Why Do Sweepers Cause So Many Popups?

We’re still in a relatively new era of pitch design. Just this past year, we saw several successful pitchers embrace a sweeping breaking ball. Last month, Justin Choi wrote about one of those pitchers, Julio Urías, and noted that his new sweeper generates a great number of popups and lift. It’s not just Urías; in general, sweepers have a tendency to generate balls hit in the air at a rather skewed clip:

It’s strange behavior not just for a non-fastball, but for a breaking pitch especially; as you can see, sweepers have nearly double the popup rate of their fellow breaking balls:

So why do sweepers generate so many popups? Before we go any further, let’s define the sweeper. It has its own sub-classification based on movement and velocity, more concisely detailed here by Driveline’s Dan Aucoin. It’s a pitch classification that stretches across curves, sliders, and knuckle/spike curves; in practice, it looks something like a slurve but with some characteristics of the knuckle/spike curve. Here’s how it fits within conventional pitch classifications (the sweeper is highlighted in pink):

So how do sweepers stack up to non-sweepers in launch angle?

The distributions are similar, but the sweeper results in a considerably larger number of popups, which is backed up by the gap in popup rate.

This is all rather broad, though; the sweeper classification encompasses a great number of pitches. Ignoring these classifications, what should we understand about the interplay between movement/velocity and popups?

There have been plenty of words spilled on this site about horizontal and vertical approach angle. We care about movement, but the direct angle of the pitch from the pitcher’s release point has an impact here, too:

There’s a clear relationship here between VAA and average launch angle, with pitches with shallower vertical break in general getting put into the air more often. Still, let’s fold in velocity to both VAA and HAA to see if that makes things any clearer:

With VAA, a pitch thrown harder and at a more shallow approach angle results in higher launch angle, but many of those lower-end VAAs are largely location driven, and velocity is still a factor. There is even less to be pulled from the relationship between HAA and velocity, though that owes more to wider approach angles being better at garnering popups.

There’s nothing readily apparent about increased velocity that would help explain the sweeper’s popup characteristics. VAA and HAA are certainly a part of the puzzle, but it also helps to see what’s going on location-wise:

Regardless of location, sweepers just about anywhere in the zone tend to generate higher launch angles compared to our non-sweeping breaking balls. Now, it isn’t necessarily a good thing to have batters be able to lift pitches low in the zone that could otherwise be groundballs, but this launch angle gap exists everywhere.

It’s worth noting that these are league-wide numbers that aren’t specific to batter types. Pitcher-batter matchups matter greatly, so it could be helpful to know how effective the sweeper is against batters with different swing types, particularly those who might have more lift in their swing. There isn’t much in the public space that we can use for swing data, but we can take the average launch angle on a batter’s hardest-hit balls as a stand-in for their attack angle. Here’s a look at sweeper and non-sweeper launch angle by batter attack angle:

Even broken down by different batter attack angles, the sweeper consistently generates more lift.

I’ve been trying to find different reasons as to why batters get under and lift the sweeper, but a more definitive answer likely doesn’t exist in the data available to us. The individual components highlighted here have some degree of connection: shallow VAAs, large HAAs, and a sprinkling of increased velocity all point toward increased launch angles. But beyond that, it’s unclear what the cause is. It could well be that batters read the pitch out of the hand as a breaking ball and fail to correct for the lack of vertical drop. It could be that such wide horizontal movement is harder to track, or that increased velocity amplifies both of these issues.

This exercise did little to prove why batters get under sweepers, but it did confirm that the individual characteristics of sweepers are the same characteristics associated with generating high fly balls and popups. Regardless, it’s unlikely that this pitch type will fall out of favor anytime soon.