VAA Post #6: Avoiding High Exit Velocities with Blast Data

My Vertical Approach Angle (VAA) posts have primarily covered pitching and have taken a pretty general approach when it comes to saying something like "elevate the fastball more because pitchers get better results at flatter VAA's." In this post, I will cover some more specific, individualized strategies that could be used by considering both pitchers and hitters on a player to player basis.

What if a hitter was developing his approach around a heat map like the one below?

Let's say 2016 Chase Utley (LHH) saw this heat map while game-planning against a certain pitcher and decided to eliminate the inner third of the plate until he gets two strikes. Everything in the outer two-thirds is fair game until that. And let's say the pitcher he is facing performs best on the inner third, like the hypothetical pitcher below.

Recognizing this, Utley tells himself that even though the pitcher will probably attack him in, he will stay disciplined for something on the outer two thirds until he gets two strikes.

For the pitcher, the strategy here would seem obvious...pitch to the inner third. But what if solely strategizing around heat map performance strengths and weaknesses caused both players to overlook where their strengths actually lie against one another? Before I answer this, I need to explain why I chose to evaluate this using exit velocity.

Why I Used Exit Velocity

Generally speaking, there are other statistics that are better to use than solely exit velocity. Exit velocity does not take other important variables into play, such as launch angle and direction. However, with a hitter's general goal of hitting a ball as hard as possible, pitcher's should be looking to prevent hard contact. Easy enough.

In this article about optimizing the swing, Alan Nathan partially writes about the differences in trying to hit a ball for distance versus solely worrying about producing contact. He concludes that higher attack angles (18 degrees on a fastball with a 6 degree VAA) produce further batted balls than lower attack angles. When many people talk about hitting a ball as hard as possible, they refer to it as being "squared up." What does that mean in terms of things we can measure? A perfectly "squared up" pitch has an offset of zero inches. This would mean that the hitter's attack angle would perfectly match the VAA, also known below as "descent angle" below (ex. Hitter attack angle is 6 degrees on a fastball with a VAA of -6).

Offset leads me to the next reason why I used exit velocity instead of another metric like xwOBA. Pitchers can make it a goal to avoid throwing to high exit velocities based on the difference in their VAA and the hitter's attack angle... sort of. While Blast Motion data is not publicly available, I believe there are ways to use the Blast data a team has collected in order to predict those of players around the league.

Process for Attack Angle Heat Maps

This is one of the hurdles to get around in this strategy. Creating heat maps for opposing teams is going to take some guesswork. I believe a few things can be done to help determine opposing player attack angles throughout the zone, most notably creating a player comp tool.

Nearly every organization is using Blast sensors at some level of the minor leagues. This part is important...teams (advance scouts, coaches) will need to bucket their own players into three groups that describe their swing: "uphill", "flat", or "steep" PRIOR to looking at the player's individual attack angle heat map. The reason for this is because teams will determine an opposing hitter's swing type partially off the ole eye test. In order to determine opposing swing types based off of video to be a repeatable process, those making the determination cannot have any bias by seeing the data of their own players first.

A more objective way to determine their attack angles, or swing type in general, is by estimating their average attack angle as a whole. While I am not sure and it is difficult to confirm for certain, it's possible that players who have close overall average attack angles would have similar attack angles throughout the entire zone. And by similar, I mean from one player to the next, not an attack angle of 10 in all parts of the zone. Hitters with similar batted ball data could potentially be placed in each bucketed swing type, as well ("uphill" if player hits a lot of HR's, "steep" if player has lot of GB's/low LA).

Once the determination is made of the hitters that fall into each type of swing, the team can then re-create attack angle heat maps for each of the three groups based on those players' data in that team's own system. Prior to a series, opposing hitters would be identified as one of the swing types and pitched to in part based on the difference between their assumed attack angle and each pitcher's VAA. I want to make that clear, as well, I am not saying this idea should be used as absolute used for every pitch.

Finding an Optimal Zone

So in order for pitchers to induce the lowest possible exit velocities (keep in mind this could include whiffs) based around the idea of maximizing "offset," pitchers need to throw to the zones that have the largest differences (absolute value) in VAA and attack angle. An example of how this could be done is below.

The heatmap on the right is an example for a hitter where we paired Blast Motion with Rapsodo. Before I go on, I think it's important to note that this is a relatively small sample size and I am only using this as an example. It's also important to know that Rapsodo does not provide point of contact, which is one of the biggest determinations of attack angle. Lastly, the pitcher VAA's are all 4-seam fastballs.

In a perfect world where both of these are available and sample size is sufficient, a pitcher could evaluate these side by side and pound the zones with the largest difference/offset within the strike zone. In this case, the low and in pitch to a left-handed hitter has the biggest difference (11.3-8.7=2.6) and would be expected to produce the lowest exit velocity. In some ways, I like to think of this approach as an expected barrel stat.

Considerations

Without having access to every piece of data required, it is difficult to know how this would actually perform. Since the best batted-ball outcomes often are not "squared up" perfectly (if this doesn't make sense, read the Alan Nathan article and check out the graphs toward the bottom), it could actually be detrimental for a pitcher to throw to the zone with the largest difference in VAA vs. attack angle, as it would increase the offset in ways I am writing about preventing. In many cases, xwOBA numbers are higher at lower exit velocities. I think it would be pretty fascinating to pair Blast numbers with all of Trackman's pitch data, as well. And if Blast can somehow provide data on swings and misses, that would help paint an even bigger picture.

Another possible way to best utilize this is to only use it on low-power potential hitters who rely more on barreling balls consistently at lower offset. Sinker/slider pitchers may be better equipped to use this strategy, as well, as they typically rely on inducing poor contact.

Lastly, we need to remember VAA is also measured as the angle of a pitch as it crosses the plate which makes the heat map imperfect since that is not always where contact is made. Something 3D might be pretty cool with this.

Back to Chase Utley

I'm not sure what Chase Utley's average attack angle is. I'd honestly be surprised if he has ever even thrown a Blast Motion sensor on his bat for that matter. But if you can't see what I am getting at, basically I think there are ways beyond overlaying a pitcher's performance and usage with a hitter's performance in a specific zone to optimize a strategy. It could very well be that the lowest expected exit velocity for Utley against a particular pitcher is in his previously noted best-performing exit velocity zone. Just because a hitter struggles in a certain zone does not necessarily mean that is the best zone to pitch to. And just because we don't have a hitter's attack angle to each zone, does not mean we can't find comparable players to use a similar strategy on.

Or I could be completely wrong.