I'm not doing my econometric model for the NCAA tournament this year. Several reasons.
#1- time constraints. I have a million other things I don't have time to do but are quite important. It takes forever to record the data I need by the day of the first game and next week happens to be my busiest work week.
#2-As Taleb says: "All econometric models are useless." If the financial crisis has taught us anything it's that you can't reliably base future predictions on past data (think housing prices, it just wasn't so obvious to the quants who are now job-hunting).
#3- I, like thousands of other Kentucky fans, am debating whether or not to even support the team accepting an NIT bid. That actually has nothing to do with not doing the model this year, but thought I'd throw it out there because it's a window into my emotions.
#4- Money is tight and I don't want to spend the money to get the supplementary data that I purchase every year.
If someone else wants to help out with #1 or #4, let me know. I still would like to record data for this season so that I can continue the model next season. Mainly because it was fun.
My sense for this season is that the tournament is (obviously) wide open. Banking on the Pomeroy efficiency numbers won't lead to 99% success rates like last year because there is much more parity this year.
My advice- After advancing mostly favorites in the first 2 rounds (including all 1-4 seeds), just pick randomly.