In the next three weeks, SEC fans will ponder, debate, agonize and pray over which school their most-coveted recruits will choose on National Signing Day.
But a mathematical model has already made the call for all of them – sort of.
A student at the University of Iowa has developed a math-based formula to predict where top recruits will sign, and it seems to be working pretty well, as reported by Jacob Bogage of the Washington Post.
The model, designed by Iowa Ph.D. candidate Kristina Bigsby, relies on social media posts and “basic biographical information” – presumably including publicly available data like hometown and school. Her work was published in the journal Decision Analysis, from INFORMS (Institute for Operations Research and the Management Sciences).
The reported accuracy rate is 70 percent – not bad for a purely math-based formula that involves zero direct interaction with recruits and coaches.
Among the biggest predictive variables in the study: Hashtags (or #hashtags, if you prefer). When a recruit tweeted a school’s hashtag, his likelihood of subsequently committing to that school surged 300 percent.
This often works:
Florida tomorrow 🐊🐊👀 #AllBite18
— Amari Burney (@amari_burney) March 24, 2017
— Jamaree Salyer ♠️ (@jamareesalyer69) December 12, 2017
Official to auburn this weekend!!✌🏾#WDE
— Harold Joiner (@hkjoiner3) January 12, 2018
But not always:
Made it to Tuscaloosa 🐘🔴 #RollTide
— Justin Fields (@justnfields) September 29, 2017
This model isn’t going to replace face-to-face recruiting for coaches, and it won’t supplant message boards and plain old heated debate for fans. But is it worth a look? You make the call.
© 2016, gridironnow.com. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.