The impact of the evolving strike zone

The strike zone is a superficially simple concept kept constant in MLB’s rulebook, except for minor changes, for more than a hundred years. But in the past six seasons, the strike zone itself has evolved noticeably, and this change has impacted all aspects of the game.

MLB’s strike zone is defined as the space between a space over home plate, which is as high as the midpoint between the top of the shoulders and the top of the pants and as low as the bottom of the knees. Yet while the rulebook boundaries of the strike zone have stayed exactly the same since 1996, the advent of Pitchf/x has allowed us to measure the de facto limits of the zone, as actually called by umpires.

Since Pitchf/x data has become available, the strike zone has become more narrow but also taller. In particular, the area toward the bottom of the zone has expanded. The addition of this new territory has not gone unnoticed by pitchers. Instead, they have begun to aggressively exploit it with lower deliveries and more pitches with sinking action

To show you what I mean, consider this graph of the average height of all pitches thrown in MLB over the last six years, with left- and right-handed batters on separate lines.

There’s a fairly dramatic dropoff, from an average height of ~2.4 feet in 2009 to only 2.25 feet now. Over the course of a single year or two in this process, the drops in pitch height might go unnoticed, as they were always less than an inch. Examined with the benefit of more time, we can see that batters are being challenged about two full inches lower in the zone than six years ago, which is considerable. To put that in perspective, it’s something like the difference in height between where Jose Altuve sees his average pitch and Matt Holliday his (Altuve stands a miniscule 5’5, Holliday a hulking 6’4).

Whereas the zone has expanded at the bottom, it hasn’t changed greatly on the sides. Here’s the average distance from the center of the plate at which left- and right-handed hitters see pitches.

This graph charts the equivalent change, but for the vertical location of all pitches (negative numbers are closer to third base, positive closer to first).  While lefties are seeing their pitches a little closer to the center of the zone, this slight narrowing of the horizontal range at which pitchers attack the zone in no way makes up for the much deeper areas of the zone that they can now pursue. Indeed, an article by Jon Roegele has found that the strike zone’s total area has expanded some 30 inches in the last five years. A bigger strike zone gives the pitcher more territory to use, and the hitter more space to worry about; on balance, the larger zone favors the pitcher substantially.

Some of the rising strikeout tide can be pinned upon this change in the strike zone. Brian Mills estimates that up to 9% of the recent increase in K/9 stems from the newer, bigger strike zone. Given the chance for a wider area in which to throw strikes, pitchers have taken advantage and put hitters far off balance. We are witnessing the result: historic strikeout rates across the board, rendering some of the old cherished baseball records and milestones more common.

But there is another, more hidden impact from the evolving strike zone.  Since pitchers can now safely get strike calls lower in the zone, hitters can’t afford to be as picky as they used to be. At constant risk of strikeout, batters are forced to make contact with pitches low down in the zone.  When they do, the bat is more likely to swing over the ball, and drive it into the ground. As a result, groundball rates are increasing.

From just under 45% in 2009, groundballs have increased somewhat to about 47% now. FanGraphs leaderboards, using a different data source, confirm the same hypothesis. It’s a modest increase, but multiplied over many thousands of balls in play, even small differences are noteworthy.

These grounders are not likely to end up as hits, either. In the age of the shift, groundballs are converted into outs more often than ever. Defense has become more intelligent: front offices are adept at predicting the batted ball behavior of each hitter, and positioning their defenders optimally to turn those wormburners into outs.

By the same logic which suggests that groundballs should be increasing, flyballs should be decreasing. And in fact flyballs are declining at a precipitous rate. From 2009 to 2014, flyball rate fell from 28.6% of batted balls down to 21.9%.

Because almost all home runs start their lives as flyballs, the fact that we are witnessing a 30% decline in flyball rate explains a lot of the decrease in home run rate that is occurring. If the only pitches David Ortiz gets are around his knees, he’ll have a hard time elevating the ball. If he can’t get the ball up in the air, there’s little chance of it clearing the fence.

This change in batted ball type also affects the kind of players who are valuable these days. A hitter whose profile is value is dependent on getting home runs may be in some trouble unless they can also drive those home runs from low in the zone. Failing that, they’d better be able to run, because the odds are good that they’ll be legging out a lot of groundballs.

Beyond the change in batted ball distribution lurks another ripple from the strike zone’s expansion. It stands to reason that if hitters are putting more balls on the ground and fewer in the air, the players who field those balls will change. Outfielders tend to see mostly line drives and fly balls, but it’s infielders who are responsible for the lion’s share of groundballs.

In accord with that idea, the number of plays accorded to infielders vs. outfielders has changed as the years have progressed. In 2009, 43% of plays were made by outfielders. By last year, that number had dropped to 40.5%. Of course, the percentages assigned to infielders increased by a corresponding amount. To put it another way, the average outfielder used to get about three-quarters as many chances to field balls as the average infielder, but in the past few years, it’s more like two-thirds as many chances. The change in the strike zone flowed all the way out to the outfield, and leaves outfielders contributing their defense on a smaller number of plays than they’re used to. Front offices may have taken note, because they are seemingly more willing to put defensively-challenged hitters at the corner outfield spots.

The impetus for the evolution of the strike zone may have come, ironically enough, from the rulebook. The Pitchf/x technology, which allows us to gather data on the locations of all balls and strikes thrown in an MLB season, is also used by MLB to grade its umpires. In making rigorous their process for evaluating umpires, MLB may have also directed them to be more consistent with the lower part of the zone, an area which was informally understood to rarely get strike calls. The umpires have slowly but surely complied, and the de facto definition of the strike zone has correspondingly evolved to better resemble the rulebook’s definition, with a lower limit much closer to the knees.

This initially sound directive to bring the strike zone more in line with the rulebook has caused pitchers to change the way they approach hitters, and forced hitters to change where they make contact with the ball.  This modification, in turn, may have altered the league-wide distribution of batted balls, making flyballs less frequent and groundballs more frequent.  Finally, as the distribution of batted balls has changed, so too have the defenders who play those batted balls: outfielders have seen their share of the defensive play plummet by thousands of chances over the course of the last five years. 

A simple change at the plate sent ripples out as far as the outfield. While it’s difficult to know whether MLB intended for its change to be as far-reaching as it has been, it’s hard to imagine that it wanted to decrease offense so severely, potentially making this revision a prime example of the law of unintended consequences.