FanPost

An Attempt to Rationalize Jon Heyman's HOF Ballot

When I was in college a job recruiter told me that they reviewed peoples' social media accounts before deciding which candidates would receive job offers. When I got home that evening I made a twitter account in hopes of using it to dupe some employer into giving me a job offer. I followed scientists to display my passion for the sciences. I followed politicians and non profits to show my social awareness. And of course, I followed the Houston Astros to show that interests outside of school and work. My tweets were even worse, as they revolved around studying, working hard and getting good grades. My friends would always get a good laugh at my tweets seeing how painfully obvious it was as it was as much of a joke as it was serious. I have no idea if it helped, but it sure didn't hurt as I was able to find employment right out of college. I didn't stop using twitter once I got a job, but now I primary use twitter as a tool to follow baseball.

This brings me to my dailly routine where the first thing I do when I get home from work is I open twitter to see what is going on in the world of baseball. Being that it was January 3rd, I wasn't expected much. But much to my surprise, my Astros twitter follows were going wild. In the span of just a few minutes I saw claims that Jon Heyman is a known plagiarist and that he has almost certainly done coke. These hyperbolic jokes could only mean one thing, and sure enough when I went to Jon Heyman's twitter he had a link posted about his published Hall of Fame ballot.

So far, there has been 163 ballots made public. Of those, 12 besides Jon Heyman have left Bagwell off their ballot. But none of those 12 have spurred nearly the response that Heyman's did. I believe this is because Heyman is still a very popular media personality in baseball, he voted for Bonds, and what I believe is even more important, he gave a detailed explanation as to why he voted the way he did. His post showing and explaining his picks an be found here. I would suggest reading it through. Or don't, I don't care.

So let's begin to unpack his ballot and talk about the logic he used to make his picks. For those of you who didn't read the article... here are his picks: Guerrero, Rodriguez, Bonds, McGriff, Hoffman, Mussina, Raines, Smith, Posada, Martinez. By far, the most interesting pick on this list is Barry Bonds. Heyman's inclusion of Bonds on his ballot shows that Heyman has no problem voting for someone who takes steroids, or more accurately, it shows that he does not find the use of steroids to be damaging enough to a players integrity, character, or sportsmanship to leave them out of the hall of fame. For the record, I feel the same way as Heyman here.

This is obviously where things get strange, and what made Astros Twitter go bonkers. There are usually three types of voters:

1. Voters who do not care about who took steroids (who vote for Bagwell and Bonds).

2. Those who were proven to have taken steroids have do not qualify has HOFers because of the integrity, char... blah blah blah (Who vote for Bagwell and not Bonds)

3. Same as voter 2, but they don't vote for anyone who they think may have possibly taken steroids (Bagwell and bonds both aren't voted for here.)

If you notice, there aren't any scenarios where Bonds is on a ballot and Bagwell is not. Heyman is a strange mixture of voter 1 and voter 3, who are usually polar opposites. Just looking at his ballot, it doesn't make a lot of sense, he obviously doesn't think people that took steroids she be in the hall (no Bagwell/Clemens), but then there, near the top of his ballot, is an checked box next to Barry Bonds. Fortunately, Heyman did what so many confusing voters don't do, he spent a significant amount of energy explaining his thought process.

For those that didn't read the ariticle I can try to some up what I believe is gist of his logic. As we already conjectured, the steroid question for him does not come down to integrity. For him, what it comes down to is understanding how much of a players talent was him, vs how much a players talent was steroids. This is absolutely fascinating to myself and it is the reason I'm writing this post. Though Heyman raised this question, he seemed far less interested in thinking about it. For Heyman it was simple, if a player took steroids, it is impossible to know what he would have been without them, and if he doesn't know how good they would've been without the steroids, he doesn't vote for them. Bonds gets a pass for already being HOF worthy before 1998, which is when Heyman believes Bonds started using steroids.

I have no problems with him voting this way. Truth is, if a guy does steroids you don't know what he looks like in the universe where he didn't do steroids. But that is no fun so I'm going to try something else, more on that later. The problem I do have with this methodology is the binary assessment of the probability someone did steroids. He either thinks someone did use steroids, or he thinks someone didn't. In all likely-hood, he probably doesn't think this way, but as a journalist, he can't really say, "I'm about 60% sure Bagwell did steroids, therefore, I don't know how good he would've been so I'm not voting for him." So what I'm attempting to do is to help Heyman. I want to make an empirical tool to explain his ballot.

