My take on negative seasons (especially negative WAR seasons) is that I generally leave them out of consideration when evaluating a player's entire career. There are three basic situations that can result in a negative WAR: First, WAR is wrong about how good the player is, and he was actually above replacement level (in which case he shouldn't actually have a negative score). Second, WAR is correct about how good the player is, but his team didn't have a better alternative available at his position. In that case, while he's below conceptual replacement level, he's not below the actual replacement level available to his team. Third, WAR is correct and the team did have a better alternative, but didn't use the alternative for whatever reason. This is generally the team's fault, not the player's fault. (In Biggio's case, the Astros chose to let him run up -2.1 WAR while chasing 300 hits in 2007 - and then kept playing him for 60 more hits even after that. Or if you look at players who came up in the '50s, some of them were rostered for years when they should have been in the minors due to the bonus baby rule - Harmon Killebrew loses 0.6 WAR and 1.4 WAA as a benchwarmer for five years before becoming a regular.)
164.  Posted: December 06, 2020 at 11:38 AM (#5992731)
Good read (talking about the comments in this thread--- currently under quarantine, so have a bit of free time to catch up on this site and others that I didn't have time to before)

I think Bill James is a victim of the harshness of internet language. What I mean is that when you write a book you have multiple times to do edits and clear up your speech to make it more 'palatable' for everyone, as a person who has many times wrote passages on the internet that I wish I could edit at a later time, it feels like he has this issue also, where he writes what he wants, maybe takes a single pass at it, and publishes it, not really realizing that the tone might be a bit harsher than he intended or in the case of this article, not fully formed or checked.

Just looking at the passage where he talks about Robinson vs Wilson, and the fact that he misrepresented the two versions of war on how they present war (in fact not only did he misrepresent the two, he didn't even notice that his complaint would have actually helped his argument if bWar would have been using it, as Wilson actual runs allowed was lower than his fip) And add in, that even in Bill James version of Win Shares, he admits that there is a margin of error of about 3 wins, so you are to take differences between the two and adjust according to your knowledge.... The same is true with War, it flat out states on the website that there is a margin of error of 1, anything within that range is pretty debatable.

As someone mentioned above, what should be people talking about how great of a season that Earl Wilson had, a relatively unknown pitcher who had a pretty good stretch of excellence where in one season he was close to being as good as a guy who won the triple crown, has ended up being a discussion as what is wrong with War (and in this case, I don't see anything wrong with it... Frank Robinson still gets my MVP vote, but Wilson might slide up higher on my imaginary ballot than 14th)

As far as replacement level, it's just a term, get over it SB, it's a baseline used for statistical purposes. We could do zero if we wanted, we could do .500 (like WAA does) but ultimately it's just a statistical baseline to bring the numbers into a somewhat acceptable format. Same reason why Bill James multiples by 3 for win shares, it just makes the numbers look more presentable. Having a commonly held baseline between the two primary producers of war at least makes it obvious the differences in their system and makes the discussion about them on the same scale, the fact that win shares has a completely different one means it's a bit harder to include it on a comparison discussion.

I've complained plenty of time about piling estimates on estimates between war or fip or cs/sb or catchers defense etc.... there are legit arguments to make about any and all estimate systems, it doesn't mean you throw them out, you look at how much you trust the estimates, and acknowledge it when you talk about the stat(every time, don't just assume the reader has read your acknowledgment previously) and you work on improving the estimates when you have time. The confidence level I have for most of the plug ins for war, puts me at a pretty comfortable level, nobody should use it as the only number in a discussion, but it's rightfully going to be one of the primary numbers listed in any rational discussion.
165.  Posted: December 06, 2020 at 11:47 AM (#5992732)
Taking a more borderline case - say, Craig Biggio, who might not have gotten in (and almost certainly wouldn't have gotten in on the second ballot) unless he reached 3000 hits. From 2002-07, he had -6.7 WAA (4.6 WAR). Is he a Hall of Famer without those years? He probably deserves to be, at least; I'm reluctant to fault him for hanging around long enough to reach a number that matched how good he was in his prime.

