Major League Baseball

The three people you meet in Hall of Fame voting

Published Dec. 10, 2015 1:16 p.m. ET

share

facebookxredditlink

Sam Miller

Last year, 549 BBWAA members cast ballots for the Hall of Fame, and each of those 549 was a unique and beautiful snowflake, with his or her own decision trees and philosophies and personal memories swaying them hither and thither regarding each player's merits. Those 549, however, are also a collection of traits and demographics that cluster in sometimes predictable ways.

Thanks to the dogged work of Ryan Thibodaux, who is collecting and publishing the ballots of as many writers as he can track down, we know what 330 of those 549 ballots looked like. We know, for instance, that at least 13 voters submitted exactly this ballot:

● Pedro Martinez

● Randy Johnson

● John Smoltz

● Craig Biggio

● Jeff Bagwell

● Barry Bonds

● Roger Clemens

● Mike Piazza

● Tim Raines

● Curt Schilling

Is it some incredible mathematical fluke that, with 34 names on the ballot, these 13 writers chose the exact same 10? Of course not. These writers (who include Jayson Stark, Jeff Passan and Tim Kurkjian) followed certain logical steps that led them to the same logical place. Others followed logical steps that led them to other logical places.

There are only so many logical steps, though, and some of them are mutually exclusive. Which means we can actually learn quite a bit about a voter's ballot simply by knowing part of his or her ballot. Put another way: If I told you that a person had filled in the first nine names of his ballot like this:

● Pedro Martinez

● Randy Johnson

● John Smoltz

● Craig Biggio

● Jeff Bagwell

● Barry Bonds

● Roger Clemens

● Mike Piazza

● Tim Raines

You could deduce that he's more than four times as likely to fill in the final spot in his ballot with "Curt Schilling" than with "Edgar Martinez," and more than six times as likely "Schilling" than "Mike Mussina," and 13 times more likely "Schilling" than "Lee Smith" -- even though, in the voting pool as a whole, Martinez, Smith, Mussina and Schilling got comparable vote totals last year. Interesting. More interesting, we can use these demographic clusters to figure out which candidates will benefit from this year's pared-down HOF voter pool.

So, then, here we go: Three People You Meet In HOF Voting:

1. The Barry Bonds Voter 

This is the easiest one. Nobody could argue Bonds' performance wasn't HOF-worthy. Even before he could be linked to performance-enhancing drugs, he was one of the best players in history, producing more WAR (by Baseball-Reference's model) as a Pirate than Jim Rice did in his career. In a world without PEDs, Bonds would be a first-ballot inductee closing in on unanimous consent. That only 37 percent voted for him tells us convincingly (and I'm not telling you anything you don't know here) that those other 63 percent were unwilling to vote for a steroids cheat. Virtually everybody who voted for him voted for Roger Clemens; virtually everybody who voted for Clemens voted for him. This is simple.

What's more interesting is to see who else the Bonds voters liked and didn't like. A person who voted for Bonds was much more likely to vote for:

Mark McGwire: 19 percent, compared to 6 percent of Bonds non-voters

Mike Piazza: 84 percent to 69 percent

McGwire, sure, obviously. Piazza, who never failed a test or appeared in a report, is more interesting. We know, because they've told us, that some voters suspect him of cheating, and have held votes back on a speculative basis. The Bonds Voter breakdown shows us just how many of those writers there are. In a world where nobody held suspicions against Piazza -- either because there were no suspicions, or because everybody would just be cool -- €”he would have been inducted last year.

Meanwhile Bonds voters were much less likely to vote for:

Edgar Martinez: 15 percent to 36 percent

Lee Smith: 12 percent to 33 percent

Mike Mussina: 22 percent to 35 percent

While the weak showing for Smith might lead you to think the Bonds voter is also a stathead voter, Martinez and Mussina are solid WAR candidates. The fact that Bonds voters don't like them tells us Bonds voters are not, as a group, a strongly pro-WAR voting bloc. This is a bit of a surprise; stathead writers, historically, have generally skewed toward an anti-moralistic way of viewing players, even cheating players. But the clustering here makes a good case that, in fact, the decision to use steroids as an automatic disqualifier has nothing to do with analytics. Some voters simply make the call, for reasons that are between them and their gods.

