you're reading...
2019 Hall of Fame Election, Sidebars

Comparing the Hall of Fame Voters

In a series that has gone on far longer than I anticipated, today I bring to you a review and a bit of analysis of votes per BBWAA ballot. Generally speaking, I believe ballots with more players are better than ballots with fewer. There were fifteen guys the voters were allowed to review who I support, and there were a handful of others who a reasonable person might. Thus, I would argue a ballot that included fewer than ten names was a mistake. Basically, the fewer the names, the bigger the mistake. I suspect you’ll see some clear trends below, and even if you don’t, I’ll be sure to point them out for you.

At this point, I don’t imagine you’ve missed any part of this series you want to see. But if you have…

[Andy Pettitte], [Scott Rolen], [Andruw Jones], [Larry Walker], [Billy Wagner], [Todd Helton], [Manny Ramirez], [Jeff Kent], [Omar Vizquel]

A Big Chart

We’re going to start with an overall chart. Yes, it’s big and not so easy to read. Alphabetical order doesn’t really make things easier either. The real thing I want to point out here is the public (8.25), private (6.72), and overall (8.01) average names per ballot. You might wonder why all of those numbers are lower than even the lowest number for any individual player. First, our individual players numbers are all public ballots, so we should focus only on the number 8.25. There’s a second reason too. There might be a mathematical term or explanation for this phenomenon, but I don’t know what it is. Let’s take an example though. There are 9.17 names on each public ballot that has Barry Bonds’ name checked. But for those 222 ballots, if you eliminated Bonds (as is the case with all ballots without Bonds’ name checked), there are only 8.17 names per ballot, or below the public average. In other words, the average of all ballots was 0.08 names higher than those ballots with Bonds’ name, minus the credit for Bonds. This little exercise doesn’t take into account that there are plenty of Bonds ballots that would include another name if they were allowed. That’s okay though. It tells us something.

Overall Chart

By Candidate


The thing to see here is pretty simple. At least as I would argue it, generally speaking, the more names that are on your ballot, the better you are as a voter. Jones and Rolen voters stand above others by this measure. Those who support Ramirez, Sheffield, Walker, Sosa, Bones, Pettitte, Wagner, Clemens, and Helton make us the next level. The third level includes those who voted for Schilling, Kent, and McGriff. I explain in a section below that I don’t give Schilling voters a hard time for finding themselves at this level. The fourth level sees only Vizquel. His supporters thought less of the candidates overall than the supporters of any other candidate.

By Candidate

Supporting 10

This next chart, as with all that follow, shows the percentage of ballots with a particular player’s name checked. In this case, it’s those ballots that have the full compliment of ten names checked. You can see that over three-quarters of all Andrew Jones’ public votes have ten names checked. Almost three-quarters of Scott Rolen’s public votes also checked ten. There’s a sizable drop after those names. Five of the next six players on the list are players associated with PEDs. It’s not surprising that this group tops others because those who won’t vote for those associated with PEDs have fewer qualified candidates from whom to choose. Those who will have an easier time getting to ten names. To me, those facts speak very, very well of Jones and Rolen voters.

I think you also see some interesting stuff at the bottom of this chart. Andy Pettitte voters could only find nine other qualified names half the time. That’s a shame to me since Pettitte was associated with PEDs. What I mean is that if you vote for Pettitte, you must vote for Bonds and Clemens. Also, Pettitte was clearly a lesser pitcher than Roy Halladay, Mike Mussina, and Curt Schilling. So we’re up to six now. Mariano Rivera makes seven. That means half of Andy Pettitte voters disagree with something above, or they thought Pettitte was better than all but two of of Ramirez, Sheffield, Sosa, Edgar, Walker, Rolen, Helton, and Kent. In other words, Pettitte voters struggled to do their jobs well.

The last thing I want to point out is the support Omar Vizquel voters give to other candidates. It’s below the average, which is just about impossible. For whatever reason, I’ve read a lot about the Dunning-Kruger Effect recently, and I think it applies here. Basically, those who have a lot of information doubt and question their information quite a bit. For example (patting myself on the back), I’ve wavered on Manny, Pettitte, Sheffield, Kent, and even Larry Walker in the past. (My Walker wavering was in the long ago past, and I understand my former arguments were poor). Dunning-Kruger goes on to hypothesize that those who know much less are far more assured in their opinions. Those who just know Omar Vizquel was a defensive genius on par with Ozzie Smith or that Jack Morris was the pitcher of the 80s are not influenced by facts to the contrary. (Note: I realize I might be bastardizing Dunning-Kruger to a degree, but I believe I am generally representing it correctly).


