Houston Astros Minor League Spotlight: The 3-2 Count

You've heard the story. And if you haven't heard the story, then you're about to.

It's a horror story. A horror story in which Giant Swarms of Nerds™ have taken over a Venerable Baseball Franchise™ and used spreadsheets and pivot tables to redefine how baseball is played. In this chapter of the story, the Nerds - this time headed by hitting coach John Mallee - have convinced their automaton Houston Astros minor leaguers to take on 3-2 pitches, because they are likely to be balls.

In not overreacting to this story, Danny Knobler of CBS Sports overreacted:

They have suggested to their hitters that they should be more selective with a full count, and according to sources they have also suggested to their minor-league managers that if the hitters don't do it on their own, it could be made mandatory at some point.

So the Nerds are training an army of FarmBots to take pitches on full counts! Anarchy!

Harold Reynolds even took some time out of his busy schedule of doing whatever it is that Harold Reynolds does to ask General Manager Jeff Luhnow about it:

(This is an exceptional time to mention that Reynolds himself had a .502 OBP on full counts, and struck out just 53 times in 426 at-bats. If anyone should know about hitting with a full count, it should be Reynolds. Of course, one has to question whether or not he even knows that he was that successful on 3-2 counts.)

As you can tell, the national media - and baseball experts everywhere - are very upset that the Houston Astros are dictating what batters do in the batter's box.

Except... are they?

Probably the most telling part of the Knobler quote above is this line: "...they should be more selective with a full count." Of course, this ignores the general idea that hitters should generally err on the side of being selective at the plate on any count, but it does suggest some rationality. Instead of just telling minor leaguers to take on a full count, the Astros seem to be telling them to only attack pitches they can drive, and to stay away from the stuff on the edges.

This doesn't exactly fly in the face of traditional hitting philosophies, which often suggest an aggressive approach with any two-strike count, trying to foul off pitches on the edges and wait for a pitch to drive. Rather, it's the idea of the front office unilaterally dictating batters' activity in the box that seems to have people up in the arms, no matter what they're telling them to do.

I decided to sample twenty-four minor league games, including in-person data I took in Lancaster, as well as Gameday data on the RedHawks and Hooks. All told, between those three teams in those twenty-four games, we have 208 full-count pitches to look at, spanning from April to July.

Balls vs. Strikes

It stands to reason that, if someone is telling minor leaguers to take pitches on 3-2 counts, then that someone has ascertained that most pitches thrown on that count are balls. However, of the 208 pitches in our sample, 153 of them (73.56%) were strikes. That isn't to say that they were thrown in the zone, however (more on that later), as that includes pitches that were swung at, whether or not they were put into play.

So, then, what about those balls put into play?

Balls in Play

Taking a pitch on a full count can only result in one of two outcomes: A walk (if it's a ball) or a strikeout (if it's a strike.) But swinging at a pitch on a full count can create three outcomes: A strikeout, a foul ball, or a ball in play.

Of the 153 strikes in our sample size, twenty-five of them were strikeouts (fourteen swinging). Fifty-seven were fouled off. That leaves seventy-one balls in play:

GB: 35 (49.3%)
IFFB: 4 (5.63%)
FB: 19 (26.76%)
LD: 13 (18.31%)

We're already seeing an indication that batters might not get very many pitches to drive on 3-2 counts, as a line drive rate of 18.31% is not exceptional. Still, we're within fairly normal ranges here.

However, what is hinted at in the "take borderline pitches on 3-2 counts" philosophy is the idea that batters should only be swinging at pitches that they can drive. A groundball rate over 49% indicates that this probably isn't happening. So what is happening? I'm glad you asked. On 33 of the 71 balls in play, the batter hit safely (26 singles, 7 doubles).

Now, I'm sure I don't have to tell you that a .465 BABIP on full counts is kind of a big deal.

Laying Off

Well... so what about when they don't swing? When the minor leaguers have taken on full counts, they've drawn walks 83.08% of the time. Overall, they have a 36.00% BB% and a 16.67% K% through our sample.

To get an idea of the locations of the pitches they were seeing, I divided the strike zone into a 5x5 grid, since I'm not cool enough to be able to make heat maps. Still, this will give you a good idea:

(Bear in mind that I used the same grid for righties and lefties, so the right side of the grid is always the outside of the plate and the left side is always the inside of the plate.)

Conclusions

We can absolutely, 100% rule out the idea that the Astros are telling their minor leaguers to take on 3-2 counts, as they have actually ended up swinging at pitches over 68% of the time over our sample, which includes twenty-four games over four months and three levels.

We can also conclude that these minor leaguers are seeing some success on full counts. The phrase "look for pitches to drive" may be a bit of a misnomer, as only 7 of the 208 pitches resulted in extra-base hits. They're generating a lot of groundballs, but they're groundballs that are getting through the infield for hits.

Astros minor leaguers seem to be extending plate appearances on full counts by fouling off balls, drawing walks, and getting hits. Whether this is the result of a fluke in our sampling method, a minors-wide policy of being more selective on 3-2 counts, or pure dumb luck (.465 BABIP!), we can't be sure. But what we can be reasonably sure of is that their approach on the count is working.