My Hall of Fame ballot

Five hundred and seventy-five Hall of Fame ballots arrived in Cooperstown by last Monday, and the results will be announced next Tuesday. I didn’t receive one of those ballots, but that didn’t stop me from thinking about who should be in the Hall of Fame. So I’m going to give you my opinions of all of the candidates on the ballot, relying primarily on Win Probability stats.

Let’s talk about Win Probability for a second. Thanks to the tremendous efforts of Retrosheet, we have the play-by-play action of virtually every game since 1957. Which means that we have detailed play info for every player who is listed on the ballot. Every single play for every single player. And thanks to Jeff Sagarin, we have the “win probability” interpretation of the play-by-play results. In other words, we have a detailed log of how much each batter and pitcher contributed to his team. Why use any other stat?

Well, there are a couple of issues with WPA. As you may know, WPA assigns a weight to every play based on how much it changed a team’s chances of winning. If a batter hits a home run, it’s a positive event for him and his team, negative for the pitcher and his team. But the exact weight is based on what we know about the game at the time.

So a home run in the first inning of a tie game doesn’t get as much credit as a home run in the ninth inning of a tie game, because a win is more likely when you take a ninth-inning lead. A lot of people have a problem with this, for understandable reasons. Once the game is over, a home run in the first inning counts just as much as a home run in the ninth.

But we’re only going to look at career totals today; most of the players on the ballot were involved in at least 8,000 plays. That’s a large enough sample size for differences within games to even out. So if a player has helped (or hurt) his team by performing better (or worse) in “late and close” situations, WPA picks that up. And if a player has proven to be that type of clutch player over an entire career, I don’t have a problem with it being on his Hall of Fame resume.

Second issue: WPA compares each player to average (a .500 team). This hurts players who had long careers because their WPA totals in later years (when they might have been worse than average but still better than a replacement) can wipe out some of the gains they achieved in their above-average years. So I have changed the baseline to replacement level, which I define as a .350 team. This particularly has a big impact when you compare starting pitchers with long careers to relievers.

A few other thoughts: WPA doesn’t include the impact of fielding. I’ll try to add some fielding comments as I go through the list. It also doesn’t include position adjustments; the batting contributions of a shortstop are considered the same as those of a designated hitter. And this version of WPA isn’t adjusted for differences in parks, though it is adjusted for differences in scoring between baseball eras. We’ll talk about all of these issues as we go through the list.

I’m not suggesting that WPA is the only criterion for Hall of Fame consideration. But I am suggesting that it can stand as a singular statistical criterion, because it cuts to the essence of virtually all other stats. So here’s a list of every player on the Hall of Fame ballot, in order of WPAR (Win Probability Above Replacement):

Mark McGwire: McGwire leads the list with 67 WPAR, 32nd among all players who played from 1957 to the present. Every player ahead of McGwire is either in the Hall, not yet eligible or named Pete Rose. You sometimes hear the argument that McGwire doesn’t deserve to be in the Hall because all he did was hit home runs. Um, no. His home runs, walks and win contributions add up to a Hall of Fame player.

Talk about ironic: the guy just ahead of McGwire is noted Hall of Fame cheater Gaylord Perry (also with 67 WPAR). And the player just behind McGwire is Dick Allen (65 WPAR), who has been kept out of the Hall primarily because of his prickly personality.

Bert Blyleven: Blyleven is 34th on the list of all players, with 65 WPAR. The fact that Blyleven isn’t in the Hall is absurd. Every eligible player who is close to Blyleven’s WPAR total is already in the Hall. I can’t add anything to the miles of words Rich Lederer has already contributed to the subject.

Tim Raines: Raines is next, at 63 WPAR, 44th overall. Hall of Famer Tony Perez has 63. Hall of Famer Dave Winfield has 62. Raines was a dynamite player and should also be in the Hall of Fame.

You may be surprised that Raines ranks so highly. He didn’t have a lot of power, but he was always on base, which also means that he didn’t make as many outs as you might expect. He was also arguably the greatest basestealer of all time. These are the types of nuances that are captured by WPA.

There was recently a great email debate between Jayson Stark and Peter Gammons on the subject of Raines and the Hall. Give it a read.

Tommy John: 60 WPAR, 50th overall, just behind Hall of Famer Bob Gibson (61 WPAR). I wasn’t really a Tommy John Hall of Fame supporter, so I was surprised to see how high he ranks on this list. Obviously his long career helped him—his pure WPA total is 22, five less than Gibson’s 27 (Blyleven also has 27 WPA). The crazy thing about John is that he finished in the top five in league ERA six times, nearly 10 years apart; three times from 1966 to 1968 and three times from 1977 to 1981.

