For this post, I will be using data from the 2008 and 2009 seasons because it is the most recent data I have at the moment. I am having issues updating my database and I hope to have more recent data in the near future.
When I first sat down to write this, all that I knew was that I wanted to do some sort of a post on pinch hitting. Pinch hitting strikes me as one of the most complicated parts of a baseball game, while also being one of simplest. If you're pitcher is having a bad day on the mound and it's his turn in the order to bat, you pinch hit for him. If the home team is down in the bottom of the ninth and it's the pitcher's turn to bat, you pinch hit for him. These are both easy decisions.
What complicates matters is if a team is down one or two runs late in the game, or even if a team is down five or six runs. What if you're down 6-1 in the 5th and the pitcher's spot is up? Is it worth it to pinch hit in such a low leverage situation? What if you're down 1-0 in the 7th and it's the pitchers turn to bat? Do you pinch hit for the pitcher to try to score and hand the ball off to a potentially worse pitcher?
When I first sat down to write this, all that I knew was that I wanted to do some sort of a post on pinch hitting. Pinch hitting strikes me as one of the most complicated parts of a baseball game, while also being one of simplest. If you're pitcher is having a bad day on the mound and it's his turn in the order to bat, you pinch hit for him. If the home team is down in the bottom of the ninth and it's the pitcher's turn to bat, you pinch hit for him. These are both easy decisions.
What complicates matters is if a team is down one or two runs late in the game, or even if a team is down five or six runs. What if you're down 6-1 in the 5th and the pitcher's spot is up? Is it worth it to pinch hit in such a low leverage situation? What if you're down 1-0 in the 7th and it's the pitchers turn to bat? Do you pinch hit for the pitcher to try to score and hand the ball off to a potentially worse pitcher?
In all of the following charts, sheet one will be reserved for starters, sheet two will be reserved for pinch hitters, and sheet three will be your cheat sheet for determining which hitter type was worth more runs. According to my data, there were a total of 9718 plate appearances by pinch hitters in those two seasons opposed to 358137 plate appearances by players that started the game. This means that a little over 2.6% of plate appearances were by pinch hitters.
Here is the expected number of runs generated by hitters while hitting in a given out state:
It makes sense to see that the runs generated goes down as the number of outs goes up. If you have more outs, then you have fewer chances to score runs. What's interesting is that pinch hitters are worth more than non-pinch hitters in situations where they come to the plate with one or two outs recorded, but starters are worth more when no outs have been recorded. Let's see if we can figure out why.
To do this, we will have to break our data into more, smaller pieces. With smaller pieces it will be easier for us to identify exactly when/why our data shifts. Lets take a look at how Here is the expected number of runs generated by hitters separated by base/out state:
According to this, there are four instances (out of a possible 24) where pinch hitters are worth more than non-pinch hitters when they come to the plate with no outs, but only two instances when they come to the plate with one or two outs. This seems backward since we determined that pinch hitters were worth more when they came to the plate with one or two outs in the first chart. These tables seem to tell us the opposite. Lets take a look at the data from a different perspective. What do the number look like when we take into account how many runs a team is down or if the score is tied when they send a pinch runner to the plate?
Same disclaimer as above, these were simply the most common situations and I didn't want the results skewed by a small sample size. Here, we see similar results to when a team is down when they pinch hit. The only situation in which a starter was worth more was when he came up to bat with no outs recorded, which also helps to confirm our first chart. It also seems as if even if a team is up it is advantageous for them to send a pinch hitter to the plate if they are in a position to do so.
So what about that second chart that seems to contradict everything else? Lets take another look at the data. This time, sheet three will show the number of times a pinch hitter came up to bat in those situations:
According to this, there are four instances (out of a possible 24) where pinch hitters are worth more than non-pinch hitters when they come to the plate with no outs, but only two instances when they come to the plate with one or two outs. This seems backward since we determined that pinch hitters were worth more when they came to the plate with one or two outs in the first chart. These tables seem to tell us the opposite. Lets take a look at the data from a different perspective. What do the number look like when we take into account how many runs a team is down or if the score is tied when they send a pinch runner to the plate?
It should be noted that this is far from a complete chart. The score of a game of baseball often differs by more than four runs, but these situations were by far the most common, and I didn't want the results skewed by a small sample size. In every situation but three, pinch hitters were worth more than starters. Two of those three situations are when the batter comes to the plate with no outs, which helps confirm our first chart. It looks like its a good idea for a team to pinch hit if they are in a position to do so. What about when a players team is up when a that player pinch hits?
Same disclaimer as above, these were simply the most common situations and I didn't want the results skewed by a small sample size. Here, we see similar results to when a team is down when they pinch hit. The only situation in which a starter was worth more was when he came up to bat with no outs recorded, which also helps to confirm our first chart. It also seems as if even if a team is up it is advantageous for them to send a pinch hitter to the plate if they are in a position to do so.
So what about that second chart that seems to contradict everything else? Lets take another look at the data. This time, sheet three will show the number of times a pinch hitter came up to bat in those situations:
Now this is interesting. Most notably when a pinch hitter came up to the plate with runners on 1st and 2nd and one or two recorded outs. As you can see, there is a sizable increase in pinch hitters in those situations. Those also happen to be the situations in which pinch hitters were more valuable than starters. This may simply be a coincidence, but nonetheless explains why that chart seems to contradict all of the others.
According to all of this data, if a team has an opportunity to pinch hit, they probably should. It seems as if the fresh bat is worth more than the tired bat, especially as pressure increases due to outs being recorded or if the game is close. That said, I'm not convinced that this data is at all definitive. It might be worth looking into how much the handedness of a pinch hitter compares to the handedness of the batter he is replacing as well as how much pinch hitting for a pitcher vs pinch hitting for a position player changes the data. I think this is a good stopping point for today though and will give us something to chew on for awhile.
According to all of this data, if a team has an opportunity to pinch hit, they probably should. It seems as if the fresh bat is worth more than the tired bat, especially as pressure increases due to outs being recorded or if the game is close. That said, I'm not convinced that this data is at all definitive. It might be worth looking into how much the handedness of a pinch hitter compares to the handedness of the batter he is replacing as well as how much pinch hitting for a pitcher vs pinch hitting for a position player changes the data. I think this is a good stopping point for today though and will give us something to chew on for awhile.
No comments:
Post a Comment