Supporting 10 or 9

I am very happy to see on this chart that Andy Pettitte voters aren’t quite as bad as represented on the full ballot chart. Lots of them support exactly nine candidates. You also see the PED candidates continue to be bunched, the Jones and Rolen voters still at the top, and the Vizquel voters continuing to do a terrible job. Let me explain what Omar’s 60.4% number means. It means two in five of his supporters thought that there were six or fewer deserving candidates on this ballot other than the unanimous Mariano Rivera and their guy, Omar.

10, 9

Supporting 10, 9, or 8

Continuing our journey, you see that more than 19 in 20 Rolen and Jones voters thought there were at least seven other worthy candidates. About nine in ten voters for eight other candidates thought there were at least seven worthies other than their guy. Then we see a drop. Fred McGriff voters, in my opinion, were wrong. He got a pop in his final year on the ballot that was inconsistent with his record. It makes some sense that Billy Wagner voters might be in this range. Some simply believe he was one of the very best closers of all-time, that closer is a position, and that there weren’t more than six others who were among the very best of all-time at their positions. I think that’s wrong, but it’s not illogical. I don’t mind that Curt Schilling voters are down here. I think there’s a real argument to be made that Schilling was the single best player on this ballot not associated with PEDs. A smaller Hall, non-PED supporter might stop not too many names after Schilling. Jeff Kent voters are just strange – or at least I can’t quite figure them out. And then we have Omar voters. More than a quarter of them think Omar, the unanimous guy, and five or fewer others are the only ones on this ballot who had careers worthy of enshrinement.

10, 9, 8

Supporting 8, 7, or 6

To me, these are bad ballots. I can make an anti-PED case for supporting only nine. It’s not a good case, but I can make it. I can make no such case for eight. And what do you know! Omar is at the top.

8, 7, 6

Supporting 7 or Fewer

This is just the reverse chart of the one above showing support for 10, 9, or 8. Please excuse rounding errors.

7 or Fewer

Supporting 6 or Fewer

Now this is just funny. Nearly one in five Omar voters find their guy, Mariano, and four or fewer others of Hall quality. Check out the Rolen voters. I’d have to call them the best voters of 2019.

6 or Fewer

Supporting 5 or Fewer

If you’re not angry yet, your chances of dying from some stress-related malady are lower than mine.

5 or Fewer

In a week, we’ll close out this series, concentrating on the players you support if you support a particular candidate.




4 thoughts on “Comparing the Hall of Fame Voters

  1. In summation, then, voters who study the Hall of Fame’s membership and use comparative analysis understand that guys like Rolen and Jones exceed Cooperstown’s standards quite handily. “Small hall” voters – those who are either woefully ignorant or intentionally superimpose unrealistic personal standards on the process – demonstrate that they are as poor at performance analysis as they are comparative analysis.

    This is pretty damning evidence that the rift between full ballot and partial ballot voters is indicative of larger trends between objective and subjective standards, between analysis and the eyeball test, between critical thinking and personal bias.

    Can we have them publicly flogged until they repent?

    Posted by BigKlu | April 15, 2019, 9:03 am
    • Yup, you’ve described today’s Hall state of affairs quite well.

      The magic voters are flogged verbally every winter. Unfortunately, when you actively dismiss (if that’s a thing) whatever comparative analysis folks say, that criticism is a badge of honor and little more.

      Posted by Miller | April 15, 2019, 1:43 pm
  2. Can hardly wait, by the way, to see who the new Negro League pick is. I know it’s less than 24 hours until the reveal, but I’m on pins and needles.

    Posted by BigKlu | April 18, 2019, 5:43 pm
    • Thanks for the support! I don’t suspect the early inductees will be much of a surprise to those aware of Negro League history. Later in the process, however, things will likely get interesting. Really excited by Eric’s work on this!

      Posted by Miller | April 18, 2019, 9:54 pm

Tell us what you think!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Institutional History

%d bloggers like this: