For example, it’s probably not too hard to model Dorian Finney-Smith’s player-adjusted shot quality, because he takes mostly basic shots. How would you suggest modeling the Doncic-specific difficulty of Luka’s running Dirk-step three?

can't wait to publish my #NBA visualization apps

1/ The difference in skill between our #1 ranked player, Rahm, and our #2 ranked player, Schauffele, is 0.52 strokes per round! This is the largest non-Tiger gap between 1 and 2 since 2004:

It has been 33 years or so since my first work on points per possession got published. I was at Caltech, studying hard, playing hard. I had no idea then it would lead to all of this. archive.org/details/sim_ba…

Question asked in an NBA chat I'm in about how many possessions are needed to trust 5-man lineup data (CC Jesse Pittsley). I ran padding values (kmedved.com/2020/08/06/nba…) on 5-man lineups since 1997-98 by game to see how much padding was needed to predict the next game. (1/x)

I did this separately for offensive and defensive units, and combined it with the DARKO time-decay approach. The results: 1) Offensive lineups take about ~550 possessions together to "stabilize". 1) Defensive lineups take about ~850 possessions together to "stabilize". (2/x)

Happy to say DARKO's win projections were the best performing "player-based" win projection models tracked this year, going 21-8-1 vs. Vegas. They would have also won the ABPR contest. The highest performing model overall was the DARKO preseason-only model as well.

NFL Preseason Results Are Predictive: a Thread I have previously found preseason are fairly predictive of NBA regular season results (x.com/kmedved/status…). Others have found similar results for baseball (CC Dan Rosenheck). I was wondering the same re: NFL. (1/x)

Lots of questions about how much to read into NCAA RAPM for NBA draft prospects. How about we do a deep dive analysis? 1/

I’ve learned a lot about football reading Ian over the years. Excited for this. Check it out for some strong football takes.

Why Steph Curry is the greatest offensive player in the history of the NBA. americaswargame.substack.com/p/stephen-curr…

Added a new stat to our BBall Index applications that I've been calculating all season. It's very similar to the structure of LEBRON but instead of using box LEBRON as the prior, it uses Daniel Myers BPM from bball-ref.

Re-releasing this Jonathan Tjarks podcast on The Meaning of Life. He's brilliant in it. Rest in Peace, Jon. We love you. houseofstrauss.substack.com/p/hos-pod-with…

These words of wisdom from Jonathan Tjarks last month (in an interview with Chris Carrino) on his perspective of life and the prospect of death were so moving and continue to stick with me. He knew what and who was important to him, and lived it with clarity and purpose

Reflecting on one of the most meaningful friendships of my life: americaswargame.substack.com/p/the-triumph-…

Jonathan Tjarks was a man of God. I have great respect for the way he finished the course. Truly a life worth celebrating. "I have fought the good fight, I have finished the course, I have kept the faith; in the future there is laid up for me the crown of righteousness..."

My version of the NBA play-by-play data for the 2022-23 season, with lineups on the floor during each event + possession count + column to filter out garbage time can be found here, along with #rstats code to replicate it: github.com/ramirobentes/N…

Blind RT, must read etc etc etc

This is a good history of how the basketball analytics community was hatched, well before it was a job.

The APBRMetrics message board was where I learned almost everything I know about basketball analytics. That and Tangotiger 🍁's blog.