In Search of the High Fastballs

Theory: Players have gotten better than ever at hitting pitches down in the zone. I don’t think this counts as a controversial theory anymore, and it goes hand in hand with what’s been casually termed the fly-ball revolution. Anecdotally, it seems like there are more and more hitters trying to hit the ball in the air. Generally speaking, this is achieved by swinging with more of an uppercut, and, generally speaking, players with uppercuts are more productive down in the zone, instead of up.

Theory: This is one of the reasons why there’s been an emphasis on higher-spin fastballs. Those are the tougher-to-hit fastballs, fastballs you mostly want to be elevated. This past offseason, I put forth the idea that the cure to the home-run spike could involve more fastballs up. If pitchers just focused somewhere else, then hitters wouldn’t so often be able to elevate the pitches at their knees. To summarize, simply: It seems like there would be a pitcher response to the hitter response. It seems like there should be more high fastballs.

But, are there more high fastballs? Turns out this is really easy to check. And the answer is, well, basically, no.

Time for a whole bunch of images, informed by Baseball Savant. For simplicity, I divided all pitches into two groups, setting the midpoint at 2.5 feet off the ground. A fastball above that would count as a high fastball. Here’s the trend in league-wide high-fastball rate going back 10 seasons:

Baseball has achieved its lowest point in the observed decade. At the peak, roughly 31% of all pitches were high fastballs. So far this season, the rate is below 27%. That’s subtle, just a few pitches per game, but the line’s direction is unmistakable. Now, there’s a problem, however. The above data uses all pitches as the denominator. There’s a greater trend actively taking place — league-wide fastball rate is slowly sinking. So why don’t we look at the same kind of plot, but this time dividing by just fastballs?

There we go; that’s something. Through 2015, the average fastball was dropping. Pitchers were working lower and lower, presumably at least in part because the zone was sinking lower and lower. Last season, you see a bounceback in the high-fastball rate, with a jump of a full percentage point. This season, the rate has been essentially the same as last season. So when it comes to the question of whether there are more high fastballs now, there are two parts of the answer. Relative to the overall decade, high-fastball rate remains down. Yet there has been a very minor recent uptick. It’s possibly something to pay attention to, but if this is any indication, teams aren’t exactly instructing their pitchers to panic and keep the ball above the belt. Pitchers look more or less normal.

Now, how about what individual teams have done? Here’s what the 2016 situation looked like:

Compare that to the 2017 landscape. Although, don’t worry too much about trying to do that with your eyes, because I’ve taken care to run all the calculations. That plot comes after this one.

Most teams haven’t budged very much. In both plots, for example, you see the Red Sox and the Nationals toward the left. Out of baseball’s 30 teams, 18 have had high-fastball-rate shifts of less than four percentage points. It’s the exceptions, though, that capture our interest. Here’s how every team has moved, in this particular respect:

The Twins have increased their high-fastball rate from just under 46% to just over 53%. They have a gain of 7.8 ticks, easily the largest out of everyone. And yet, the magnitude of their change is kind of dwarfed, if you look to the right. The Orioles, for example, are down 8.8 ticks. And the Astros are down a relatively massive 12.1 ticks. By high-fastball rate in 2016, the Astros ranked 10th. Now they rank 30th. That sort of change can’t be an accident, as the Astros are leaning into the low-fastball approach.

Speaking of things that probably aren’t accidents, the Twins are using the Astros’ catcher from a year ago. And, on a slightly smaller scale, the Nationals are using one of the Orioles’ catchers from a year ago. But I don’t think the meat of the story here is the backstops. The Astros have had 11 pitchers throw at least 100 fastballs on the year. Nine of those 11 pitchers have reduced their high-fastball rates, and three pitchers have reduced their rates by at least 10 percentage points: Dallas Keuchel, Tony Sipp, and Lance McCullers. Charlie Morton has dropped his own rate by nine percentage points, and Morton was already a sinker-baller. The Astros are making sure their sinkers sink. Keuchel has baseball’s lowest high-fastball rate among starters. Morton, too, is close to the bottom, and McCullers is now keeping the ball about as low as Kyle Hendricks.

I don’t want to act like the Astros are the only team up to something, but theirs is a particularly interesting data point. You’d think that teams would be demonstrating more of a willingness to explore the upper part of the strike zone. After all, the bulk of the recent home-run increase has come lower within the box. The Astros, I’m certain, are aware of where the home runs have been, but they’re staying low, remarkably low, without a care in the world. Maybe it’s as simple as Keuchel improving his command again, and Morton just being a guy who was available for the right price. Maybe the Astros would throw more high fastballs in an ideal world. But you could take them as a sign — there’s no true urgency here. Maybe hitters are better than ever against pitches down, but if you throw the right pitches down, you can win 71% of your baseball games.

Quickly, in closing: Few pitchers have reduced their high-fastball rates by as much as Grant Dayton. For Dayton, that’s a problem. And few pitchers have increased their high-fastball rates by as much as Blake Treinen. For Treinen, that’s a problem. Individual data points go to show that it’s not always about executing a plan; sometimes these data points waver because a pitcher himself is wavering. But on the team level, you expect to find more signal. You expect to find even more still when you examine the league overall. Overall, the high fastballs haven’t surged, not yet. Perhaps they still will. Perhaps the hitters need to do a better job of forcing it.

Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.

Newest Most Voted
Inline Feedbacks
View all comments
5 years ago

Thanks for writing this. I’ve definitely noticed more work upstairs in Minnesota, particularly when they were playing Texas. I’m wondering if there’s an uptick in usage with 2 strikes? I’m intuitively thinking high heat is used more as an out pitch and more randomly located earlier in counts, but I don’t know if the data backs me up.