Our (Tentative) Plan for Our College Sports Models This Fall

Ok, it’s been a while since we’ve tried to forecast our progress with our predictive models, but with fall college sports getting closer, it seems like time to set some goals for ourselves.

On the business side of this blog, one of the best routes to viability is becoming a leader in collegiate predictive models. Bracketology, bowl projections, probabilities galore, across as many sports as possible. ESPN is broadcasting more and more college sports, which is both a good indicator of and a good vehicle for rising interest in these sports. There are lots of overlaps between the consumer bases of individual sports. We are familiar with the NCAA’s procedures for football and men’s basketball championships at the Division-I level, and with things like conference structure. It combines nicely into a potential niche for us.

To that end, last summer we identified ten sports we wanted to expand our models into, already having models built and in use for football and men’s basketball. These were the ten:

  • Women’s basketball
  • Baseball
  • Softball
  • Men’s and Women’s Lacrosse
  • Men’s and Women’s Soccer
  • Women’s Volleyball
  • Men’s and Women’s Ice Hockey

The idea is that we’d have four models running per season, with tournament appearance probability, championship probability, and bracketology or the equivalent for each, and with expansion hopefully on the medium-term horizon in terms of depth (more features sharing more information), quality (more accurate and precise predictions), and breadth (more sports than even these twelve). We’d keep our NHL model going. We’d finish our NASCAR, Formula 1, and IndyCar models and launch those. We’d consider modeling in other professional and amateur sports (the WNBA comes to mind as one where we’re unaware of a prominent playoff and championship probability model, Texas high school football comes to mind as one where there might be a market large enough to justify the time investment).

With that, then, our hyper-tentative target launch dates for our fall models are as follows:

  • Football: July 28th (One month from kickoff)
  • Women’s Volleyball: August 4th
  • Women’s Soccer: August 11th (ten days from kickoff?)
  • Men’s Soccer: August 18th (ten days from kickoff?)

There’s a lot we don’t know, so if you have information on any of the following areas that would be of help to us, please share it:

  • A handy FBS bowl selection flow-chart
  • Women’s Volleyball selection criteria, bracketing rules
  • Women’s and Men’s Soccer selection criteria, bracketing rules
  • Women’s Volleyball scheduling information, specifically the first date games will be played this fall
  • Women’s and Men’s Soccer scheduling information, specifically the first respective dates games will be played this fall.

We are hoping to have both the FCS Playoffs and FBS Bowl Projections as part of our football model, which would constitute an expansion of that model’s scope, so look forward to that. I’d say it’s a priority over the volleyball and soccer models, but that could change depending on feasibility. Thanks for any help you can provide. We’ll see whether we can pull off launching three new models on the back of significantly expanding an existing model. Lots more to come.

The Barking Crow's resident numbers man. Was asked to do NIT Bracketology in 2018 and never looked back. Fields inquiries on Twitter: @joestunardi.
Posts created 3299

Leave a Reply

Your email address will not be published. Required fields are marked *

Related Posts

Begin typing your search term above and press enter to search. Press ESC to cancel.