Of course the internet baseball community hates Barry Bonds. Despite being clearly one of baseball’s greatest players, his career will always be overshadowed by his presence as Mr. BALCO, as the center of the steroids scandal which loomed over baseball for much of the last decade. Not only that, but he had a tendency to come off as short and angry with the media. A combination of cheating and nastiness hardly endears people to your cause.
But, naturally, I am not fully satisfied with anything until I can quantify it. Luckily, the Baseball-Reference Elo player rater exists. With the help of the users of Baseball-Reference, who have cast over one million votes rating players, we can see just how the internet baseball community (at least, that rather large part that uses Baseball-Reference) view players. Voters are welcome to use objective criteria from WAR to batting average, and obviously nothing is preventing them from hitting one player’s button if they simply like that player better. To me, finding a subjective measure of the opinions of greatness to put next to our objective measures is incredibly interesting.
Make no mistake: Barry Bonds is clearly one of the top hitters — if not the top hitter — to ever play the game. Between the home run record, 175 career wRC+, 168 career fWAR, and 172 career bWAR, it is impossible to argue with his on-field accomplishments. So when we see the Baseball-Reference crowd has him rated as the 26th best player of all time, right between Cal Ripken Jr. and Ken Griffey Jr., we know it goes beyond what’s on the field. Observe, the massive difference between how Bonds is rated by the people and how we would predict he would be rated simply by WAR (using the Baseball-Reference version as it is the one presented to the voter), looking at the top 50 players in the rankings:
For a better look, here are the distances from the trendline plotted against rating:
Click on either image to embiggen, or click here for an interactive version of both in which you can see which marks represent which players.
It is clear that WAR is not the only thing taken into account by the voters — nor should it be — but it is a major factor. The r-squared on the line of best fit is a solid 0.64, suggesting WAR can explain the majority of why each individual vote is made. However , we see Bonds, the second-ranked player by WAR, sitting in 26th place and more than twice as far from the line as the next farthest player (Albert Pujols, presumably because his excellent career may only be a bit over half over). The difference comes out to a whopping 80 WAR.
To a lesser extent, we see this happening with a few other steroid-linked players. Rafael Palmeiro and Jason Giambi, for example, are about 40 points of Elo rating lower than where their WARs would suggest. Roger Clemens (and, oddly, Cy Young) see a similar effect to Palmeiro and Giambi, as can be seen here. Oddly enough, we don’t see any impact with Alex Rodriguez. He features prominently in the rankings at number 18, with WAR predicting a rating only 2 points higher than his actual rating — perhaps the active career bias we see with Pujols is cancelling out the effect of steroid scandals.
Take this for what it’s worth — the users of Baseball-Reference are only a subset of the baseball community as a whole and the internet baseball community as well. But I think it is a representative subset of many of those who would read this page and many of those who consider themselves fans of the game as a whole. When we see Barry Bonds come on a Hall of Fame ballot and perhaps fail to earn induction, remember it isn’t just the voters who penalize him for his involvement in the steroid era.
Print This Post