My take on negative seasons (especially negative WAR seasons) is that I generally leave them out of consideration when evaluating a player's entire career. There are three basic situations that can result in a negative WAR: First, WAR is wrong about how good the player is, and he was actually above replacement level (in which case he shouldn't actually have a negative score). Second, WAR is correct about how good the player is, but his team didn't have a better alternative available at his position. In that case, while he's below conceptual replacement level, he's not below the actual replacement level available to his team. Third, WAR is correct and the team did have a better alternative, but didn't use the alternative for whatever reason. This is generally the team's fault, not the player's fault. (In Biggio's case, the Astros chose to let him run up -2.1 WAR while chasing 300 hits in 2007 - and then kept playing him for 60 more hits even after that. Or if you look at players who came up in the '50s, some of them were rostered for years when they should have been in the minors due to the bonus baby rule - Harmon Killebrew loses 0.6 WAR and 1.4 WAA as a benchwarmer for five years before becoming a regular.)

Sometimes it's probable the team is receiving a higher financial reward for playing these guys also, the Reds and Expos most probably did with Pete Rose, and it's not like in Biggio's case that Chris Burke or Eric Bruntlett would have been a clear better option. (they might have had a better year that year, but looking at their career, that was an exception for them, and it's hard to believe that Burke was playing at his real level that season)

I'm in the camp that I do add up the counting numbers, but don't include negative war or waa even for players on their last year of a hovg (or better) career. Now of course if I'm comparing Willie Mays to Albert Pujols, yes then those years matter, but when I'm looking at the hof argument and the in/out line, then they don't matter.
166. Eric J can SABER all he wants to Posted: December 06, 2020 at 11:53 AM (#5992734)
In Biggio's case, the Astros chose to let him run up -2.1 WAR while chasing 300 hits in 2007

Missed the edit window. Obviously this should have said "chasing 3000 hits."
167. DL from MN Posted: December 06, 2020 at 12:00 PM (#5992736)
Hall of Merit ran an election for 1966. Frank Robinson won, Earl Wilson didn't get a vote. He was my #4 pitcher after Marichal, Koufax and Bunning.
168.  Posted: December 06, 2020 at 12:21 PM (#5992742)
Hall of Merit ran an election for 1966. Frank Robinson won, Earl Wilson didn't get a vote. He was my #4 pitcher after Marichal, Koufax and Bunning

How much of that happened because people overlooked his batting? As a pitcher, I don't see any reason for him ahead of those three, (or even Kaat or Gibson to be honest) (obviously you didn't overlook it, just curious how many downplayed it or ignored it? and of course that is both leagues combined, outside of Robinson how many other players from the AL even made the list?)
169. epoc Posted: December 06, 2020 at 01:12 PM (#5992748)
Adding to the Q in 144, I looked at the RAR-to-WAR conversion for some other players for 1966, and I can't make any sense of it:

Robinson, F 74 RAR, 7.7 WAR, 9.6 runs/win
Robinson, B 45 RAR, 4.6 WAR, 9.8 runs/win
Wilson, E 63 RAR, 7.8 WAR, 8.1 runs/win
Yastrzemski 52 RAR, 5.5 WAR, 9.5 runs/win
Stange, Lee 17 RAR, 2.1 WAR, 8.1 runs/win
Kaline 54 RAR, 5.6 WAR, 9.6 runs/win
Freehan 15 RAR, 1.2 WAR, 12.5 runs/win
McLain 17 RAR, 2.3 WAR, 7.4 runs/win
Marichal 88 RAR, 9.8 WAR, 9.0 runs/win
Koufax 87 RAR, 9.7 WAR, 9.0 runs/win
Mays 82 RAR, 9.0 WAR, 9.1 runs/win
Aaron 72 RAR, 7.9 WAR, 9.1 runs/win

