Guestie: Blind bracket building

Posted: March 14, 2012 by terryvandrovec in Guesties
Tags: , , , , , , , , , , , ,

By Rich Jensen
Guest blogger

The NCAA tournament selection and seeding process will never be perfect. There is simply no perfect way of evaluating teams that play only a small subset of each other, and especially when the difference in resources available to teams enables some teams to barely stir from home arenas for months at a time while others end up traveling more than the Harlem Globetrotters.

However, there is a clear and simple way to remove a significant source of bias in the selection process:

Use a blind.

In clinical tests, doctors are not allowed to know whether they are administering drugs or placebos to patients, because knowing this would affect the way patients are treated and would taint the data.

Similarly, when the selection committee includes conference commissioners, university athletic directors and individuals with
strong connections to universities and conferences, bias is unavoidable as long as the committee knows the identities of the
teams it is evaluating.

So why not eliminate the bias?

My modest proposal is to prepare software and index cards that contain a short code identifying the top 100 schools in RPI (ideally, a better metric would be used, but that’s a whole other argument), along with the conference auto-bids, as they are determined. Data on the index cards would include SOS, top 50 wins, top 100 wins, winning percentage, etc. (like the RPI metric they’re based on, these are flawed metrics, but again, why fight more than one battle at a time, right?)

The conference name would not be included.

The only additional data would be a schedule that would show the score and RPI of opponents, with no note as to whether a game was in or out of conference, at home, on the road, or at a neutral site, and this information would only be available on request by the committee, not by individual members.

Once the teams are seeded and placed in their respective regions (the S-curve would be retained), a computer program would determine the location of play based on the weighted mean distance between schools, with the top seed in each pod receiving more weight than the other three seeds. Teams would be placed at sites beginning with the highest number one, down to the lowest number two.

No consideration would be given to teams from the same conference meeting up in the first rounds, nor would consideration be given for teams playing first round games at their home arenas.

Would this be a perfect system?

Of course not.

Would it be better than the chaotic and biased selection process in place today?


About the author: Jensen has had his blood pressured raised recently by the NCAA over its treatment of his favorite school, South Dakota State, when it comes to the bracket process.


Leave a Reply

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

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s