Luke Neuendorf

Data sample is from the 2013 - 2024 seasons. 1570 unique season + team combinations.

58.5% of teams committed fewer penalties in the second half of the season than in the first half, suggesting that teams tend to become more disciplined as they gain experience over the course of the year.

Do college football teams stay consistently disciplined?

Only 17% of the variation in penalty counts from the first 6 games predicts the next 6.

Penalties appear to be influenced more by context — game situations, officiating, and pressure — than by team identity alone.

I recently build a 4th down decision model for college football. However, the win probability model is failing to capture late game decision making logic.

I've started working on a Monte Carlo simulation model of CFB games and it is turning into a rats nest quickly 😅

This is a cool visualization! The NE vs ATL chart is wild

Thanks @statsbylopez.bsky.social, Tom Bliss, and everyone else is involved in running this, it’s a fun competition!

And the y-axis label on the second plot should say "Rush Yards Gained", as there is no averaging going on in that plot.

When not taking the aggregated mean of rush yards gained by the number of defenders in the box, the relationship is less clear. However, the first slide still shows that on average, adding defenders to the box stops the run. The first approach smooths out play-to-play variability.

It's an obvious conclusion that adding more defenders to the box will stop the run, but i never thought the relationship would be this clear.

This is a regression between the number of defenders in the box and the aggregated mean of rush yards gained for each number of defenders in the box.