August 8, 2012

PITCHf/x Mailbag

Swing Tendencies on 3-0 Counts

by Dan Brooks and Harry Pavlidis

Printer-friendly

Dan Brooks and Harry Pavlidis, the minds behind Brooks Baseball and the PITCHf/x Hitter and Pitcher Profiles, will be answering your statistical questions using PITCHf/x data on a regular basis at BP. To submit a question for consideration in their next mailbag, email them at mailbag@brooksbaseball.net or cram your question into 140 characters and send it to @brooksbaseball or @harrypav.

For our first PITCHf/x mailbag, we’ve decided to take a look at a deceptively simple question. We’re not so good at simple, however, so we took lemons and made a small storage building out of them.

How often do major league players swing on a 3-0 count?
—Derek Turner

Well, Derek, the answer to your question is 6.95 percent of the time. But let’s go ahead and look more deeply at that question. For example, how often do hitters swing in all counts?

And what if we restricted that table to just pitches in the strike zone?

So, on 3-0 pitches that are strikes, batters swing 9.72 percent of the time—a fairly substantial increase from the 6.95 percent. Of course, these rates are nearly tenfold less than how often batters swing on 3-2 counts in the zone.

What if we further subset by pitch types? To do this, we used pitch type data from the Pitch Info database that powers the hitter and pitcher PITCHf/x profiles, as well as the Brooksbaseball.net pitcher cards. Pitchers throw a fastball variant (either sinker or four-seam) 93.8 percent of the time in a 3-0 count. As you can see in the table below, if that pitch is in the zone, batters swing at those pitches 9.65 percent of the time.

What about when that pitch is off-speed, like a changeup, splitter, curve, or slider, and is also in the strike zone?

It actually turns out that batters swing a little more at offspeed pitches 3-0 than fastballs 3-0, which is a little counterintuitive and makes this seemingly simple question even murkier. Why are the 3-0 off-speed pitches more likely to yield a 3-0 swing? Selection bias?

What if the 3-0 off-speed pitches were reserved for the batters who are known to have the green light? To answer this question, we took two groups of hitters: those who swung at more than 10 percent of 3-0 pitches (“green-lighters”) and those who did not (“red-lighters”). (Note: we realize that the propensity to swing or not swing does not necessarily indicate whether or not a player has or does not have the “green light.” We’re just using those as easy-to-remember group names.)

The group of green-lighters consists of 112 players like Ryan Howard (35.7 percent), Albert Pujols (28.8 percent), and David Ortiz (24.7 percent). The group of red-lighters consists of 308 players like Mark Teixeira (8.2 percent), Bobby Abreu (1.1 percent), and Jose Reyes (0 percent).

Because the distribution is strongly skewed toward players who don’t swing, there were many more below 10 percent than above 10 percent. On the whole whole, the green-lighters were swinging on 3-0 counts 16.5 percent of the time, and the red-lighters were swinging 3.1 percent of the time.

We next took those same groups of hitters and restricted the analysis to pitches in the strike zone. Green-lighters swing more, of course—up to 23.8 percent. Same story for the red-lighters, who slide up to 4.3 percent swings in the zone. This is roughly a 50 percent increase in swing rate when the pitch is in the zone, without regard to group.

Okay, we’re all convinced that we can group hitters into “swings a lot on 3-0” and “swing less than a lot on 3-0.” What types of pitches are these guys seeing, though? More off-speed for the green-lighters or not?

Overall, the green-lighters were thrown a four- or two-seam fastball 92.1 percent of the time in those counts, the “red-lighters” slightly more at 94.5 percent. This difference in pitch types seen may be the result of advanced scouting of the green-lighters’ swing tendencies. Thus, there is a slightly decreased chance of green-lighters seeing fastballs, which may account for the overall greater likelihood of a swing on those pitches in the 3-0 count.

So, yes, the increased swing likelihood on off-speed pitches relative to fastballs may be selection bias rather than an actual increase in swing likelihood. Guys who swing 3-0 will get off-speed pitches a bit more as a result.

So, Derek, er, what was the question again? Oh right, “How often do major league players swing on a 3-0 count?”

The raw data: 6.95 percent

Harry and Dan: It depends on the hitter, but pitchers seem to know who the green-lighters are and try to surprise them with off-speed pitches.

This is why we dive into raw data and dissect the simple. We learn or confirm things about baseball, and that’s the idea.

Okay, next question…

To submit a question to the BP PITCHf/x Mailbag, send your question, name, and where you’re from to mailbag@brooksbaseball.net.

Dan Brooks is an author of Baseball Prospectus. Follow @brooksbaseball
Click here to see Dan's other articles. You can contact Dan by clicking here
Harry Pavlidis is an author of Baseball Prospectus. Follow @harrypav
Click here to see Harry's other articles. You can contact Harry by clicking here