First some assumptions that make this a dumb, pointless, stupid waste of time.

1. I'm going to use fWAR. In the article Heyman talks about not completely believing in WAR. Of course he says something about Lou Whitaker vs Jeter and therefore he doesn't vote for Larry Walker? Yep, I'm definitely wasting my time.

2. To find out how much people are helped by steroids I'm going to use Jon Heyman's Ballot. This is dumb, this is wrong, but it's an attempt to rationalize his ballot, so I try to find a way to use his ballot to do something impossible.

3. I'm going to assume Heyman almost voted for Clemens. Probably the worst assumption on this list, and also the most important. If you stop reading now I don't blame you and I apologize for wasting your time.

4. I'm going to pretend Heyman feels the same about the integrity question with Clemens and other steroid users. If you read the article, you'll find out this assumption is just flat out wrong, but I don't care. I don't see him being that much more shitty of a person than Bonds.

5. I'm assuming everyone's increased performance from steroids. I could vary this by person, but it makes this task more complex as are the answers. Also, it's more easy to input your own biases on who you assume benefited more or less from steroids.

6. Catchers need 40 WAR to get in, Pitchers and other position players 52. This is roughly 80% of the average WAR for each group in the HOF. This was easier than looking at ever way at every position and doing some percentile that would make them eligible.

Now that I've convinced you I'm wasting your time, the first thing I did was I took the entire HOF ballot and filtered out everyone that doesn't meet the 4th assumption. This leaves 16 Players. 2 of Heyman's picks didn't make this list, Lee Smith, and Trevor Hoffman... stupid relief pitchers, let's ignore them for everyone's sake.

Of those 16 guys, there are (at least) three definite steroid users, Manny, Bonds and Clemens. Clemens did not get Heyman's vote. This is how I'm going to approach assumption #2 and #3 to get a number for how much steroids benefited the players that used them. These are not good assumptions but I need to use Clemens because I'm able to hold one variable constant, the probability that he took steroids and he was the highest WAR player to be left off of Heyman's Ballot. Assuming now Heyman almost voted for Clemens, that would mean that Clemens almost reached the cutoff of 52 WAR, but not quite, so Heyman would assume that steroids made Clemens 62% better.

If you apply this 62% to all remaining players, only Bonds gets in (Obviously since only Bonds has a higher WAR than Clemens on the ballot). So it clearly shows that if Heyman thought a player did steroids he could apply the 62% reduction and get a undeserving WAR, but if he doesn't apply the 62% that player is deserving. But that isn't exactly fair, because in all honesty, it is hard to know if a player did or didn't do steroids. Being 60% confident is a lot different than 99%, even though both scenarios would indicate that you think that person did steroids.

What we have now though, is a way to solve that probability. By multiplying the 62% regression from steroid use by the probability of needing to do that regression, whatever probability makes the players adjusted WAR to be equal to the HOF WAR cutoff is the critical probability that a person would need to think about when voting for the HOF. Below are the results.

2I7KhGQ.0.jpg

So there it is. So instead of arguing yes and no, Heyman could just be arguing what the right two columns show, whether he thinks someone was greater or less than the probability stated. Though he'd have to back up that 62% number which people would hate, but I actually really like it. Seems like a reasonable thing to do. So now he just has to say that he thinks there was a greater than 57% chance Bagwell took steroids.

And for the record 57% seems pretty high when you evidence consists of:

1. His HR total was 4 as a 22 year old in AA and then 15 as a MLB player at 23. (And his ISO went from .124 to .143 ... what a roided out machine, lol)

2. He was teammates with a guy who later did steroids in San Diego.

3. He was big, followed by a period of getting smaller after he couldn't work out due to injury.

4. In 2007 Jerome Solomon wrote that he took PEDs in a blogpost. This was not a blogpost about investigative journalism that uncovered his PED usage, rather, it was one sentence in a short blogpost about booing Barry Bonds lol. And then the very same Jerome Solomon wrote this...

Even if you are one of the lost ones who believes steroids users are vile human beings, life cheats and have no place in the Hall of Fame, there is no evidence Bagwell used performance-enhancing drugs. None.

Not exactly the evidence that I would say that there was a greater than 57% chance he did steroids. But I'm biased, so wtf do I know. Anyway, I'm sure I'm not alone but I'll be excited to see Bagwell in Cooperstown.