David Gerth

I heard you said “the MLB”

I learned statistics in university.

I learned how to be a statistician on twitter.

I was offered many jobs because of twitter.

I retain some connections I've made over twitter.

Twitter, or what it was circa 2019, was instrumental to the success I've had if I have had any to begin with.

Very excited to welcome @bgawlowski.bsky.social back to the site as our new prospect writer!

Bat speed aging curve comparing two methods, one with no correction for age and another that corrects for age (fewer younger players and older players)

Bat speed aging curve, but this time I account for the fact that we are less likely to observe very young/old players. The corrected curve declines less rapidly early and more rapidly later compared to the regular curve, which seems like the correct inference:

The cover of Atlas Shrugged, but instead the title reads, "Atlas Needs You to Spot Him Like 20 Billion Dollars."

Looks great Thomas, congrats!

didn't think we could hit another low with ex players on panels but think we got there with will compton/taylor lewan

How Should You Think About Your Priors for a Bayesian Analysis? Here, I riff on how students should think about priors for a Bayesian analysis. It seems daunting, but it just means you should use your head and not expect software to think of your data/model for...

I like this: svmiller.com/blog/2021/02...

Article: dgerth5.github.io/swing-type-p...
Shiny App: david-gerth.shinyapps.io/predicted-sw...

Lastly, I put together a Shiny app for the 2024 season so you can see the predicted swing types for all 650 hitters, along with a summary of how each of their swings work and what the results are

I also look at predicting hitter specific swing types given location + count, and see how hitters differ in their approach

New article today about work I’ve done on classifying swing types and how hitters vary their swing based on location and count. Personally, I’ve found it hard to group swings together in my head based on the new data, so this attempts to do that.

YouTube video by owen reiser LISTERS: A Glimpse Into Extreme Birdwatching

This is easily the best thing YouTube has recommended to me in the past five years. www.youtube.com/watch?v=zl-w...

I think that was a home run, and I don't think that fan reached, but do I think folks are generally gettin' just a little too comfortable at the wall this year!

An acceleration of an existing trend this year: First-time head coaches hiring former HCs as coordinators.

I traveled around the NFL this summer to understand what that type of partnership really looks like, from honest reflection and tough lessons to truly ruthless competition:

Nice stuff Jonathan. Not sure if you are familiar, but there is an INLA analogue called INLAbru for spatial statistics. It's a nice package to test out if you haven't already

Had some time and decided to join the CFB Data Pick'em this year. Here is the writeup of my model: dgerth5.github.io/cfb-model-20...

Just pushed my first submission. This was a ton of fun to work on

During grad school, I sent actual letters through the actual mail to five GMs, asking how a 25-year-old with an MFA in poetry and no other credentials could get involved in baseball. Only one responded, and the answer was some (polite) version of, "I regret to inform you that your dream is doomed."

FanGraphs Turns 20 Years Old Today! Where does the time go?

FanGraphs Turns 20 Years Old Today! blogs.fangraphs.com/fangraphs-tu...

But pivoting to AI is not a business strategy. Telling journalists they must use AI is not a business strategy. Partnering with AI companies is a business move, but becoming reliant on revenue from tech giants who are creating a machine that duplicates the work you’ve already created is not a smart or sustainable business move, and therefore it is not a smart business strategy. It is true that AI is changing the internet and is threatening journalists and media outlets. But the only AI-related business strategy that makes any sense whatsoever is one where media companies and journalists go to great pains to show their audiences that they are human beings, and that the work they are doing is worth supporting because it is human work that is vital to their audiences.

Found this @jasonkoebler.bsky.social story about why AI isn’t a viable business model for journalism to be extremely well done and oddly hopeful

www.404media.co/the-medias-p...

Something I'd be curious to see is a comparison between a model w just a hitter random effect vs model w the underlying data

wow, all these Dems looking for the left's Rogan, as if a thrice-weekly, very long sports podcast that loves stats and science BUT ALSO baseball players kissing THOUGH ONLY IF THEY WANT TO isn't right there WOW

underrated worst part of the lottery is Morey getting bailed out. It's like seeing SBF getting pardoned

Take Me Out to the Mudcats An Experience with A-Ball’s Most Exciting Squad

Our recent piece with @downonthefarm.bsky.social spotlights the Carolina Mudcats, who may be MiLB's most exciting team! We take you on a gameday experience with analytical evaluations and the good old' eye test. Read it here: downonthefarm.substack.com/p/take-me-ou...

I keep trying to hammer the point that LLMs do a much better job replacing the work of MBAs and coders than liberal arts majors.

How to Fit Monotonic Smooths in JAX using Shape Constrained P-Splines Let’s say you have a trend you are trying to model that you know to be monotonically increasing or decreasing; this could be something like default as a function of risk, power usage as a function of ...

I've had instances in the past where GAMs make crazy predictions out of sample and interested to see if this methodology helps that. Here's the article: statmills.com/2025-05-03-m...

Read this nice article on Shape Constrained Additive Models yesterday. Tried it out today on some generated data, comparing a standard GAM to a SCAM with upward trend (correct) and SCAM with downward trend (incorrect). GAM and correct SCAM are similar, with the SCAM being less wiggly.

Two of the offices without senior officials are Houston and Tampa, which, whew, that’s not good.

A slow moving catastrophe that won’t be widely understood until it’s too late.

And they say you can't make nice graphics without ggplot