Reaching on an Error Should Improve OBP
Right now it does not, and that is very silly
There’s a war going on beneath the surface of the statistical community. You may not see it — you’d have to be quite a nerd to do so — but nonetheless, it is there. The forces are drawn up on two separate sides of a very simple question. What should happen to a batter’s On-Base Percentage when they reach base as result of an error? Statistician Tom Tango put out this Twitter poll yesterday:
The opinion is split evenly among the options. Interestingly, the answer that received the fewest votes, “Go Down,” is what currently happens in the described situation. I voted the option labeled “Go up, obviously,” and will defend that choice.
First and foremost, we must look at the statistic in question: On-Base Percentage. The name implies that it tracks the percentage of time a batter ends an at bat standing on base. When the batter reaches base on an error, they end up standing on base. Does that not make the case clear enough? Despite this logic, under the current system the player in question gets an excellent view (while standing on base) of his OBP sliding down a few ticks. To make the point more clear, here is a flow chart showing how On-Base Percentage should logically be calculated.
Nice, simple, and logical — just the way the stat should be.
The argument against this line of reasoning is that reaching on an error is not demonstrably a result of the batters skill, but an arbitrary result of the defense being played against them.
We’ll get to the flaws in this argument later, but first, assume it’s true that batters have no control over errors. It still doesn’t matter. OBP is a results-based statistic — it tracks what happened, not how it happened. One must look deeper than AVG (average) to see how many singles, doubles, and triples were hit. OBP says nothing about whether the hitter reached because of a hit, walk, or was hit by the pitch.
Why separate only reached on error? The comparison to Batting Average is further apt because a hit is quite often dependent upon the defender. If a batter hits the ball into right field while Jason Hayward is there, it is far less likely to land on the turf than if J.D. Martinez is. The batter does not control this, yet is still credited with a hit. If we pick and choose which results matter to a statistic, it significantly weakens the stat.
Now, to the flaws in the original argument, or more accurately flaw since there is really only one. Batters have the ability to force errors. Very few skills in baseball can be defined as being available a will, but there are certain attributes that make errors more likely to happen depending on the hitter.
The first is having a batted ball profile skewed towards hitting ground balls, since grounders produce significantly more errors than any other batted ball type. This effect is best seen when combined with above average speed, since the infielders will have less time to make the play and thus are more likely to make mistakes. Players like Ichiro and Dee Gordon see far more of their batted balls turn into errors than a player like Yadier Molina.
Second, a player can force more errors is by hitting the ball hard consistently. Last season on average the baseball came off of Giancarlo Stanton’s bat at 95.1 MPH — much faster than Ryan Hanigan’s 82.2. The faster the ball moves, the harder it is to field, producing a clear relationship between skill and higher error rates. Some errors obviously happen without being caused by the batter, but that’s baseball. Bartolo Colon hit a home run once — should it not count because he‘ll probably never do it again?
At the end of the day, OBP is a simple stat and should not be unnecessarily complicated. The question should simply be, “Is the batter on base?” A separate statistic can always be kept to explain how they got there.
