Statcast Pitch Tracking: Using Sabermetrics for Fantasy Baseball

When you're evaluating a pitcher by Pitch Info on FanGraphs, sometimes you see pitches that don't seem to accomplish anything. If Pitch X is seldom a strike, rarely chased outside of the zone, and not generating whiffs or a favorable batted line against, why doesn't the pitcher scrap it entirely?

Some pitchers could improve by dramatically decreasing how often a particular pitch is thrown or even removing one from their arsenal, but the rate of this is much lower than you'd expect at first glance. The reason why is a key piece of baseball that Pitch Info doesn't really get into: platoon splits.

Baseball Savant gives fantasy managers unlimited access to platoon splits by pitch type, allowing us to make an informed decision about whether a pitch mix change would truly prove advantageous. The best way to access this information is by typing in a pitcher's name in the search bar on the top-right corner of Baseball Savant and scrolling to the second table, labeled "Pitch Tracking." Next, toggle "Handedness" to either righties or lefties and give it a few seconds to refresh. You're ready to rock!

Interpreting Pitch Tracking Data

The Pitch Tracking table can be intimidating when you first look at it, so let's get that out of the way right now:

We start with the total number of pitches thrown, pitches thrown versus RHB and LHB, and pitch usage percentage. Adding pitch velocity, plate appearances, at-bats, hits, singles, doubles, triples, homers, and batted ball events (BBE) is probably unnecessary on this chart, but they are grouped so you can ignore them fairly easily. Similarly, average exit velocity and launch angle feel extra when the xStats that consider them are right there.

One of the biggest downsides of platoon splits for individual pitches is that you're almost always working with small sample sizes, meaning that fortunate bounces could skew the results. Luckily, the table's inclusion of xStats lets you know both how hitters fared against a particular pitch and how they should have fared. Whiff% is tempting, but remember that it isn't the same as the SwStr% cited from FanGraphs. Years are listed in reverse order, so the most recent is on top.

Let's consider Max Scherzer as an example of these numbers in action. The veteran ace threw five different pitches in 2021: fastball (46.7%), cutter (9.5%), change (14.9%), slider (19%), and a curve (9.8%). Per Pitch Info, Scherzer's fastball and curve get him ahead in the count with Zone% rates above 50 while his slider and cutter have impressive SwStr% rates to put hitters away. His change sticks out because of its mediocrity: 13.4 SwStr%, 39.4 Zone%, and 31.1% chase rate.

The lazy analyst might determine that Scherzer should scrap his change, but Statcast Pitch Tracking reveals that it's his go-to weapon against lefties. Scherzer used his change 20.1% of the time vs. LHB in 2021, and hitters didn't do much with it with a .177 average and .228 slugging. xStats suggest that those numbers were deserved with a .200 xBA and .281 xSLG.

None of Scherzer's other secondary offerings generated similar results. Lefties hit his cutter for a .221 average with a .243 xBA and slugged at a .485 clip with a .488 xSLG. His curve fared a little better with a .265 average (.219 xBA) and .441 slugging (.389 xSLG), but it was clearly worse than the change. Finally, Scherzer only threw 19 sliders to LHB all season. He hasn't thrown a significant number of sliders to lefties in a season since 2016, and they brutalized it for a .277 average and .511 slugging percentage that year.

Scherzer's change might not produce the strikeouts that fantasy managers are looking for, but the rest of his repertoire generates plenty. If Scherzer expects a lefty to put the ball in play, his change is clearly the best option available to him.

What's Below the Pitch Tracking Table?

There is a ton of information beneath the Pitch Tracking table, but most of it is available elsewhere or of limited use in fantasy. You may have seen analysts citing Vertical and Horizontal Movement for individual pitch types in the past, but that's too abstract for average fantasy purposes. Furthermore, the graphic that compares pitch movement to a hypothetical "average" pitch draws from all offerings within two miles per hour and six inches of extension in either direction.

That's a problem because the net is too broad. Returning to Scherzer as an example, his slider is a high-spin breaking pitch intended to strike out RHB while his change is low-spin to limit the contact quality of LHB. Scherzer gets nearly the same extension on all of his pitches and less than an mph differentiated them by pitch velocity last season, so they're compared to the same "average" pitch despite doing completely different things.

Run Values by Pitch Type are way too abstract. The Spin Direction table includes active spin if you're looking at a specific pitcher instead of the leaderboard, but this author feels that it's easier to put it into context with the leaderboard. Plate Discipline metrics are available in cleaner forms elsewhere, and the Expected HR by Ballpark is inherently flawed since it doesn't consider altitude. Shift statistics might be legislated out of existence soon, and Pitch Distribution is covered by the Pitch Tracking table.

Parting Thoughts

Platoons matter in baseball and Baseball Savant gives us the tools we need to identify which pitches are a given arm's preferred weapons to go after righties and lefties alike. As always, it is best to combine this information with other data sources to determine who you want to roster to dominate your fantasy leagues.

Download Our Free News & Alerts Mobile App