College Football

For more than 125 years, the process by which the highest level of college football determined its national champion was marred by controversy. Various mathematical formulas created by biased individuals and/or a plethora of subjective polls oftentimes resulted in more than one team claiming a national championship in any given year.  This remained commonplace through the 1997 season, at the end of which, the Associated Press (AP) poll named Michigan national champions, while the Coaches poll determined Nebraska was the best team in the country.


In 1998, the demand for a singular national champion was seemingly satisfied with the creation of the Bowl Championship Series (BCS). The stated goal of the BCS was to match the top two teams in the country in a true national championship game. However, at the conclusion of the 2003 season, there was another split national championship - LSU defeated Oklahoma in the BCS National Championship Game, but the AP voters decided USC was #1 after defeating Michigan in the Rose Bowl - and the controversy continued.  While most pundits agreed that the BCS was an improvement compared to the system it replaced, there was still widespread appeal for a playoff.


Thus, in 2014, the College Football Playoff (CFP) was created to replace the oft-criticized BCS. It was determined that a selection committee would be charged with selecting the four best teams to compete for the College Football National Championship. The winner of the CFP would be the crowned the undisputed national champion. This system was designed, in part, to eliminate the controversies which had plagued the sport since its inception. However, the reality is that the criteria by which teams are selected for inclusion in the CFP remains a matter of intense debate.


Should it be the four teams that are surmised to have the best chance to win a playoff game on a neutral field? The four teams that are surmised to be the most deserving based on résumé? Or some combination of the two? I generate an objective rating for all three approaches, updated weekly during the season, for each of the 130 FBS teams.

Best

The Best rating is a team's expected net point margin versus an average FBS team on a neutral field. In other words, it identifies which teams would have the best chance to win a game on a neutral field, if selected for the CFP. It is designed to be purely predictive, or forward-looking. 

Most Deserving

The Most Deserving rating is a measure of how deserving a team is of inclusion in the CFP on a 0 to 100 scale, where 100 is best. It is designed to be purely results-oriented, or résumé-based, and becomes progressively meaningful each week.

Composite

The Composite rating is a comprehensive measure of overall team strength on a 0 to 100 scale, where 100 is best. For those who remain undecided in the “best versus most deserving” debate, this rating is for you. 

College Basketball

Each year, for as long as I can remember, I have joined the millions of people worldwide who complete March Madness brackets with the immensely challenging goal of attempting to correctly predict each tournament game. Like many others, my bracket is usually busted on the very first day of the competition. Over time, I resigned myself to accepting the fact that I will likely never complete a perfect bracket.


So, in 2017, I modified my goal: create a prediction method to identify which tournament matchups are most likely to result in upsets. The results of the method to date are as follows:
 

  • 2019 – 99.1 percentile of all brackets entered on ESPN.com
  • 2018 – 50.5 percentile (thanks, UMBC)
  • 2017 – 76.8 percentile