What Goes Into An MLB Draft Model: Vertical Approach Angle

One of the more common terms used in the analytics community is ‘model’. Whether it’s a model built to predict how much a specific metric can predict the future BABIP of a player, or looking at the best ways to predict run production; all models have one thing in common, using data to create a more accurate representation of players.

Draft models are more unique compared to other models though, as sometimes you have to work with extremely small samples of data (which can be the case for high schoolers), or zero data at all (which can be the case with prospects who suffered injuries). On top of the, at times, lackluster sample sizes, many organization’s models actually factor in scouting reports on certain aspects that can’t necessarily be quantified by data, such as command, approach, and character.

Every team’s model is unique. Some teams have kinetic sequencing and biomechanical evaluations as the most important variables. Other teams put more emphasis on pitch metrics, and batted ball profiles. This article will be focusing on the latter.

Pitch metrics are a broad subject. The basics being release speed, spin rate, extension as well as many other terms. The more modern metrics being movement profiles, spin axis, and most importantly, vertical approach angle, which this article will be centered around.

Before we get started, there are some things I’ll be noting. First of all, I’ll be referring to it as VAA. Second, plenty of research has been done on VAA of late, and most of the findings suggest there’s very little relevance in regards to offspeed and breaking balls. So I’ll be focusing on fastballs only. Lastly, while VAA is always a negative number, in this article, numbers will be written as a positive number so it’s easier to read.

What is VAA?

Trackman defines VAA as the following: How steeply up or down the ball enters the zone, reported as the angle in degrees, as the pitch crosses the front of home plate. A negative number means it is sloping downward, while a positive number (rare) means it is sloping upward.

How do we get VAA?

When talking about VAA, the most important thing to note is how important location is. It’s number one variable in TrackMan terms, is plate location height. 

The average VAA in college baseball for 2020 was 5.49°. When you expand it to just 2019, it’s 5.54°, and when you batch the two together, it’s 5.52°. With VAA though, averages are meaningless, unless it’s isolated to a specific quadrant of the strike zone. So here’s the average VAA by zone:

Knowing that on Trackman, the strike zone is between 1.5 to 3.5 feet above home plate. You can figure out that assuming every 0.665 feet above the strike zone, the average VAA will also decrease (technically it’ll increase as it’s measured as a negative number) about 0.68 degrees. This means every 0.1 foot above the strike zone, on average, a VAA will decrease around 0.082 degrees.

Plate location height is far from the only metric that plays into it. Induced vertical break (IVB) also has some correlation. We know VAA is the measurement of the ball as it enters the zone. So logically, the vertical carry of a baseball will make a VAA appear slightly flatter. Similar to that is release speed, the harder a pitch is thrown, the quicker it’s moving, so there’s less gravity-pull on it. Release height is the last major variable in VAA for obvious reasons. The lower a pitcher throws from, the easier it is to create an illusion of a “rising fastball” which pitches thrown with VAA’s under 4.0° tend to do.

Why is VAA Important?

Not many things directly correlate to success at the next level, but the ability to miss bats (aka pure stuff) tends to translate most. Which is why teams have begun targeting pitchers with flat VAA’s. Jeffrey Hakanson (Rays), Tyler Brown (Astros), and Holden Powell (Nationals) are three examples of relievers who got heavy interest due to many reasons, but VAA-related traits playing a heavy factor. Slade Cecconi (Pirates), Christian Chamberlain (Royals), and Kade Mechals (White Sox) are all great examples as well.

Likewise to spin rate, velo, and movement, VAA doesn’t change when you change playing levels. Opposing batted ball profiles, wOBA, wRC+ and slash lines do.

Unfortunately, there is no public data on VAA in the MLB. So we’ll be using numbers from the college side in every statistic we use to keep the data relevant to each other. So here’s whiff rate by pitch zone in college baseball last year:

Similar to the MLB, the higher a pitch, the higher the swing-and-miss rate. This isn’t surprising though, everybody knows this. But here’s where this correlates to VAA. In a sample of 10,000+ pitches thrown, we limited the data to pitches thrown up in the zone (between 2.83-3.5 plate location height) and found out the swing-and-miss rates sorted by VAA. Here’s that:

The difference between an average VAA fastball (around 4.9° in college baseball, and 4.4° in the MLB) up in the zone and a massive outlier is 8.6%, which is massive obviously. So now that we see how important a fastballs VAA is, let’s look at real life examples.

Amateur Pitching Prospect Examples

Typically looking at real pitchers or pitches helps picture specific things like a fastball having a “rising illusion” and this is a perfect opportunity to highlight what makes a VAA flat, or steep.

You can’t talk about vertical approach angle and NOT mention Jeffrey Hakanson, a fifth round pick in last year’s draft to arguably the most progressive team in baseball (Tampa Bay).

Of Hakanson’s 359 fastballs thrown since 2019, 100 are available with trackman data. His average VAA on pitches down the middle was 3.4°. And on pitches up in the zone, a whopping 2.86°. The reason for this is quite simple, his 4.37 release height is 1.51 feet shorter then the college average.The 17.02 IVB, and 92.14 release speed help out as well as both are well above average.

Hakanson’s lowest VAA on a called strike, is a 2.35°, but the video for the game is quite bad. So his next closest called strike (or at least it should’ve been) is this pitch against Auburn.

This pitch had a 2.97 PlaLocHei, so how it was called a ball is beyond me, but more importantly it had a 2.56° VAA. Now why? Well it had 24.41 inches of IVB (likely closer to around 21.5 because of amateur data’s inaccuracy) with just 3.77 inches of HB at 2,377 rpm and 92.9 mph. All of these numbers played a major factor, but his 4.65 release height played the biggest, and the 7.21 extension helped out as well.

Now let’s look at a more neutral fastball VAA, arguably the best pitcher in college baseball last year: Kansas City’s 4th overall pick. Asa Lacy’s average fastball in 2020 was thrown at 93.9 mph with 21.6 inches of VB, both of those are well above average and in the upper percentile ranking, but his average release height of 6.47 makes his fastballs shape appear less than you’d hope given the movement/velo. Lacy’s average VAA up in the zone was 4.4°, not great, and down the middle it was around 5.3°. Both less than ideal.

So let’s look at one of Lacy’s only fastballs thrown with a 3.5 PlaLocHei to see just how flat his VAA can get:

This pitches 4.2 VAA clearly didn’t turn out well, and the 6.38 release height doesn’t help. The fact that this pitch also likely had a > 99% spin efficiency (22.6 IVB, 2191 rpm) makes it interesting though.

Now, let’s limit the data to Lacy’s 18 swing-and-misses on the FB last year. The average VAA of every pitch is 4.87°, but as we know, that’s meaningless. Trimming it down to pitches up in the zone, it’s down to 4.13°, obviously much better than the 4.4° on every pitch thrown up in the zone.

One last pitch to look at:

Shockingly, a pitch thrown at 93.6 mph with a 3.38° VAA got a swing-and-miss. This pitch did have a 4.02 PlaLocHei though, so the location did play a massive part in the end result and VAA.

Conclusion

Vertical approach angle is a great metric when used correctly. Most people who put emphasis into it realize how location-dependent it is, unlike other metrics like release speed, spin rate, and vertical/horizontal break which are more useful on an average basis. VAA is more useful in one of two places: a pitch by pitch analysis, or taking other variables (release speed, release height, etc) and producing a VAA constant, which would be able to be compared to other pitchers without needing a larger sample in specific PlaLocHei.

While vertical approach angle is important, it is far from the most important metric. So we’ll revisit what goes into a draft model in the coming months before the 2021 MLB Draft.