My Hall of Fame ballot
4
|
The Hardball Times found this 1/3/2008 on www.hardballtimes.com [flag] |
Tags:
MLB
Hall of Fame
Comments (16)
-
-
studes Thanks, Guy. It is certainly possible to introduce a ballpark factor into WPA -- just use a different table for each ballpark, making assumptions about the run environment. We've worked on that at THT, but I'm not sure what the status is.
I agree 100% regarding relievers, but I also think you can make a case that some relievers ought to be in the Hall, even if a perfect replacement level ought to be much higher, just because relief pitching has become an imortant role. That's why I didn't both with it too much.
But if we were to try to develop a "perfect" system, we'd definitely want a different replacement level for relievers.
-
-
philosofool Nice article. This is more about WPA than about hall of famers, but won't players on good teams get a higher WPA than those for bad: batters on good teams will have more plate appearances with men on base, for example? I don't know the WPA formula. Would pitchers be affected (e.g. a K in a one run game is worth more than in a three run game--benefiting, ironically, pitchers with poor offenses)?-
studes You might want to read this article for background about WPA:
http://www.hardballtimes.com/main/article/the-one-about-win-probability/
To answer your specific questions: No, generally players on good teams won't get more WPA than players on bad teams (except to the extent they deserve it). The type of player who is likely to be helped or hurt most by WPA is one whose team played a lot of close games -- likely a .500 team.
Which leads to the second point that yes, both pitchers and hitters are affected evenly. A strikeout or hit in a close game is worth more than the same event in a blowout.
If a great pitcher pitches for a poor offense, he'll get more WPA than if he played for a team with a great offense. That's as it should be, because he'll have contributed more to the team's winning.
-
-
ubu49 Fine, detailed summary, and conclusions, but I think you might be doing Jack Morris a disservice in trying to build up BB.
I'm not familiar with WPA, but, going by the statistics available at Baseball-Reference, it would seem that Jack is rather more than equal in accomplishment:
Jack Morris: 47 WPAR, 102nd overall, just ahead of Kevin Appier. Some BBWAA members voted for Morris but didn't vote for Blyleven because he never had that "one big year." My counterargument would be that Morris never finished higher than fifth in league ERA. Blyleven did it five times.
True, but Morris was top-five twice, and 5 times in the top ten. Now Blyleven has 10 times... but with a lower career winning %.
In career Cy Young voting, Jack's 66th, to Bert's 98th, .71 to .45, and was considered 7 years to BB's 3. Morris also has the edge in MVP shares (and years considered 5-2), .18 to .09.
Then there's:
To me, the argument for Morris in the Hall comes down to one game, and I don't think one game is enough for the Hall.
Which is really not the case:
Morris has three 20 win seasons to Bert's... well, one (a year in which he also lost 17!), another 19 gamer, two 18's, and two 17's.
He was a mainstay of three World Series winners, going 4-0 in the first two, and while he was 0-2 in the last, he was 21-6 that regular season, a big reason Toronto got there.
Blyleven has good series stats too, but not as good: 2-1 to 4-2.
Bert has his 287 wins, making him 26th all-time... but he's also 10th in losses. He pitched 4 more years that Jack, to win just 33 more games. He was actually 43-41 in those four years (reminiscent of, but not as bad as Don Sutton's last three campaigns, when he was 29-28... to hit that magical 324 wins!).
Don't think it's far-fetched to say Jack might have dragged it out a while longer and equalled Bert's total, but he seems to have had the sense not to linger too long with diminishing skills.
Anyway, I hope for Bert's sake he gets in, he's at least the equal of Sutton, and better than a few others already at Cooperstown. Please though, leave WPA to the Works Progress Administration, and give Jack a break. He might well have been the better player.
-
studes Sorry, but there is no way that Morris was better than Bert. Won-loss records and appearances in the postseason are meaningless for pitcher evaluation, if you've got better info. All they mean is that one guy played for a better team than the other.
I certainly agree that contributing to wins is the most important thing. WPA measures that far better than a won/loss record does.
For more on the Bert/Jack comparison, read this:
http://joeposnanski.com/JoeBlog/2007/12/29/best-pitcher-of-the-1980s/
...and remember that Blyleven had already been pitching a full decade before the '80's even came about.
-
ubu49 Dave, Sorry, but if the WPA system rates Jack Morris on the same level as Kevin Appier, then
it
simply
doesn't
work.
"Won-loss records and appearances in the postseason are meaningless for pitcher evaluation, if you've got better info. All they mean is that one guy played for a better team than the other"
Damn! And here I always thought winning in big games when the chips are down actually counted for something!
Maybe baseball should emulate our feel-good-about-yourself education system and figure a way to stop actually declaring a winner in all these pesky games? Maybe just give everybody a nurturing pat on the butt and congratulate them on their Win Shares score?
The problem I see with Win Shares, WPA, and other such systems, is that while they're purportedly designed to level the field, what they end up doing is blurring the picture... and worse yet, belittling the accomplishments of other athletes.
Invariably, some perhaps unjustly neglected X-guy's fans will prop him up, dust him off, point to his stats and tell us he was not only one of the best of his era...
but waaaay better than this other guy who actually achieved more...
because everybody X-guy played with was s#*t, he wasn't in a big market city, or had the misfortune to get a knee injury trying to stick a twenty in some stripper's g-string...
because WinShares, or WPA, or some other ouija-reliable system says he was.
Stop, already.
Is being on a good team the only reason a player himself is good? Obviously, and using your own argument, no. It's very much a two-way street. You might consider one of the reasons Morris' teams performed well behind him, was their confidence in what they knew he could do. He was an important part of their success.
Read the anti-Morris blog you recommended (the guy really despises him, regardless of what he says), and saw very little of interest, beyond the complaint that he got too many runs behind him (kinda like the nitwit Emperor's comment in 'Amadeus' that Motzart used "...too many notes."), and wasn't as good as Roger Clemens (Oh Wow!), or a host of lesser players who had a good year or two, but mostly don't approach JM on the career-performance radar. Also, I'd never contend he was the best pitcher of the 80's... but neither was Blyleven (and he wasn't that in the 70's either).
I'm all for Bert being in the HOF (Which should soon be the changed to "Hall of the Overpaid and Famous" as today's players drift in). He'll be right at home beside Sutton, Sutter, Eckersley, and Drysdale.
Sorry, but there is no way Bert was 'better' than Jack.
-
studes Well, I agree with you about one thing. Individual players shouldn't be awarded wins or losses. Teams win games, not individuals.
That is an interesting point: that Morris "inspired" his teammates. But as you can see from this article:
http://www.baseballprospectus.com/article.php?articleid=1815
...Morris didn't really get any more run support from his teammates than the team's other pitchers did. So what was the nature of this inspiration?
WPA is what it is, just like runs and wins and every other mainstream stat. It's not like WIn Shares, which contains a lot of opinion and judgement calls. It's a straightforward stat. You can reject what it measures, just like I reject wins allocated to pitchers as a valid measurement, but that should be based on what the system measures, not because you don't like the results.
The thing is, WPA should be perfect for those who champion Morris's cause. Players who perform better in more critical situations, when the score is tight or whatever, do better in WPA. The fact that Morris doesn't kind of undermines one of the arguments for Morris's "greatness" (ie. that he pitched to the score, won the big games, etc.).
Stat fact: Appier's lifetime ERA was 3.74 at a time the league ERA was 4.51. Morris's was 3.90 at a time the league ERA was 4.05. Even if Morris's ERA was inflated because he stayed in games longer, that's a mighty big difference. You do approve of ERA, right?
-
studes Sorry about that last line. Guess I was feeling snarky.-
ubu49 Stat fact: Appier's lifetime ERA was 3.74 at a time the league ERA was 4.51. Morris's was 3.90 at a time the league ERA was 4.05. Even if Morris's ERA was inflated because he stayed in games longer, that's a mighty big difference. You do approve of ERA, right?
