November 15, 2014

Simulations and Plate Discipline

This Tom Tango post about batters reaching base against a consistent pitching machine got me thinking about plate discipline. Tom basically asks the question, if two teams face a consistent pitching machine and one does much better than the other, how much is attributable to the machine, and how much to the batters. The machine is consistent in that it throws only fastballs, but the fastballs find the strike zone 50% of the time.

This would be an interesting process to simulate. The pitching machine is a random number generator that half the time produces a pitch in the zone, and half the time out of the zone. From FanGraphs, we have access to plate discipline statistics. So, for example we could create a team of players who are good at reaching base, a team of players that are poor at reaching base, and run a simulation based on these stats, seeing what the long-term expected outcome of times reaching base is for both teams. If we record the result of the pitches for each game, we can see if the pitching machine had a good game (threw a lot of strikes), or if the batters performed poorly (swung at pitches outside the strike zone more than usual).

This also got me thinking about what are differences in plate discipline between players who get on base a lot and players who don’t. Here are two extreme teams. I basically looked through the list of batting title qualifiers and created a team of the best and worst on-base percentage at each position. I was flexible with the outfield assignments. Jay Bruce was simply the worst player whose position was already filled in, so I used him as the designated hitter.

The result is two teams of everyday players, one really good at getting on base, and one really bad. The average OBP of the good team is .389; of the poor team .281. So for every 100 plate appearances, the good team will get on base 11 more times than the bad team, 6 or 7 times more in a 27 batting out game.

If you look at the average line, what becomes clear is that the high OBP teams swings less. They swing less both inside and outside the strike zone. Note that the outside the zone swing rates are close. It’s not like the batters who don’t get on base well can’t recognize a ball versus a strike. When they do swing, the poor batters miss a lot more, especially on pitches outside the strike zone.

So the difference between players who excel at getting on base and those that don’t isn’t so much strike zone recognition as it is recognizing pitches that will lead to a high contact percentage. They are willing to take pitches in the zone that are not the best for them to hit. The also have the ability, I suspect, to foul off pitches on the edge of the zone.

Maybe tomorrow I’ll do the same analysis for slugging percentage and see what that produces. I might try to write a simulator as well this winter.

Leave a Reply

Your email address will not be published. Required fields are marked *