iiinova — where your digital self keeps evolving

Reposted by Scott Powers

American Soccer Insights Summit @americansoccerinsights.com · 01/07/2025

We are thrilled to announce that the American Soccer Insights Summit will return in 2026. The summit will be held at Rice University on January 30-31. Register today at: www.ticketleap.events/tickets/amer...

American Soccer Insights Summit. Save the date: January 30 - 31, 2026. Registration Open.

0129

Scott Powers @saberpowers.bsky.social · 27/08/2024

New York Friends: I'll be on a panel at the first annual Symposium on AI & Sports hosted by the Columbia-Dream Sports AI Innovation Center on Thursday, September 12. Registration is free! (and there is a remote option) lnkd.in/g2deuMMS

000

Scott Powers @saberpowers.bsky.social · 21/08/2024

I'll be at Saberseminar this weekend! If you'd like to meet, book some time with me using my Calendly. calendly.com/saberpowers/saberseminar

010

Scott Powers @saberpowers.bsky.social · 21/08/2024

From novice to expert—STaRT@Rice has the workshops to match your research journey. Grow your skills and network with us! I’m excited to lead a session on ridge regression and the lasso in R. Register here: start.rice.edu #STaRTatRice2024 #Shapingthefuture

000

Scott Powers @saberpowers.bsky.social · 10/06/2024

The American Soccer Insights Summit is seeking sponsors to keep registration costs accessible and to fund travel for deserving students. If you are interested in being a sponsor, please reach out: americansoccersummit@gmail.com. Re-posts are appreciated! americansoccerinsights.com/sponsors

010

Scott Powers @saberpowers.bsky.social · 30/05/2024

That did not occur to me. I will check that out. Thanks!

010

Scott Powers @saberpowers.bsky.social · 21/05/2024

I just checked. There is a slight exacerbation with two strikes, but only slight. It's not strong evidence in favor of my hypothesis.

011

Scott Powers @saberpowers.bsky.social · 21/05/2024

So what's going on here? I *think* I did the calculations correctly. Further investigation is required. Perhaps batters are sometimes adjusting their swings just to make contact and foul the ball off. If so, we would expect to see this phenomenon exacerbated with two strikes ...

100

Scott Powers @saberpowers.bsky.social · 21/05/2024

Let's compare batter results on their "fast" swings (above that batter's average bat speed) vs. their "slow" swings. MLB avg (FB only, min. 50 comp. swings) "fast" swings 84% contact 44% hit into play 31% squared up "slow" swings 85% contact 34% hit into play 23% squared up

100

Scott Powers @saberpowers.bsky.social · 21/05/2024

Adi Wyner asked a good question on the Wharton Moneyball show last week: We observe that bat speed correlates negatively with squared up rate ACROSS batters. But what about WITHIN batter? For a given batter, when they swing harder, are they less likely to square the ball up?

210

Scott Powers @saberpowers.bsky.social · 14/05/2024

Conclusion: Let's exercise caution with our interpretations of swing length. It's probably true that longer swings lead to more swing-and-miss. But I need more convincing that the new data provide strong evidence in favor of this hypothesis.

000

Scott Powers @saberpowers.bsky.social · 14/05/2024

When hitters are fooled by off-speed pitches, they start swinging early. The "point of contact" ends up out in front of the plate, increasing the measured swing length. Here, being fooled is correlated with both miss-swings and swing length. We see the opposite for fastballs.

100

Scott Powers @saberpowers.bsky.social · 14/05/2024

What is going on with those long Cruz swings? Are they misses because they are long? Or are they long because they are misses? Here's the distribution of contact/miss swing length by pitch type. Miss-swings are longer for breaking balls and off-speed but shorter for fastballs!

100

Scott Powers @saberpowers.bsky.social · 14/05/2024

Here is a comparison of Cruz's and Soto's swings. At the risk of making the figure too complicated, I also annotated contact/miss. The upshot is that Soto never takes swings > 8.5 feet like Cruz does, and these swings are almost always misses.

100

Scott Powers @saberpowers.bsky.social · 14/05/2024

I wondered whether we could use this to measure swing adaptability: Which hitters are appropriately adapting their swings based on the pitch? Standard Deviation in Swing Length (min. 150 swings) #1 out of 186: Oneil Cruz (0.9 feet) ... #186 out of 186: Juan Soto (0.5 feet)

100

Scott Powers @saberpowers.bsky.social · 14/05/2024

This isn't unique to Cruz. It seems generally true across players. Players who swing hard tend to swing long. BUT for an individual player, swinging harder doesn't correlate strongly with swinging longer. I am curious what causes differences between swings for individual players.

