Aggregation DistortionI continue to be surprised by how aggregating multiple subgroups into one can distort the picture and make understanding more difficult.Sep 22Sep 22
Building a Better, Simpler Model of Running Back SuccessFor several years, I’ve been publishing predictions of NFL rookie success odds for wide receivers and running backs over at 4for4. The wide…May 20May 20
Myths and Reality of AI Writing CodeI’m a big fan of Jensen Huang, and it is obviously silly for me to try to give him any kind of advice, but if I were in his shoes, I would…Feb 29Feb 29
What Happened to Value in 2010?Last month was another good one for growth stocks, continuing the trend that started in early 2023. Like many investors, however, I am on…Feb 16Feb 16
Adjusting for Strength of ScheduleAnyone who knows me knows how closely tied I am to the University of Washington: I’m an alum, my wife is an alum, my parents are alums, my…Jan 6Jan 6
A More American College Football Playoff SystemLike many sports fans, I was shocked by the College Football Playoff (CFP) Committee’s decision to leave out Florida State, the undefeated…Dec 6, 2023Dec 6, 2023
Reflections on Ben GrahamA couple weeks ago, Ben Claremon of the Compounders podcast wrote an article about his investing journey that started with the approach…Sep 18, 2023Sep 18, 2023
Why Do We Teach Under-Parameterized Models?When I first studied machine learning, over 20 years ago, the course started with classical statistical techniques (so-called “maximum…Aug 31, 2023Aug 31, 2023
Limits to Super-IntelligenceWhile the community struggles to sort out how the capabilities of LLMs compare to those of humans, it may be easier to put limits on the…Apr 3, 2023Apr 3, 2023
Something to See HereLast month, I wrote up some of my thoughts on ChatGPT. That post was a reaction to other opinions I’d seen, namely, the “it’s so over” and…Mar 26, 2023Mar 26, 2023