If anyone can make any sense of this, I would be really curious to know the explanation. The NL seems to be pretty straightforward: Marichal, Koufax, Mays, and Aaron all have a conversion rate of 9 runs/win. But the AL is all over the place. Even hitters on the same team (Kaline and Freehan) have radically different conversion rates. Compare Earl Wilson's hitting conversion with only DET to Freehan's:

Wilson, E 12 RAR, 1.3 WAR, 9.2 runs/win
Freehan 15 RAR, 1.2 WAR, 12.5 runs/win

Why would a hitter with *fewer* RAR be worth *more* wins than a hitter with more RAR on the same team? This is just the conversion of RAR to WAR, so it has nothing to do with how the RAR are accumulated in the first place.
170.  Posted: December 06, 2020 at 01:22 PM (#5992756)
Why would a hitter with *fewer* RAR be worth *more* wins than a hitter with more RAR on the same team? This is just the conversion of RAR to WAR, so it has nothing to do with how the RAR are accumulated in the first place.

Without looking too deeply into it, at a glance I would say it has to do with playing time. Either innings played or plate appearances.

Edit:That is a question that Rally would need to answer to be honest though, maybe Walt.
171. epoc Posted: December 06, 2020 at 01:36 PM (#5992758)
Without looking too deeply into it, at a glance I would say it has to do with playing time. Either innings played or plate appearances.

Compare Freehan to Kaline to Willie Horton:

Kaline 572 PA, 9.6 runs/win
Freehan 544 PA, 12.5 runs/win
Horton 578 PA, 10.9 runs/win

Same team, similar playing time, radically different RAR:WAR conversion.
172. Eric J can SABER all he wants to Posted: December 06, 2020 at 01:36 PM (#5992759)
Wilson, E 12 RAR, 1.3 WAR, 9.2 runs/win
Freehan 15 RAR, 1.2 WAR, 12.5 runs/win

Why would a hitter with *fewer* RAR be worth *more* wins than a hitter with more RAR on the same team? This is just the conversion of RAR to WAR, so it has nothing to do with how the RAR are accumulated in the first place.

I think WAR accounts for a pitcher's effect on his own run environment? So, Wilson as a good pitcher made the run environment lower and his hitting would therefore be more valuable. Not sure if they actually account for that in evaluating pitcher hitting though.
173.  Posted: December 06, 2020 at 01:54 PM (#5992772)
Kaline 572 PA, 9.6 runs/win
Freehan 544 PA, 12.5 runs/win
Horton 578 PA, 10.9 runs/win

Same team, similar playing time, radically different RAR:WAR conversion.

Not sure what is going on, it's possible that there is playing time issues (as both Horton and Freehan played more innings than Kaline---roughly about 6-8 games more) or other issues.... I'm with you, I would like to know the difference but honestly it really does look like those 60-80 extra innings as a defender are figuring into this.

Edit: mind you Freehan's is a bit high using that assumption, so not sure that explains all of it.

Although if you look at it, Freehan had 135 games started, and 131 complete games, Horton had 133 games started (137 total) and 120 complete games, while Kaline had 136 games started and 98 complete games
174. sunday silence (again) Posted: December 06, 2020 at 01:57 PM (#5992773)

155 re boyer/matthews, what is the basis for saying 2nd half of career favors boyer?

HI Baxter. It's been awhile since I looked at that and I was convinced by the fielding numbers I looked at. I promise to revisit this as I continue my series of posts on Egregious Throwing Errors under thread labeled: Is the Hall leaving out too many players? Today's subject Jim Ray Hart so you'll have to wait a bit.
175.  Posted: December 06, 2020 at 02:05 PM (#5992776)
155 re boyer/matthews, what is the basis for saying 2nd half of career favors boyer?

Facts.... It's not huge, but no matter how you slice it, Boyer post 30 was the better player.
176. sunday silence (again) Posted: December 06, 2020 at 02:10 PM (#5992778)

My take on negative seasons (especially negative WAR seasons) is that I generally leave them out of consideration when evaluating a player's entire career.