100

Scott Powers @saberpowers.bsky.social · 14/05/2024

By contrast, here is a visualization of within-player variance, specifically for Oneil Cruz. We see there is not much relationship between bat speed and swing length across Cruz's swings. Cruz doesn't seem to swing harder when swinging longer.

100

Scott Powers @saberpowers.bsky.social · 14/05/2024

I'm particularly interested in comparing *between-player* variance in bat path and *within-player* variance in bat path. For example, here is between-player variance in bath path. We see that there is a correlation: Batters who swing hard tend to swing long.

100

Scott Powers @saberpowers.bsky.social · 14/05/2024

HUGE thanks to all the folks at MLB who put in the work necessary to bring us pitch-by-pitch bat-tracking metrics publicly available on Baseball Savant. This is an awesome dataset, and I appreciate the opportunity to play around with it.

media.tenor.com

Wowww Amazed GIF

ALT: Wowww Amazed GIF

131

Scott Powers @saberpowers.bsky.social · 21/03/2024

Better late than never, mRchmadness (w/ Eli Shayer) is up for 2024! We're NOT in the business of predicting outcomes. But if you have those predictions from another source, our tool maximizes your chances of winning your pool (using pick data from ESPN). saberpowers.shinyapps.io/mRchmadness/

001

Reposted by Scott Powers

Esteban Navarro Garaiz @estebanng.bsky.social · 12/02/2024

⚾ Save the date - April 13! Join me at the UConn Sports Analytics Symposium for a historical journey through Baseball Analytics. We will discuss its present state and open problems, how we got here, and envision the future. Don't miss out! statds.org/events/ucsas... #UCSAS2024

statds.org

UCSAS 2024 : Keynote Sessions

011

Scott Powers @saberpowers.bsky.social · 05/02/2024

Luke Stancil, Naomi Consiglio and I estimated a touch-by-touch point win probability model for women's college volleyball. We measured core skills (serve, receive, set, attack, block, dig) all on the scale of Points Gained. The preprint is now on arXiv: arxiv.org/abs/2402.01083

arxiv.org

Estimating individual contributions to team success in women's...

The progression of a single point in volleyball starts with a serve and then alternates between teams, each team allowed up to three contacts with the ball. Using charted data from the 2022 NCAA...

000

Scott Powers @saberpowers.bsky.social · 31/01/2024

Another one of my favorites: pebblehunting.substack.com/p/baseball-a...

000

Scott Powers @saberpowers.bsky.social · 31/01/2024

Having worked in MLB front offices as a data scientist for several years, sometimes it *is* hard to be romantic about baseball. Thank you @sammillerbb.bsky.social for rehabilitating that feeling within me over the past 11 months. I loved this piece. pebblehunting.substack.com/p/mike-trout...

150

Scott Powers @saberpowers.bsky.social · 31/01/2024

💯 This is my biggest lament with the task of creating player projections. You're always wrong, and all you can do is try to be (a little) less wrong.

001

Scott Powers @saberpowers.bsky.social · 17/01/2024

If you want to build a personal website, I recommend trying Quarto. Big thanks to Sam Csik for putting together the most delightful tutorial I've ever read. ucsb-meds.github.io/creating-qua...

ucsb-meds.github.io

Creating your personal website using QuartoPaletteTerminalR ProjectTerminalR ProjectFont AwesomeFace...

120

Scott Powers @saberpowers.bsky.social · 17/01/2024

Today I debuted my personal academic website! saberpowers.github.io For fun, I included a page on which I am collecting R&D jobs in sports that come across my radar.

saberpowers.github.io

Scott Powers

120

Scott Powers @saberpowers.bsky.social · 13/01/2024

First-years Brady Detwiler, Devin Abraham, Tyler Emanuel and Wyatt Bellinger developed Points Saved by Tackle (PST). For made tackles, they trained a neural network to predict the counterfactual EPA of the play if the tackle had been missed. www.kaggle.com/code/bradyd/...

010

Scott Powers @saberpowers.bsky.social · 13/01/2024

First-years Lou Zhou and Rahul Herrero developed Play Value Without Penalty (PVWP). They used random forests to gauge yardage penalties based on the expected yardage had the defender needed to make a tackle instead of committing the penalty. www.kaggle.com/code/louzhou...

100

Scott Powers @saberpowers.bsky.social · 13/01/2024

Juniors Jonah Lubin and Charlie Wells developed TacklrTrackr and Adjustable Tackle Metric Hub, two apps that allow users to visualize eight attributes they developed to measure tackling; find similar players; and create custom leaderboards. www.kaggle.com/code/jonahdl...

100

Scott Powers @saberpowers.bsky.social · 13/01/2024

I want to highlight three Big Data Bowl submissions made by Rice University students...

111