That thought led me to decide to do some retroactive spot checks of the BBWAA's performance on past Hall of Fame ballots. I wanted these checks to be pretty simple. I don't want some new, proprietary statistical formula, or a catchy acronym. I just want a simple reality check on whether or not the BBWAA's votes, as a body, generally tracked with player performance.
To do that, I needed a couple of simple things. First, I needed the annual vote totals, which are handily available on the Hall of Fame's website. (Though, I must say, the elimination of every players voting history from their new web design is somewhat vexing.)
Second, a needed a standardized performance measurement of some kind. I decided to use WARP3 scores as the performance measure, for no other reasons than that they are internet-accessible and standardized. It's not perfect measure by any means, and can be downright misleading as just a flat, rolled-up number. But all I'm looking for is something that will be directionally correct, not precise to the nth degree. I just need a basic hammer to drive a basic nail, not a Paslode IMCT Impulse Cordless Framing Nailer.
So, in short, here's what I did. I picked a random Hall of Fame ballot (1981 in this case) and ranked every player on it by total votes received and WARP3 score. Then I subtracted their rank in the voting from their WARP3 rank to get a basic delta that would show me whether or not they were overrated or underrated by the BBWAA. Like I said, nothing too complex here.
For instance, the top guy on the ballot was Bob Gibson, with 337 votes. He also had the top WARP3 score, 119.8. So, subtracting his voting rank (1) from his WARP3 rank (also 1), he scores 0, meaning he was rated by the BBWAA exactly where he should be. Congratulations writers, you have past your first (and easiest) test. Now for some more results:
Underrated Players
Bill Mazeroski - 3 WARP rank minus 20 vote rank = -17
Luis Aparicio - 5 WARP rank minus 18 vote rank = -13
Dick McAuliffe - 21 WARP rank minus 33 vote rank = -12
Sam McDowell - 22 WARP rank minus 33 vote rank = -11
Leo Cardenas - 16 WARP rank minus 27 vote rank = -11
Vada Pinson - 13 WARP rank minus 23 vote rank = -10
Richie Ashburn - 2 WARP rank minus 11 vote rank = -9
I'll stop there because I don't think many people care about Lindy McDaniel and Claude Osteen.
There are at least a couple noticeable trends. The heavy defense, weak offense guys don't seem to do too well. Mazeroski scored as well as he did in WARP almost entirely due to remarkable defense, something the writers seemingly couldn't care less about. Moreover, there isn't a single genuine power hitter in the group. Pinson had a touch of pop, but he was essentially a singles and doubles hitter his whole career, and without a string of batting titles the voters apparently decided he should reside with the rest of the banjo hitters near the bottom of the ballot.
Also, note that most of these guys played the bulk of their careers in the Rust Belt. Pittsburgh, Baltimore, Detroit, Cleveland, and Cincinnati aren't exactly teeming with sportswriters who will stuff the ballot box for the local guys on the ballot. Chicago (Aparicio) and Philly (Ashburn) have a few more scribes who might be willing to pad the total of their readers' favorites, but apparently not for a pair of leadoff hitters whose teams never won the big one.
