A Vote for Transparency: How Secret Ballots Skewed the Hall of Fame Election Results

Lewie Pollis is a senior at Brown University and a former baseball analytics intern for the Cleveland Indians. He also writes for ESPN Insider. Follow him on Twitter @LewsOnFirst.

The year was 2011. I was a green-behind-the-ears aspiring sabermetrician who would pore over every little piece of baseball-related data I could get my hands on in an attempt to better familiarize myself with the numbers. So as I sat on a Greyhound bus with nothing better to do on a dreary January afternoon, I found myself looking over some pre-announcement ballot counters for the newly minted Cooperstown class of 2011.

As my gaze moved back and forth between the spreadsheets and the fogged-up bus window, I gradually came to a realization: the writers who made their Hall of Fame ballots public had voted differently than those who kept their choices secret. Not only that, but from my perspective, those who opened up about their picks had voted more correctly than those who remained anonymous. This phenomenon occurred again in 2012, and then again in 2013.

Now that the 2014 results have been released, we’ve got some good news and some bad news. The good news is that the influence of secret ballots on the Hall of Fame vote seems to be smaller than it was last year. The bad news is that secret balloting still very much matters, and this year it actually seems to have made a difference in the results.

Last year on its website the BBWAA posted a list of every voter who made his or her ballot public shortly after the results were announced, but as of this writing they have not yet done so. Instead, I copied down the last pre-announcement vote totals from Baseball Think Factory’s HOF Ballot Collecting Gizmo and leokitty’s collection of ballots few minutes after the results were announced; using these lists of ballots instead of the BBWAA’s is not ideal for this exercise, since Baseball Think Factory does not list individual ballots and leokitty does not have listed sources for all of hers, but since the two had virtually identical results they seemed like acceptable alternatives. I used Baseball Think Factory’s numbers for this analysis because its Ballot Collecting Gizmo had more ballots (205 ballots to 184).

In the 2014 Hall of Fame vote, 22 players received at least seven votes—we’ll call them the serious candidates for induction. Extrapolating the results for anonymous voters from the full results and the Ballot Collecting Gizmo, here’s how the two groups voted on the serious candidates:

A few quick observations:

• If everyone had voted like the public voters, Craig Biggio would have been elected to the Hall of Fame this year.
• If everyone had voted like the public voters, Rafael Palmeiro would still be on the ballot for 2015.
• Assuming no public voter cast his ballot for a candidate who finished below Palmeiro, but counting private balloters who did, those who opened up about their votes checked off 0.7 more names than their more secretive peers, with an average of 8.84 votes per ballot compared to 8.14.
• Anonymous voters were almost twice as likely to vote against Frank Thomas, more than twice as likely to vote against Tom Glavine, and more than eight times as likely to vote against Greg Maddux as those who owned up to their ballots.
• Secret balloters were more than twice as likely to vote for Don Mattingly.
• Jack Morris beat Mike Piazza among private voters.
• Despite finishing higher than they in the total vote, Lee Smith finished behind Curt Schilling, Edgar Martinez, and Mike Mussina on publicly listed ballots. However, he beat all three of them as well as Roger Clemens and Barry Bonds among anonymous voters.

The good news is that the differences between public and private voters’ support for players are generally smaller than they have been in the past. The 12-point gap between Bonds’ public and private vote shares is the largest on this year’s ballot, but it would have ranked sixth-largest in 2013. We saw a full 16-point differential for Jeff Bagwell last year; now that gap is statistically insignificant. Perhaps most impressive is that the ever-polarizing Morris, who fared 11 points better with anonymous voters than with public voters in 2013, got identical support from both groups this year.

But that the situation has improved does not mean it has been fixed. As you can see, some of the gaps are so large as to be statistically significant, meaning we can say with a high degree of confidence that the two groups of voters were truly evaluating the candidates differently. We can say with greater than 99-percent confidence that Bonds, Tim Raines, Curt Schilling, Mike Mussina, and Thomas were evaluated more skeptically by those who kept their ballots secret, and we can be at least 95 percent certain that Smith, Piazza, Clemens, Glavine, Mattingly, and Maddux were considered differently by the two groups.

Though the pattern is less pronounced than it has been in years past, at first glance it also appears that there is a relationship between how deserving a player is of enshrinement and how much more support he received from the public voters than from the private voters. Each of the top five finishers received significantly more support from the open writers than the anonymous ones, and five of the six serious candidates whom the latter group liked more than the former finished in the bottom seven. Put it another way: 15 of the 19 candidates for whom I would have voted in a no-maximum ballot got more support from the public crowd, but none of the three I would have left off (Morris, Mattingly, and Kent) did.

It turns out this relationship is more than just anecdotal. The correlation between the voting discrepancies and Jay Jaffe’s JAWS scores is 0.64, with a 0.62 correlation between the public-private gaps and Adam Darowski’s Hall Ratings; assuming these are good proxies for Hall of Fame candidates’ worthiness, approximately 40 percent of the variation in how the two groups of writers voted for each candidate can be explained by how deserving he was of enshrinement.

I don’t think the relationship is entirely causal. Despite my having been overly dismissive of the idea in years past, I think it’s fair to say that the writers whose ballots have not been revealed are more likely both to be retired and to think about the game in terms of older and more subjective measures that those of us in the sabermetric community do not consider. (Not that I’m advocating for pure objectivity in the Hall of Fame conversations. After all, when I was filling out my ballot for a mock vote last week, I almost checked the box for Eric Gagne.) But to some degree, I think there is some not-insignificant number of voters who do not take their ballots as seriously as they should because they know their names will not be attached to them.

Take Greg Maddux. There is no good reason why he should not be in the Hall of Fame. (If you have one, tell me. I’d actually be very interested to hear it.) Unless you were to leave him off your ballot because you knew he was getting in and you wanted to use your 10th vote on someone for whom it would matter more, there is no excuse for not checking his name. Yet while only one writer admitted to not voting for him, 15 voters did under the cover of anonymity. Would all 15 nameless naysayers have voted for Maddux if their names were on their ballots? Probably not. But surely a few of them would have voted differently if not for their anonymity.

The story of the lone named Maddux denier is instructive. In case you haven’t heard, it was MLB.com Dodgers beat reporter Ken Gurnick who owned up to checking Morris’ name and no one else’s. It’s not worth pointing the flaws in his argument or the double-standards of his own voting record because the rest of the blogosphere already has ad nauseum, and that wave of mean-spirited criticism is probably part of why Gurnick has said he will never vote for the Hall of Fame again. It’s unfortunate that he’s been so bullied for his opinion, but if you take away the malice, that’s exactly what is supposed to happen. If a professional sportswriter cannot satisfactorily defend his opinions about the game he covers, he or she deserves to be challenged on it.

There is no reason that revealing one’s ballot should not be mandatory for all voters. I don’t think all or even most of the BBWAA veterans who did not reveal their ballots publicly failed to do so because they were embarrassed about their choices, but if even a few did, that is unacceptable. Voting for the Hall of Fame is a privilege, not a right, and if you make a living writing about baseball and you are incapable of convincingly arguing for your choices and snubs, you voted wrong.

In a world in which writers had to make their ballots public, Craig Biggio would probably get to stand with Maddux, Glavine, and Thomas on the Cooperstown steps in July. Protecting the identities of the people who snubbed him—many of whom will probably end up voting for him next year—is not reason enough to keep him out.