If you think average is a better baseline for Hall of Fame membership, you might not support John. Also, WPA doesn’t spread credit between pitching and fielding, and John depended on his fielders quite a bit because he wasn’t a strikeout pitcher (unlike Blyleven). Factoring that in would probably lower his rating a bit.

Dave Parker: 57 WPAR, 59th overall. Why doesn’t Parker get more attention? If you use Bill James’ Hall of Fame Monitor, Parker rates a 125 (where 130 is a “virtual cinch”). I’m forever reading about Murphy, Dawson and Rice. Why not Dave Parker?

Admittedly, we’re getting into some gray area here. Other players with similar WPAR totals are Hall of Famers Paul Molitor (57) and Roberto Clemente (56) and non-Hall of Famers Reggie Smith (59) and Dwight Evans (58). And Ron Santo (55). But Parker deserves stronger consideration than he’s gotten and, if he were voted in, he wouldn’t be out of place at all.

Rich Gossage: 54 WPAR, 74th overall. Among pitchers who spent virtually their entire careers as relievers, the Goose is first in WPAR. Trevor Hoffman is next with 51 WPAR. Bruce Sutter has only 30. I’d vote Goose into the Hall for sure.

Andre Dawson: 54 WPAR, 75th overall. You may think of Dawson as a right fielder, primarily because he played right in 1987, his MVP year in Chicago. But according to Tangotiger, he spent as much time in center as Dale Murphy.

I never appreciated how good a center fielder Dawson was. Sean Smith has developed his own system for rating outfielders, including a terrific approach to fielding measurement, and he concludes that Dawson trails only Raines in Hall of Fame qualifications. Dawson was that good in center field. I used to be against Dawson’s inclusion in the Hall because of his low OBP, but Sean’s ratings have changed my mind.

Harold Baines: 53 WPAR, 78th overall. Baines was a great hitter with a long career, but his extended years as a designated hitter will keep him out of the Hall, as they should.

Dale Murphy: 49 WPAR, 89th overall. Sean rates Murphy behind Rice and Parker, and just ahead of Baines. I would give Murphy a boost because he was very popular and a good role model, but I probably wouldn’t vote for him. It’s close, though.

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.

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.

Jim Rice: 46 WPAR, 113th overall. I’m guessing you’re surprised that Rice ranks so low. That was my reaction, too. I also know that a number of voters picked Rice but not Raines. So let’s dig a little bit to see if we can make sense of it. Here are the top 10 WPAR seasons for both Rice and Raines:

 Raines        Rice
 1987   6      1978   9
 1986   6      1983   6
 1984   6      1977   6
 1983   6      1979   4
 1985   5      1982   4
 1989   5      1986   3
 1982   4      1975   3
 1992   4      1984   3
 1981   4      1985   3
 1990   4      1980   3

Counting only their top ten seasons, Raines outscores Rice, 49 to 42. Rice had that tremendous 1978, with 46 home runs and 139 RBIs, and his 9 WPAR led the AL by a good margin (league leaders typically have about eight WPAR). But he only had four other seasons with four or more WPAR. Raines had 11, including four seasons of six WPAR.

Taking a closer look, in 1979 Rice batted .325/.381/.596 with 39 homers and 139 RBIs and finished 5th in MVP voting. Yet he accrued only four WPAR, the same as Ken Landreaux (.305/.347/.450). How did this happen? Well I don’t know for sure, but I have a feeling that the fact he batted .209/.270/.330 in “late and close” situations (7th inning or later with the batting team tied, ahead by one, or the tying run at least on deck) had something to do with it.

Now, that was an unusually poor performance by Rice. In his career, Rice batted .274/.337/.453 (.265 GPA) in late and close situations. Not very good, but not downright terrible. On the other hand, Raines batted .316/.414/.442 (.297 GPA) in late and close situations, which gives a boost to his WPA numbers. You may not believe in clutch hitting, but Tim Raines did.

Don’t forget that Rice’s batting stats were boosted significantly by Fenway Park (where he batted .320, vs. .277 on the road), which isn’t factored into his WPAR stats. Also, Rice wasn’t a particularly good fielder. Bottom line: Murphy, Dawson, Baines and Parker are more qualified Hall of Fame candidates than Jim Rice.

By the way, Rice had as many opportunities as other players to contribute to his team’s wins. I estimate that his Leverage Index (a measure of the criticality of his at bats) was 1.09 (where 1.00 is average). For a quick comparison, Dave Parker’s was 1.05. Raines’ was 0.96, presumably because he was a leadoff hitter and came to bat less often with men on base.

