Whiff+ updates, Swing Modeling, and Other Baseball Related notes

Philadelphia Phillies Opening Day 2026 photos, highlights

    Rejoice! Baseball is back! The 2026 MLB season kicked off last Wednesday, and new data is already being plugged into Whiff+ and other models which I will get into later. Anyways, here's a quick update on Whiff+ and some other projects I've been working on.

Whiff+ Expansion

    Whiff+ has been expanded to include every major pitch type, using the same methods I went over in the original Heat+ article. All of the non-fastball models are very much in their newest stage, and I haven't taken much time to go in and refine them. That being said, they are correlating well with whiff rates. 
    Sorry for the blurryness of that image, I'm looking for a new site to host this blog. Anyways, correlations are strong and show that high Whiff+ numbers correlate very well with whiff rates. Splitters have the best correlation, at an R^2 value of .742. Slider is the lowest at .56. All are between .56 and .75. 
   
    I have also been working on a command based model. The model looks more at how effectively the pitcher is executing the pitch. For example, a fastball down in the zone is less effective than a fastball up in the zone, even if they have the same exact characteristics. The COMMAND model would reward the pitcher who can consistently locate the fastball at the top of the zone. 
   
    For predictive measures, I compared both Whiff+ and COMMAND to Fangraphs Stuff+. 
    Whiff+ and COMMAND both predict whiff rates far better than Fangraphs. This makes sense, as my models have been trained to predict whiffs where as Fangraphs is trained on run values and outcomes. That's why Fangraphs is considerably better than Whiff+ at predicting xWOBA across all pitch types except fastballs and sweepers. 
    Seeing that Whiff+ is better at predicting xWOBA than Stuff+ is a very encouraging size, as the fastball model is far and away the one I've spent the most time on. 

Swing Modeling

    Unlike pitch modeling, swing modeling is a very new idea in the Sabermetrics world. Statcast has been tracking swing data for a few years now, and no one has made a very good swing model yet. I decided to take a crack at it, creating a power based model and a contact based model. There's no public app to search through the database yet, but I am working on one. If there was, looking at some of the top 10s and bottom 10s is very strange and shows how little we (I) know about swing modeling. 
    As we can see, correlations aren't that great. POWER and CONTACT are solid at predicting what they are trained to predict (hard hit and contact respectively), but aren't awesome. Driveline has the best swing model I've seen, and if Eno Sarris uses it it must be solid. But even that one isn't great, as swing characteristics have less impact on pitch outcome than pitch characteristics do. I do plan on playing around with this a little bit, but I believe the stats themselves (Tilt and bat speed most notably) are better predictive and descriptive stats than any swing model out there now. 

Other Baseball Notes

    Baseball is back! This season, I want to continue to be more of a baseball fan, and not just a Red Sox fan. I want to try and release power rankings every few weeks, but I am busy with my actual job, school, and this internship in the summer. Additionally, I'll try and use these models to highlight some cool players or anything I find interesting. Like I said, I hope to move this blog somewhere better at some point, but I haven't really looked to far into that yet. 











Comments

Post a Comment

Popular posts from this blog

Introducing Heat+, a Brand New Fastball Quality Model

Image
          A few weeks ago, Camden Quick, who I will be interning with on the cape this summer, developed his own Stuff+ model to use in college ball. Taking after him, I’ve been working on my own pitch model. This is the first iteration of the model, and I plan on creating a better viewing application to go alongside with. With that being said, here is the preliminary version of Heat+.      Heat+ is a pitch model designed only for fastballs at the moment. Heat+ was created in R, using XGBoost’s gradient boosting techniques to capture more. I took fastball data from 2023, 2024, and 2025, giving my model 3 different seasons to work with. The model is trained to predict the run value of a certain pitch. To define the run value, I used statcasts built in run value metric, “delta_run_exp”. This is the change in run expectancy from before the pitch to after the pitch. For a more detailed explanation on how pitch models such as this are made, read ...
Read more

What should the Sox do about infield depth?

Image
Vaughn Grissom has a lot to prove this spring. Maddie Meyer, Getty Images With the signing of Alex Bregman, the key positional battle (second base) has likely been solved. This leaves the Red Sox with an infield likely to consist of Bregman, Rafael Devers, Trevor Story and Triston Casas. That's a strong infield, but there are a lot of durability questions still. This leaves infield depth as one of the most important areas the Red Sox should look to figure out. For this position, I see these 5 players as the only 5 to have a reasonable shot at making the opening day roster: Romy Gonzalez, Nick Sogard, David Hamilton, Vaughn Grissom, and Kristian Campbell. With Cora and the front office expressing a desire to use Hamilton in the outfield, I consider him the most probable candidate to make the team as a utility and speed guy. Hamilton ranked in the 99th percentile in terms of sprint speed, and contributed 6 Baserunning Runs and 4 Basestealing Runs, both ranking above the 90th percenti...
Read more

Red Sox Game Preview: Red Sox vs Blue Jays, 2/23/25

Image
Alex Bregman gets his first start for the Red Sox       Both Red Sox marquee signees will make their spring training debut's today, as Garrett Crochet gets the start and Alex Bregman starts at third for the Sox. Crochet, 25, was acquired from the White Sox for a major haul of prospects this offseason. Bregman recently signed a 3 year, 120 million dollar deal with opt-outs after the first two years. Here's your lineup for todays Grapefruit League clash:          Roman Anthony, LF     Alex Bregman, 3B     Triston Casas, 1B     Trevor Story, SS     Rob Refsnyder, DH     David Hamilon, 2B     Ceddanne Rafeala, CF     Blake Sabol, C     Nate Eaton, RF          2024 All Star MVP Jarren Duran was a late scratch to the lineup as he battles a sore calf, Alex Cora notes it's nothing serious and he expects the star, and catcher Connor Wong, to be back tom...
Read more