Among this group of underrated players, the BBWAA only changed their tune about Aparicio, and his case is genuinely strange. Aparicio first appeared on the ballot in 1979 and received a healthy 28% of the votes cast. The next year he improved a bit, to 32%. Suddenly, with the 1981 ballot, his support dropped like a rock, with over 60% of his prior supporters deciding he was no longer worthy of their vote. This same kind of inexplicable drop ultimately doomed Luis Tiant’s candidacy a few years later, but apparently the voters decided Aparicio was worth saving. In 1982, he got all of his previous supporters back, and then some, collecting over 40% of the vote and ultimately being elected in 1984. In fact, eleven of the seventeen players who received more support than Aparicio on the 1981 ballot still appeared on the 1984 ballot and Aparicio passed every single one of them, getting the highest vote total of any player that year. What this means, among other things, is that in 1981 there were 120 voters who thought Nellie Fox was a Hall of Famer but that Luis Aparicio wasn’t, and just three years later that gap had swung completely the other way, with 95 voters casting their ballots for Aparicio but not Fox. In other words, 215 voters suddenly changed their minds about the relative position of the two men in baseball history. And they wonder why people question them.One more note on the guys who were underrated in 1981. I don’t, in any way, believe that Dick McAuliffe or Sam McDowell belong in the Hall of Fame. But I do know that Sam McDowell was just as good a pitcher as Lew Burdette or Roy Face or Don Larsen. Better in some cases. All of those guys got significant support, ranging from 23 to 48 votes, begging the obvious question as to how guys of that caliber got a few dozen votes while McDowell didn’t get any. Same goes for McAuliffe. He wasn’t a Hall of Famer on his best day, but he was a solid shortstop whose 64.8 career WARP3 score stands very nicely with the group of Ted Kluszewski (59.3), Harvey Kuenn (59.1), Elston Howard (58.2) and Roger Maris (56.6). The fewest votes any of those guys got was Kluszewski’s 56, yet McAuliffe got shut out entirely.
Speaking of some of those guys…
Overrated Players
Roger Maris – 28 WARP rank minus 12 vote rank = +16
Don Larsen – 35 WARP rank minus 21 vote rank = +14
Elston Howard – 26 WARP rank minus 14 vote rank = +12
Harvey Kuenn – 24 WARP rank minus 13 vote rank = +11
Gil Hodges – 11 WARP rank minus 3 vote rank = +8
Lew Burdette – 25 WARP rank minus 18 vote rank = +7
(Note: I skipped over two guys, Glenn Beckert and Gates Brown, who each received one vote and were therefore technically overrated using this system. In the grand scheme of things, those two stray votes really aren’t in focus here. Whether or not writers should be casting spare votes in the direction of guys who no one really sees as a Hall of Famer is a topic for another day.)
What a shock, the three most overrated guys on the ballot were all Yankees. Skip a spot and you get a Brooklyn Dodger. Flabbergasting, isn’t it?
In all seriousness, this will be a very interesting trend to follow as I look at additional ballots. Maris and Larsen both have unique claims to fame outside of playing for IBM, err, I mean the Yankees, so their respective vote totals could be expected to receive boosts. That’s not really the case with either Howard or Hodges. They were very good players on very good teams, but neither had that single signature accomplishment or record that would garner them extra votes. I think the likelihood is that they were just more publicized than guys who were similar to them, almost certainly because they played the bulk of their careers in New York. The only other player on the ballot who could be identified either completely or mostly with a New York team was Thurman Munson, and though he wasn’t wildly overrated, he did, in fact, place a bit higher in the voting (16th) than his WARP score (18th) warranted.One ballot is far too small a sample to draw any conclusions, but there are already a few things that need to be tracked. It will be interesting to see if a player’s style of play (slap hitter versus power hitter, or power versus finesse pitchers, for instance), or the number of years they’ve been on the ballot, prove to be significant factors in their levels of support. The most disturbing possibility is that there could be geographic bias on the ballot. It’s not as if anyone is surprised that players from New York get more votes. There’s simply more writers from there, ergo more voters who saw them play. On top of that, New York teams have historically played more games on television, giving their players more exposure to the writers that do the voting. A boost in their vote totals isn’t a shock in any way.
But what does that indicate? To me, it begs the question about whether or not the voting process should be changed. Each vote cast is supposed to be done objectively, with the writers chosen specifically because they saw more ballgames and therefore had more information at hand to render an objective opinion. The presence of any kind of bias means that the writers’ objectivity is compromised. If the writers really are throwing unwarranted support toward players they simply saw more often, at the expense of equal or better players in smaller markets, doesn’t that mean the process is broken? Isn’t the presence of any bias in the voting process, geographic or otherwise, an indicator that the writers can’t be objective, and therefore shouldn’t be voting?
