After a great deal of hours put into data compilation, statistical analysis, and number crunching, I have finished my model. My goal was to be able to compile a formula that projects the likelihood that any one of the given teams in the field of 68 wins the NCAA Tournament. There are a good amount of extra variables, such as distance traveled and overall team tournament experience, that I wished to include. However, I was not able to compile some of these due to time constraints and my bracketology work.

In the end, the final version of the model includes a whopping 29 variables. The data collected and analyzed featured statistics and results from the NCAA Men’s Basketball Tournaments ranging from 2010 to 2015. Based on previous years, it appears that the most important variables in determining a team’s likelihood of making the final four include: strength of schedule, preseason AP poll points, offensive and defensive efficiency, and offensive rebound percentage. Once a team makes the final four, the amount of tournament wins and the tournament winning percentage of the coach of each team becomes more relevant.

Tournament Projections

Do with the information and projections above what you will, and good luck to anyone filling out  a bracket – March is the best time of the year!

 

Advertisements
Comments
  1. Mike Woz says:

    Do you take last 10 games into consideration? I’m surprised not seeing Seton Hall. Since the tournament is all about getting hot at the right time – ala Connecticut 2014; 2011.

  2. JLCook33 says:

    Though it was under consideration, a team’s record in their last 10 games going into the tournament did not end up as a variable. I completely agree that the tournament is about finishing the regular season and conference tournament on a high note. However, obtaining historical last 10 game records for teams going into the tournament in 2010 through 2015 proved to be too time-consuming of a task. Instead, a metric like that, along with distance traveled and coach tournament wins, should be accounted for in a qualitative manner.

    It’s also of note that those two Connecticut teams were led by exceptionally talented scoring guards (Kemba Walker, Jeremy Lamb, Shabazz Napier, Ryan Boatright). That is another pattern that IS accounted for in the model.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s