I mean I do the same thing, probably most of us do. I usually take the prime seasons e.g. Boyer had 7 prime seasons averaging 7 WAR (wotever it was I cant recall) plus 4 or 5 pretty good seasons say 20 WAR and then anything less than 2 WAR just forget it or add a footnote like he was hurt, or the war, or could still maybe PH...

I was basically just piling on Bill James cause the Logic of it didnt make a lot of sense, which sort of contradicts the premise of his approach. In terms of sentiment and probably in terms of HoF yeah sure, I can usually just omit it.
177. sunday silence (again) Posted: December 06, 2020 at 02:15 PM (#5992780)
Facts.... It's not huge, but no matter how you slice it, Boyer post 30 was the better player.

CFB do you remember the discussion we had a year or two ago? You were more skeptical of my Boyer arguments vis a vis Santo. Did you know James was actually plumping for both of them for HoF? He actually has Boyer at 19th all time 3b here, but I think he's higher on another James list:

https://www.billjamesonline.com/wrapping_up_third_base/

Also I think I under appreciated Santo's fielding, hopefully I will revisit that.
178.  Posted: December 06, 2020 at 02:24 PM (#5992782)
CFB do you remember the discussion we had a year or two ago? You were more skeptical of my Boyer arguments vis a vis Santo. Did you know James was actually plumping for both of them for HoF? He actually has Boyer at 19th all time 3b here, but I think he's higher on another James list:

As a Cardinal fan, I intentionally try to make myself more sceptical of pro-Cardinal arguments... I don't always succeed, but anytime I see a pro-Cardinal argument, I try to force myself to see differently. (again I rarely succeed at that, but I try)

I'm on the fence for Boyer, simply because in my heart I think he belongs, but I know that is colored by 30 years of growing up and hearing he belongs. I pushed for Santo because he was clearly worthy, to the level that it was absurd he was not in. With Boyer, he's closer to Rolen territory for me and having a well known Pro-Cardinal bias, I can't push too hard for either until my confidence level is higher (same with Molina when he becomes eligible.... I think he's worthy but the numbers don't really support it, and I have to admit I'm bias, and that to talk about Molina without including Sundberg in the discussion is idiotic on any level)
179. snapper (history's 42nd greatest monster) Posted: December 06, 2020 at 02:38 PM (#5992785)
You were more skeptical of my Boyer arguments vis a vis Santo. Did you know James was actually plumping for both of them for HoF?

To me, both belong, along with Nettles and Bell.
180.  Posted: December 06, 2020 at 03:08 PM (#5992789)
I know some people might look at Rolen and Santo relatively equal war and argue that if I think Santo is clearly over the line, then Rolen should be, and my response is ... No... Do I think they are both over the line, yes, but I think it's pretty clear that even with the same war, that Santo is absolutely the superior player. Santo's health wins it for me... Santo played 15 seasons, 12 of those seasons he played 149 games or more (heck 7 of those he played 160 or more and another he played 154 in a 156 team game season) the other three years of his career was his rookie year where he played 95 games and a couple of his last seasons where he played 133 and 117... games and retired...

Look at Rolen, similar career war (70.5 for Santo vs 70.1 for Rolen) Rolen played in 149 or more games 5 times, over 140 7 times he needed 17 years to reach 8518 pa, while Santo did 9397 in 15 seasons... Santo was more reliable player, didn't need as much rest or was less injury prone etc. Rolen rate numbers are better, but it's absolutely clear to me that Santo was the better overall player throughout their career.