Sorry about that last line. Guess I was feeling snarky.
Snarky's no problem at all. I'm used to boards where the debate is often far more unpleasant. I do try to clean up my posts here, as it's a very well-mannered club. Zing away, and no hard feelings.
That said, ERA's subject to enough deceptive variables to make it no better an indicator than Winning%. The all-time leaders are almost all dead-ball era pitchers, and the two best modern starters are Koufax and Whitey Ford... neither of whom makes your WPAB top 25 list:
http://www.hardballtimes.com/main/article/ten-things-i-didnt-know-last-week41/
Don't you approve of ERA, Dave?
As for Appier, he pitched for 16 years and won 169 games... and Jack? Well, it appears he won 237 games over his first 16 years. But hey, what's 68 wins among friends? (Apparently about four-and-a-half years for Kevin Appier). KA had 402 starts in those 16 years, Morris 477. That's still alot more wins for only 75 more starts, enough to make the comparison highly questionable.
What was Kevin's shoe size? Maybe he at least had bigger feet?
I'm not a statistician, but I do admire the effort that goes into devising these formulae... but IMO, the WPA and WPAB systems seem to be giving too much weight to the certain variables. Based on the list here:
http://www.hardballtimes.com/main/article/ten-things-i-didnt-know-last-week41/
Longevity and strikeouts (and ERA?).
Koufax one step ahead of Jerry Koosman, and behind Ryan, Blyleven, Tommy John, Don Sutton, Drysdale, and Phil Neikro????? HaHa.
It looks like a Bill Clinton poll where you figure out the answer you want first, then go looking for a demographic that will provide it.
I'd take a rotation of non-top-twenty listers Sandy, Warren Spahn, Whitey Ford, Ferguson Jenkins, and Mike Cuellar, or Catfish at their best, over a Blyleven, Tommy John, Sutton, Nolan Ryan, Phil Neikro, and Eckersley team... any day.
I'll stop now, never actually been a big Morris fan anyway. Only joined in because your dismissal of him was pretty cavalier. I have no problem with Bert's selection, but you might consider that if it takes masses of stats to keep his hopes alive, but Jack apparently comes readily to BBW's minds, there's probably alot more going on here than just one game.
BTW, if you're going to push for somebody, why not Dave Stewart? Know it's a terribly short starter career, dumb Dodger management, but he was a great pitcher for a few years, and a class act.
Peace... and your feet smell and your mother dresses you funny!
ubu
-
ubu49 Dave, just one more point:
If Jack didn't in fact get more than normal run support from his team mates...
doesn't that make his performance even better?
-
studes Well, we're probably talking past each other at this point. This has become one of those threads where people don't compare apples to apples. But a few notes:
- Ford isn't on the WPA list because the data doesn't cover much (certainly the best part) of his career. Same with Spahn.
- Koufax does much better with WPA than WPAB because he had some awful years in Brooklyn at the start of his career. People tend to think of Koufax just at his peak -- not the years it took to get there.
- I don't get where you call Morris's performance "even better." He was the best pitcher on a team that scored a ton of runs. His ERA was very good, but not great. His record reflects what he did and the quality of his teammates.
- You may be right that WPAB favors longevity too much. To offset that, I could raise the replacment level closer to WPA. Judgment call. Some people have advocated a system that only values the big years, and gives no negatives for below-average years. Sounds like that's a system you'd prefer too, given that you'd pick your first team of pitchers "at their best".
I used both WPA and WPAB to inform my comments. Morris does relatively worse in WPA than WPAB.
- Regarding Stewart, I'm not "pushing" for anybody for the Hall. My article was a list of the guys eligible for the Hall this year, and what I thought of them. FYI, Stewart has 8 WPA, 28 WPAB, far behind Morris.
Your Stewart comment shows where you're coming from. Overall, he was good, not great, but he did have some great years with Oakland (particularly those 20-win seasons). That's not where I come from. I like to look at entire careers, and I like to use stats that are a better reflection of contributing to wins than Win%.
- The nut of our disagreement: I think ERA is much better than Win% once you correct for era, which is what I did in the Appier and Morris comparisons (and what you didn't do in citing your examples). Win% is a reflection of both the pitcher and his teammates. In many cases, it's more a reflection of teammates than the pitcher.
-
-
-
-
-
-
-
-
walshj58 Nice work, Dave, as always.
Do you think some sort of evaluation of peak should be included in deciding on Hall of Fame worthiness?
What was Sandy Koufax's WPAR?
-
studes Hi John,
I wrote about Koufax and WPAR (called it WPAB then) here:
http://www.hardballtimes.com/main/article/ten-things-i-didnt-know-last-week41/
He ranked 16th among all these pitchers in WPA and 25th in WPAR. I do agree that some sort of peak evaluation ought to be included in Hall of Fame worthiness. I didn't talk about it in the article, but I also looked at each player's WPA, to see if anyone stood out, and no one really did.
-
-
bads85 Great artilce, Studes.
How does WPAR compare to WSAB (which incorporates defense) with these players?
-
studes Thanks, JP. That's a good question. It may be something I can get to in a month or two, time permitting.
-
Links (4)
This Week's Links (12/31-1/4)
Published 1/4/2008 by Vegas Watch at Vegas Watch
Hall of Fame edition, because there wasn't a whole lot else going on this week. Looks like Gossage is in, while Blyleven, Rice, and Dawson all have a shot. Raines does not. Dave Studeman's imaginary ballot. OMDQ's imaginary ballot. Heyman's ballot really was bad. Jay Jaffe on this year's SP candidates (subscription only). Posnanski on Morris, and the Pozcars results. More on the Blyleven-Morris "comparison".
Friday Links (4 Jan 08)
Published 1/4/2008 by Geoff Young at Ducksnorts
... Two more takes on the Hall of Fame: Dave Studeman examines the current candidates, while Patrick Sullivan ...
Worthless
Published 1/11/2008 by Maddog at Another Cubs Blog
... who gets an unusual amount of support while Raines got fucked over this year. Raines ranks 44th in Win Probability Above Replacement among all players since 1957. Sadly, 2 others who most definitely belong in the Hall rank just ahead of him in Mark McGwire and Bert Blyleven. Raines was ...
A New Statistic
Published 1/13/2008 by Vegas Watch at Vegas Watch
... Hall of Fame ballot but might not use it -- it's not about the inflated muscles. With Mark McGwire, it's all about his flawed numbers. McGwire has gotten exactly 128 votes (23 percent) each of the last two years in hall of fame voting. It should never get any higher and any intelligent baseball fan can see it with his or her own two eyes." It immediately becomes clear that I am not an intelligent baseball fan. Neither is he. Nor he. Not intelligent. It's always good to start an article by a) apologizing, and b) ...
Great analysis, Studes. As you know, I'm not the world's biggest WPA fan, but I agree with you that over a career it should still provide a fair assessment most of the time.
A question and a quibble:
Question: Do you think there's a practical way to incorporate park factors into WPA? And if so, any plans to try to do that? (Rice's mark would certainly be still lower without the benefit of Fenway).
Quibble: I don't agree with using the same replacement level for both relievers abd starters. We know that short relievers have a substantial advantage over starters, mainly because they face hitters only once and can throw harder (more Ks). And since high-leverage innings are a team resource -- the team can elect who gets to pitch them -- a team would basically never need to use a .350 pitcher in high-LI situations. So I think you need to set replacement level for closers much higher, probably something like .500.
That said, Gossage in particular I think threw longer outings (often 2+ IP), and so probably enjoyed less of a "reliever boost" than today's closers. Perhaps he should be measured against a lower benchmark than current closers, but still higher than starters.