2. The Curt Schilling Voter

The Curt Schilling voter is far more likely than the Schilling non-voter to vote for:

Jeff Bagwell: 72 percent to 51 percent

Mike Mussina: 42 percent to 20 percent

Tim Raines: 72 percent to 50 percent

Alan Trammell: 29 percent to 22 percent

and less likely to vote for

Jeff Kent: 9 percent to 18 percent

Lee Smith: 18 percent to 31 percent

Sammy Sosa: 3.3 percent to 6.6 percent

Here's what's interesting about Curt Schilling: He produced 80.7 bWAR.  That puts him 12 wins ahead of any pitcher in history who isn't either in the Hall of Fame or on this year's ballot. He's a WAR superstar. He's a WAR no-brainer. He also appeared on only 39 percent of all ballots last year, because some people just really don't care about WAR.

But some do, and the rest of the Schilling voter's ballot is proof. Here are last year's candidates, sorted by bWAR:

Bagwell, Mussina, Raines and Trammell are all strong WAR candidates -- especially if you're a voter who leaves Bonds and Clemens off. They also produced a grand total of one MVP or Cy Young Award, and danged if I can think of a single record any of them holds. Kent, Sosa and Smith are bad WAR candidates. They also have, in just their little group, two MVP Awards; one of them once (briefly) held the all-time single-season home run record, and another of them held (for a while) the all-time saves candidate. It's the Hall of Fame, not the Hall of WAR, somebody has probably written in a first-draft HOF column.

3. The Short-Ballot Voter

Many voters last year complained, and will complain this year, that there are more worthy candidates than there are ballot spots, thanks to the "No Juicers Get In!" backlog. But some voters find a way to choose just seven, or four, or even two candidates. Of those who voted for seven or fewer, the preferred candidates (relative to the rest of the voters) are Fred McGriff, Alan Trammell and Lee Smith. They especially hate Bonds/Clemens, Larry Walker and Jeff Bagwell.

These are strange tendencies. Bonds/Clemens, sure. But Walker? Bagwell? McGriff!!? One explanation goes like this: This is a protest vote. Not finding 10 (or even eight) candidates on a list of qualifiers that arguably went 13 or 14 deep shows either an intent to punish an entire generation or, more likely, and more generously, a nostalgic preference for an older era. A longing for an era without graphics on the TV screen, without cameras in the bases, without three pitching changes in the middle of the eighth inning, without four rounds of playoffs, when players were chastised for lifting weights, when fastballs weren't so darned fast, and when writers could call the second baseman "gritty" and the center fielder "lazy" without some blogger accusing them of racial stereotyping. You see the same thing in the Lee Smith voters:

Love: Mattingly, McGriff, Trammell 

Hate: Bonds, Mussina, Schilling, Sosa 

There are voters who just don't like post-1993 baseball. They find a way to make room for Fred McGrif -- maybe because his black-ink seasons all came before 1993 -- and they still don't generally like Tim Raines, for reasons I'm not clear on.

The Hall of Fame purged scores of non-active writers from the voter rolls this year -- basically, writers who haven't actually covered the sport in a while. Thibodaux has confirmed around 30 of the purged, of whom 21 had a public ballot last year. So what can we learn about the purged voter?

● They aren't Bonds voters. Just 33 percent voted for Bonds, compared to 39 percent on all public ballots.

● They aren't Schilling voters. Just 24 percent voted for Schilling, compared to 45 percent on all public ballots.

● They are short-ballot voters. Only 29 percent voted for 10 names last year, compared to 57 percent of all public ballots.

Thibodaux speculates that the purged voters he hasn't confirmed will skew even more toward these tendencies.

By this point in the article, we now know what these tendencies mean for Hall of Fame candidates. They mean we have a voter pool that is less absolute about steroids usage, that is more likely to use WAR to guide their decisions, and that values the post-1993 era of baseball. Depending on what sort of logical steps lead you to your decisions, this might qualify as progress.

All praises due Ryan and the Play Index at Baseball-Reference, which provided all the tools necessary for this piece.

share

Get more from Major League Baseball Follow your favorites to get information about games, news and more

in this topic

Major League Baseball