I've mentioned it many times on any Larry Walker thread, but health is absolutely a skill that should be considered in hof discussions. When a team builds itself around a great player, that great player's health is a major factor in how well they will contend.
181. JJ1986 Posted: December 06, 2020 at 03:40 PM (#5992791)
Just going by WAR, Santo had 3 MVP level seasons and Rolen had 1; Santo has six seasons above 6 WAR and Rolen has 3. I think JAWS does enough of the work that we don't really need a more exact stat, but adding up a player's WAR isn't enough to determine career value since a 9-WAR season is worth significantly more than three 3-WAR seasons.
182.  Posted: December 06, 2020 at 04:19 PM (#5992797)
Just going by WAR, Santo had 3 MVP level seasons and Rolen had 1; Santo has six seasons above 6 WAR and Rolen has 3. I think JAWS does enough of the work that we don't really need a more exact stat, but adding up a player's WAR isn't enough to determine career value since a 9-WAR season is worth significantly more than three 3-WAR seasons.

Agreed, even though Rolen has a better waa, I think it's pretty clear that Santo was the better hof candidate.
183. baxter Posted: December 06, 2020 at 05:19 PM (#5992806)
174 Look forward to seeing your Mathews commentary; I remember Hart as being a good hitter, but could not field very well.

175 On Boyer/Mathews what facts? They are born the same year; Matthews is a few months younger; from 1961 to the end of their careers, Mathews, for example, has more more WAR 36.8 to 31.5. That's 8 seasons, roughly half of Mathews' 17 seasons.

Boyer was a fine player; apt comparison made to Santo, w/Santo being more valuable b/c of consistency.

Mathews is inner circle. Double the 2nd half of his career, he still would be HOF'er.
184. Rally Posted: December 06, 2020 at 06:31 PM (#5992812)
‘Edit:That is a question that Rally would need to answer to be honest though‘

I’m just as stumped as everyone else on this.
185.  Posted: December 06, 2020 at 06:53 PM (#5992814)
175 On Boyer/Mathews what facts? They are born the same year; Matthews is a few months younger; from 1961 to the end of their careers, Mathews, for example, has more more WAR 36.8 to 31.5. That's 8 seasons, roughly half of Mathews' 17 seasons.

Generally the assumption is that post 30 is second half of a players career, considering both were either out of the league by their age 36 season useless there after, the concept is to look at either age 30-to whenever or age 31....

Going by War You have Mathews from age 30 to the end of his career at 28.3... Boyer at 31.4. If you move the bar to 31 to end of the career you get 22.7 vs 23.4 Boyer also had more pa, innings played etc.... there is literally no way to make an argument that Mathews had a better late career than Boyer.
186.  Posted: December 06, 2020 at 06:59 PM (#5992817)
175 On Boyer/Mathews what facts? They are born the same year; Matthews is a few months younger; from 1961 to the end of their careers, Mathews, for example, has more more WAR 36.8 to 31.5. That's 8 seasons, roughly half of Mathews' 17 seasons.

you are including Mathews age 29 and 30 season in the discussion while ignoring Boyer's age 29 season. I get that argument on the face of it, since they were born in the same year, but ultimately you are including a year or two of Mathews (and Boyer's even) that nobody has historically considered to be the second half of their career, second half isn't generally denoted as half the plate appearances of their career, it's generally considered what they do after they turn 30.
187. snapper (history's 42nd greatest monster) Posted: December 06, 2020 at 07:16 PM (#5992820)
there is literally no way to make an argument that Mathews had a better late career than Boyer.

That's sort of irrelevant isn't it? Mathews put up a HoF career before he hit 30.
188.  Posted: December 06, 2020 at 07:38 PM (#5992823)
That's sort of irrelevant isn't it? Mathews put up a HoF career before he hit 30.

Not debating that, just replying to the question that was posted above. Mathews is absolutely a hofer, more so than Boyer... .I don't think anyone would make that argument, the discussion is that Boyer had a better second half to his career than Mathew. I don't know where the discussion came from, it was quoted above and I'm replying to that. I think there are three things that are accurate. 1. Mathews is an inner circle hofer. 2. Boyer is a borderline debatable and 3. Boyer had a better second half to his career.