Chuck Finley: 44 WPAR, 125th overall. Chuck Finley was a very fine pitcher, but not a Hall of Famer.

Don Mattingly and David Justice: 37 WPAR. Well, that’s a surprise. Mattingly and Justice tied? If it makes you feel better, Mattingly is ever so slightly ahead of Justice, 37.04 to 36.88.

Lee Smith: 37 WPAR, 180th overall. I know there is support for Smith because of his saves total, but can’t we all just admit that saves aren’t a good measure of baseball greatness? Rollie Fingers is a bit ahead of Smith, with 39 WPAR. So is Tom Gordon.

Alan Trammell: 30 WPAR, 257th overall. Unless my eyes deceive me, Alan Trammell is the highest-ranking shortstop in WPAR among all shortstops eligible for the Hall but not in the Hall since 1957 (Barry Larkin has 46 WPAR and ought to be an easy choice when he becomes eligible). Jim Fregosi is next behind Trammell with 26 WPAR.

How should we rank shortstops? Well, players with the skills to play shortstop are rarer than those who can play the outfield, so we really should change the baseline we’re using. Skyking posted a little summary of the modifications you can use for this. Basically, you should give a shortstop credit for five runs above average per 1,000 innings. Trammell played approximately 19,000 innings at shortstop, so give him credit for 95 runs above average. It takes nine to 10 runs to convert a loss into a win, so Trammell gets 10 wins above average added to his WPA total for a position adjustment.

I have to use WPA totals here, not WPAR. In his career, Trammell had 13 WPA; adding 10 for the position adjustment takes him to 23. Following the methodology he used in the THT Annual, Tangotiger reports that Alan Trammell was an average fielder at short, so he gets no extra fielding credit.

Twenty-three WPA is pretty good. This is very rough, but when I add a position adjustment for some other players, Rice has 13, Parker has 23 and Raines has 34.

Robb Nen: 28 WPAR. Nope.
Chuck Knoblauch: 22. Nope.
Brady Anderson: 20. Nope.
Jose Rijo: 19. Nope.
Todd Stottlemyre: 18. His dad had 33.
Rod Beck: 18. Nope.

Dave Concepcion: 16 WPAR, 625th overall. Here’s what SI’s Jon Heyman said about his vote for Concepcion:

This is his 15th and last year on the ballot, and he’s probably going to get his usual 10 percent of the vote again. The reason I am in that 10 percent is that I think he was perhaps the best all-around shortstop of his generation and an underrated piece of the Big Red Machine. Great defender (five Gold Gloves) and superb stealer (321 stolen bases), his career looks a lot like Hall-of-Famer Phil Rizzuto‘s to me — without the announcing, of course.

I’m not sure about the “best all-around shortstop” tag, but Concepcion was indeed a great fielder. In fact, he’s the only person on the ballot whose primary claim to fame was his glove, and WPA doesn’t include any credit for glovework. Let’s see if we can adjust for that omission.

Let’s start with his actual WPA (not WPAR) total of -1. He played almost the same number of games at shortstop as Trammell, so let’s add 10 wins, bringing his WPA total to nine. Next, let’s factor in his fielding prowess.

According to Tango, Concepcion was about 20 plays per year better than the average shortstop—not in the company of Ozzie Smith and Mark Belanger, but in the next tier. Concepcion played about 15 full seasons, which means he saved 300 plays more than average. At .75 runs per play, that’s 225 runs, or roughly 22 wins.

22 plus nine is 31 WPA, more than Rice, Trammell and Parker and almost as many as Raines. If you believe these assumptions (and I think they’re roughly correct), Concepcion’s glove gets him into the Hall.

Travis Fryman: 12. Nope.
Shawon Dunston: 6. I loved watching Dunston play, but nope.

That’s it, every player on the Hall of Fame ballot. For my money, the names of Blyleven, Raines and Gossage should definitely be called to Cooperstown on Tuesday. I’d let my little steroids voice guide my McGwire vote, and I’d vote for John, Parker, Dawson, Concepcion and Trammell if I were a “large Hall of Fame” person.

A lot of BBWAA voters seem to have a problem with baseball statistics. Jon Heyman took a little dig at people like me by saying that he’s actually seen all the players on the ballot play. Well, I’ve seen all the players on the ballot play, too, but I’m not too proud to admit that I can learn more by spending extra time with baseball statistics, particularly the right baseball statistics.

And I hope you can see that this little statistic called WPA, along with the cutting-edge fielding research being conducted by Sean Smith, Tangotiger and many others, can shed a lot more light on the subject.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Weird games
Next: The international game »

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>