Jose Fernandez and where batters are making contact

Jose Fernandez pitched another fantastic gem of a start, going seven innings and striking out 14 New York Mets batters with no walks en route to a dominant 1-0 victory of the Miami Marlins. It was more par for the course of the recent version of Fernandez; in the month of May, he upped his strikeout rate to nearly 38 percent while dropping the worrisome walk rate of April from nearly 12 percent down to 7.4 percent. Early in the season, we were worried that Fernandez was trying to be effectively wild and that his current season was a strange one. However, so far, it has been an effective campaign, and it seems Fernandez has been reveling in this new style of pitching.

That drop in walks, however, from April to now is rooted in real differences in play, particularly it would seem in command. Take a look at the heatmaps for Jose Fernandez's pitches before and after the end April.

April

May-June

You can clearly see that Fernandez has tightened up his location and focused on two areas of interest: pitches inside to primarily right-handed hitters, likely the fastball, and pitchers down and away to righties, likely the Defector curve/slider. In his few starts in April, things were a little more all over the place, with his primary focus being closer to the center of the plate in a dangerous area for pitchers in which hitters can take serious advantage.

Interestingly, this has come up this season because hitters are making contact in what seems like dangerous locations for Fernandez. Last year, Jose Fernandez allowed a .343 BABIP to hitters, and this year his BABIP is also elevated at .331. One might see that and think that with a pitcher this skilled, it seems unlikely that this should continue; players with such strong skill at missing bats usually allow slightly weaker contact than usual. However, Fernandez's exit velocities are actually elevated this season, as he has allowed a 92.6 mph average velocity off the bat.

FanGraphs' August Fagerstrom has a theory on this, and it has something to do with just how elite Fernandez has gotten at missing bats so far.

Most pitchers generate their weak contact outside the zone. Against Fernandez, however, nobody is able to make contact out of the zone, so the only time a hitter puts the ball in play against Fernandez is when it comes against a hittable pitch.

You can see the difference in the contact maps for guys like Fernandez versus other elite starters.

All of that contact is dead center in the strike zone this season. Compare that to last season's version of Fernandez.

All of that contact is still centered, but there appears to be an overall larger spread in all directions, most notably in the lower part below the strike zone and in the upper outer corner from a right-handed hitter's perspective. Now look at this compared to Jake Arrieta, against whom hitters are making contact on out-of-zone pitches at about a 56 percent rate.

Again, the concentration of most contact is in the middle of the strike zone, but the range fans out inside, outside, and low in a far wider area than it did for Fernandez. Arrieta owns an identical 31.3 percent swing rate on out-of-zone pitches as Fernandez, so the difference here is all in contact. That out-of-zone contact is weaker, and thus it helps Arrieta towards his .247 BABIP this season.

Now, the trade-off for this is an intriguing one. The average ball in play on pitches out of the strike zone I imagine is lower than the average batted ball off a down-the-middle fastball. Still, even if it is largely a negative event, it probably is not worth less than a swinging strike. A swinging or called strike is worth some value less than the -0.27 runs an out is worth. If you just estimated the average strike to be worth a third of an out, that would make it worth -0.09 runs. How many outs would you have to get on balls in play on a swing in order to make up for the value of a swing-and-miss? Assume an average distribution of types of hits on contact out of the zone (another probably non-truth, as you would think line drives would be less prevalent) and the average hit value comes out to 0.62 runs. To get to about -0.09 runs, you would have to get outs on about 80 percent of balls in play given these numbers.

Of course, that makes a few assumptions that are probably not true, and the real value is probably something closer to maybe 75 percent. Batters on average get hits on 33 percent of balls made contact, so you have to get about eight percent more outs to break even with just getting whiffs on the same pitch. In other words, the fact that Fernandez is missing bats on those pitches is not likely hurting the average value of his pitches as compared to forcing weaker contact. Given that this seems to be working for now and that Fernandez's swings on out-of-zone pitches are slowly coming up, I think Marlins fans should be very happy with the developments as the season has wore on. Don't mind the harder contact and BABIP, that is just a byproduct of him being an elite pitcher everywhere else.