None of this implies Boyer was better, nor does this say that Boyer is a deserving hofer, it says that basically from age 31 on, Boyer was the better player, simple as that.
189. Howie Menckel Posted: December 06, 2020 at 07:43 PM (#5992825)

Hall of Merit 3Bs, ranked
this was 12 years ago, when we had only elected 18 3BS at the time

Mathews, Santo, and Boyer are among them (as are Allen, Evans, and Nettles)
190. baxter Posted: December 06, 2020 at 08:18 PM (#5992829)
189 thank you; I see one of the votes placed Matthews 6th behind Schmidt, Brett, boggs, baker and santo
191.  Posted: December 06, 2020 at 08:38 PM (#5992830)
189 thank you; I see one of the votes placed Matthews 6th behind Schmidt, Brett, boggs, baker and santo

It's an outlier, I'm not sure that I would take it too seriously, he ranked it because of peak, but ignores the fact that Mathews peak was in 154 game seasons and to be honest he used an arbitrary number of 5 and 7 seasons, when Mathews peak was 9 seasons of extreme high consistent play. Jaws was created after this thread, and there is a reason why Mathews is second there also...

In reality, there is absolutely no real argument for Brett, Boggs or others ahead of Mathews, Schmidt yes... after that, there really isn't a discussion. If you put Arod as a third baseman, then you have an argument for second greatest of all time, until then it's not really close.

(Edit: that was over the top, if you want to put Boggs, Beltre or Brett in the discussion I can actually see it, but I think that ultimately Mathews wins out the argument individually against each one... Sean was locked into a 5 and 7 year peak in that discussion, and Mathews had a 9 year peak)
192. Howie Menckel Posted: December 06, 2020 at 08:59 PM (#5992832)
no problem, Baxter

I see I failed to include my usual "fwiw" - important, because I'm not claiming "HERE IS THE FINAL ANSWER" in any way.

'for what it's worth' means different things to different people, and of course I am not locked into agreeing with exact orders or anything.
193. Zach Posted: December 07, 2020 at 05:08 PM (#5992954)
On the negative value issue, I think the best policy is to ignore it:

First, most players with a regular job are contributing *something*. Being good enough to receive playing time should usually be a positive or neutral factor in a Hall of Fame argument.

Second, everybody sucks at the end of their career. Early retirement vs eking out a couple of extra years is a matter of personal preference and shouldn't affect a HOF vote unless it's really egregious. I mean, how long did Rickey Henderson try to keep playing? The man just liked baseball.

Third, it doesn't seem to be much of a factor in HOF voting. I'm not aware of anyone who has played himself out of the HOF, or any voting constituency for dropping people from the ballot because of a few subpar years at the tail end of the career.

Finally, my basic approach to the HOF is that everybody gets to make their own case and put their own best foot forward. That means that the argument focuses on the good things a player did, rather than precisely accounting for the small change.
194. snapper (history's 42nd greatest monster) Posted: December 07, 2020 at 05:38 PM (#5992959)
I mean, how long did Rickey Henderson try to keep playing? The man just liked baseball.

Rickey, because of the awesomeness that is Ricky, was never really bad. The sum total of his negative WAR seasons at the end was -0.7.
195.  Posted: December 07, 2020 at 11:27 PM (#5993022)
I'm not aware of anyone who has played himself out of the HOF, or any voting constituency for dropping people from the ballot because of a few subpar years at the tail end of the career.

It's obvious he didn't, but in a few articles in the 70's there was talk about how Mays was hurting his hof chances.... it wasn't many, and it was something memorably stupid enough that Bill James pointed to it, but yes there has at least been articles talking about how a particular player might hurt their hof by their late career... mind you obviously it's one or two writer perception issue over the electorate.
196. Rally Posted: December 08, 2020 at 09:28 AM (#5993046)
On Mathews vs. Boyer for second half of their careers: Not something I've ever pondered before this thread, but looking at them I'll call it a tie.

Going by seasonal age, Boyer had a slight edge from age 30 on. But it all comes down to how you define things. Seasonal age makes Boyer one year older, but they were born in the same year and he's actually 5 months older.

