2017-12
There is a flaw in the original design of the Hall of Fame’s selection process which inevitably causes it to sometimes fail. Why it has taken me 40 years of writing about the Hall of Fame and studying the Hall of Fame to see what the cause of the problem was, I don’t know, but I console myself with the thought that nobody else has apparently put their finger on it, either.
Explaining how I got there. I had the thought that perhaps I could construct a model of Hall of Fame voting, which I could then use to compare alternative methods. In other words, suppose that we KNEW absolutely who should be in the Hall of Fame? In real life we don’t know absolutely who should and should not be in. We can only have the perfect knowledge of this in a model.
In my model there were going to be two components to each voter’s vote: the underlying facts, and the "perceptual error" of each voter. I don’t see Bill Mazeroski as a Hall of Famer; obviously some voters did. One of us has a perceptual error.
All of us have perceptual errors all the time. My understanding of the world, your understanding of the world, anybody’s understanding of the world. . . it is always colored by perceptual errors. The perceptual error is different from the gray area in the voting. In my model, I would going to assume that each player has a "True Value", and that each voter has a "Perceptual Error" of each player’s True Value. Thus a player—let us call him Duke Snider—has a True Value—let us call that 12,656—and that each VOTER has a perceptual error of that value, which may be as high as 2,000 or as low as negative 2,000.
One voter may focus unduly on Duke’s sometimes negative relationships with teammates, or may overrate the effects of the fact that Duke hardly ever had to face a left-handed pitcher, or may overrate the effects of playing in Ebbets’ Field with good hitters around him, and may thus have a negative 1,500 perceptual error, thus valuing Duke at 11,156, while the Hall of Fame standard may be 12,500. He thus sees Duke as being a little more than 10% short of a Hall of Fame standard. Another voter may focus on Duke’s positives, may overrate him as a defensive player or may ignore the benefits he got from his park, and may have a perceptual error of +1,500, thus may value Duke at 14,156—about 13, 14% OVER the Hall of Fame standard.
The relationship between the perceptual error and the gray area is that the perceptual error is one of the main causes of the gray area. The perceptual error causes a gray area, and also different people have different ideas about what the standard should be, which also causes a gray area. There could be other things which contribute to a gray area, as well.
THAT idea—that I could build a model which would help us to test voting systems—has not yet come to fruition, because it is just too complicated, at least yet. Maybe I’ll make it work; maybe not. Anyway, let’s go back to the Duke.
Duke’s True Value is 12,656, which just above the Hall of Fame standard. (I chose Duke because he is just a little bit better than the AVERAGE Hall of Famer.) But do you see the problem?
Duke is not going to be elected to the Hall of Fame.
Let’s assume for the sake of argument that all voters (100% of the voters) agree about what the Hall of Fame standard should be, 12,500 units. If all the voters had a perfectly accurate view of Duke Snider, he would be elected, because he is over the line. But if each voter has a perceptual error, then he’s only going to get maybe 55, 57% of the vote, something like that. He’s not going to get to 75%.
I have long assumed—and I may have written—that the Hall of Fame’s original sin is the creation of multiple entry systems. If you have different panels or different groups selecting Hall of Famers, they inevitably are going to use different standards. That’s unavoidable. I have long believed that THAT was the original cause of the Hall of Fame’s problems, the two-panel solution.
But now I realize that there is another flaw that goes before that, and which did in fact occur before that in history. It’s the 75% standard.
If we assume that
a) the standard for future Hall of Fame selections is an average Hall of Famer, and
b) different voters will have different opinions of each candidate,
then what that means is that an average Hall of Famer will be rejected in future voting. An average Hall of Famer hits .300 with 400 homers and 1300 RBI, let’s say. A player then appears on the ballot with a .300 average, 405 homers and 1315 RBI. He’s not going to make it. It’s inevitable. Because of the perceptual error, he’s NOT going to get to 75%.
And what that means is, you will have to create a clean-up system to pick up the players who SHOULD have been selected, because they do in fact meet the Hall of Fame standards, but who were NOT selected because the system required 75%. And that’s what happened. That’s what has happened again and again. We have tried to close that back door again and again, but it won’t stay shut. It CAN’T stay shut. It can’t stay shut, because the system is designed to reject some players who SHOULD be elected.
We all understand WHY the 75% rule was adopted. The 75% rule was a way of saying "we don’t want anyone in the Hall of Fame unless we are SURE that he belongs." But the problem is, that works too well. It overachieves its intention, thus causing a blowback. It forces you to cover for the players who are wrongly rejected. For every action there is an equal and opposite reaction. The problems in the Hall of Fame voting structure are an inevitable REACTION to the success of this provision.
Irrelevant Addendum
Anybody notice that Jim Sweeney, point guard for the Boston College basketball team in the 1978-79 Point Shaving Scandal, looks just like Pete Rose?