I decided that it might be interesting to look at the correlation between selectivity and OPS across the different pitch counts within subject. I’m not sure exactly what this correlation means. I don’t think it’s outlandish to say that your level of selectivity would influence how well you hit. When you are in a better count you can swing at a higher quality of pitch which should allow you to be more productive – but I’m not totally sure if it’s causal or not. It could just be picking up something in the situation which has an effect on both your selectivity and your productivity – like the quality of pitch you are thrown. If it is causal then it might give us some idea of how a player changing their batting approach might influence their productivity. Since it’s within subject it also allows us to hold many of the other factors constant.
The chart below shows the Batting Eye, Selectivity and OPS of all batters in the league broken down by pitch count.
It is probably easiest to see the relationship between Selectivity and OPS if we plot them against each other
We see that there is a very strong positive relationship between Selectivity and OPS. As a player gets ahead in the count they can wait for better pitches to hit and it definitely shows up in their OPS. If they fall behind they have to press a bit and go after pitches which are not as good. This is, basically, why people say that the first strike is the most important (though if you look at the chart… it’s really with TWO strikes that a batter becomes much less selective – and therefore less productive.)
So there’s a strong correlation collapsed over all batters but what about within a single batter? To take a look at this I (randomly) decided to look at two very different players. The first is Albert Pujols, the best hitter of our generation, and the second is Jeff Francoeur, one of the least selective batters in the league who is trying to improve his approach to batting this year.
First let’s look at the charts for each player. Pujols first and then Francoeur.
The first thing to notice here is that Albert Pujols is REALLY REALLY good. Yet even he seems to be mortal once the pitcher has two strikes on him. Maybe strike two is actually the most important pitch. The second thing to notice is that Jeff Francoeur will swing at almost anything except on the first pitch and a 3-0 count.
Let’s check out the correlation of Selectivity vs OPS for each player.
We see here that even though Pujols and Francoeur are very different hitters they both exhibit a significant correlation between their selectivity on a pitch count and their OPS on that pitch count. As they become more selective and require a higher quality of pitch before they swing they are able to produce more on those pitches. Albert Pujols clearly understands this and it’s a large part of why he is the best hitter in baseball. If he swung at bad pitches and allowed the pitcher to get two strikes more he would be a much less effective hitter. Jeff Francoeur…well… he says he wants to walk 60 times this year, and that he’ll be a completely different player if he does. Maybe he’s starting to get it – but he hasn’t proven it yet. And while the walks would be nice I hope he realizes that being selective for the sake of getting better pitches to hit – and not just being more passive to take walks – is the key to his evolution as a batter.
I’ll have more coming up on Francoeur later this week so stay tuned.
Hey, this is good work.
I’m just reading through some of it now and wanted to make some comments.
You mention that the relation ship of selectivity to OPS may not be causal. I think that’s important. Does OPS by count just total the result of all plate appearances that ended with the last pitch being thrown on the specific count? If that’s the case then the correlation between these measures is in a large part the result of the count. That is, count has a causal relationship with OPS and count also has a causal relationship with selectivity.
In all counts with 2 strikes or 3 balls Strikeouts and Walks have a huge influence on the OPS outcome of the pitch. Also, batters will obviously be more or less selective based on how close they are to striking out/walking. To confound the relationship even more, some of the correlation could be coming from the pitchers giving batters better pitches to hit (type/location) depending on how much they feel they need a strike, given the count.
I guess in the end, this is just reinforcing our perceptions that batters take more (more willing to pass on borderline pitches) in hitters counts, swing more in pitchers counts, and that PAs that end in hitters counts are better for the batter than those that end in pitchers counts.
I think the math behind Batting Eye and Selectivity might still be going over my head so I’ll try to understand that more later.