Both were still star level players to age 33. Both had an 8 WAR season at age 30 or 31. At age 33 Boyer had 6.1 WAR, Mathews 5.4. Neither had a significantly above average season after age 33. Both had their last seasons as regulars at age 35. Mathews was done at 36, Boyer had 3 years as a part timer from age 36-38.

Too close to call. It's a tie. If I had traded one for the other at age 30 (assuming a working flux capacitor, since their seasonal age 30 was a year apart) and ended up with these results, I wouldn't regret the deal no matter which side I ended up with.

Assuming no time travel and trading them in a specific year, here's their future WAR:

1961: Ed 36-31
1962: Ed 28-23
1963: Ed 23-18
1964: Ed 15-13
1965: Ed 10-7
1966: Tie 5-5
1967: Tie 2-2
1968: Tie 0-0

Slight edge for Eddie, but not a mismatch.
197. CFBF's Results are Certified Posted: December 08, 2020 at 10:02 AM (#5993051)
Rickey, because of the awesomeness that is Ricky, was never really bad. The sum total of his negative WAR seasons at the end was -0.7.

Rickey's ability to keep drawing walks at the end of his career when he had nothing left in his bat and couldn't have scared anyone is really something else.
Page 2 of 2 pages  < 1 2

You must be Registered and Logged In to post comments.

<< Back to main

### Support BBTF

Thanks to
Backlasher
for his generous support.

You must be logged in to view your Bookmarks.

### Hot Topics

NewsblogCollege baseball pitcher Sang Ho Baek dies after complications from Tommy John surgery
(19 - 1:17am, Jun 22)
Last: Tulo's Fishy Mullet (mrams)

NewsblogSo no one told you OMNICHATTER! was going to be this way, for June 21, 2021
(27 - 12:26am, Jun 22)
Last: Howie Menckel

Sox TherapyAnd Breathe
(28 - 11:34pm, Jun 21)
Last: the Hugh Jorgan returns

(1856 - 11:03pm, Jun 21)
Last: Dandy Little Glove Man

NewsblogOT Soccer Thread - In Which Euro 2020 Is Played in 2021
(248 - 10:51pm, Jun 21)
Last: AuntBea odeurs de parfum de distance sociale

NewsblogBaseball Reference Adds Negro Leagues Statistics, Rewriting Its Record Book
(162 - 10:32pm, Jun 21)
Last: sunday silence (again)

NewsblogRays call up top prospect Wander Franco
(6 - 9:07pm, Jun 21)
Last: Walt Davis

NewsblogEmpty Stadium Sports Will Be Really Weird
(13254 - 8:48pm, Jun 21)
Last: Never Give an Inge (Dave)

NewsblogThe story of honoring Negro League history and a search for buried treasure
(1 - 8:31pm, Jun 21)
Last: Starring Bradley Scotchman as RMc

NewsblogJuneteenth Weekend OMNICHATTER, for June 18-20, 2021
(137 - 1:08pm, Jun 21)
Last: Edmundo got dem ol' Kozma blues again mama

NewsblogI Am Breaking My Silence About the Baseball Player Who Raped Me
(2 - 7:15am, Jun 21)
Last: Never Give an Inge (Dave)

(2 - 5:25pm, Jun 20)
Last: Joyful Calculus Instructor

NewsblogOT - Catch-All Pop Culture Extravaganza (June-August 2021)
(3 - 2:46pm, Jun 20)
Last: Ben Broussard Ramjet

NewsblogHow baseball’s response to RFK’s death inspired some players to fight back and sit out
(1 - 8:45am, Jun 20)
Last: willow19

NewsblogFrom the spitball to Spider Tack: A brief history of foreign substance use by pitchers in baseball
(2 - 12:46am, Jun 20)
Last: the Hugh Jorgan returns

Page rendered in 0.8692 seconds
48 querie(s) executed