May 19, 2019

SUR data to be used during Harvard symposium

I’ve always wanted to be accepted in the Ivy League and with the help of Edward Egros, a sports analytics journalist with the Fox affiliate in Dallas, I’ll finally make it. Kind of.

Over the last year, Edward and Charles South, a statistician who is passionate about sports, have been working on a study applying advanced statistical models to determine what’s the best way to forecast the outcome of games (and as an extension, how to pick games against the spread).

A few weeks ago their findings were presented for an R Users Group in Dallas, and next month they will present a poster at the New England Symposium on Statistics in Sports (NESSIS) at Harvard University.

The Seldom Used Reserve database contains over 4,300 college football games and is used by many visitors to this site and a large group of Reddit users to model systems for predicting college football games, as well as in research and master’s level projects.

Over the years I’ve heard from degenerate gamblers, graduate students, fantasy football players, statisticians, mathematicians and, now, a journalist who will use the data at an Ivy League institution.

Thank you to Edward and Charles for their work and including Seldom Used Reserve.

%d bloggers like this: