15 March 2009

In an effort to assist readers in filling out their NCAA tournament brackets, I've looked at results from past years (from the beginning of the 64-team era) to determine if there are any upset tendencies they can use to their advantage. There's nothing really earth-shaking, but here are my takeaways:

1) 1 seeds do far, far better than 2 seeds. Picking all 1 seeds for the final four is almost never a bad strategy.

2) 10-12 seeds win their first round matchups roughly a third of the time; interestingly, they also do well in second round matchups, and 10s and 11s do well in the sweet sixteen. 10 seeds are more likely to reach the elite eight than everyone except seeds 1-4 and 6. They have never made the final four, however.

3) The 8-9 game is a tossup in the first round, but that doesn't mean those seeds are interchangeable. In the second round 8 seeds beat 1 seeds 20% of the time, but 9 seeds almost never do.

4) 5 seeds almost never reach the elite eight.

5) Once in the elite eight, 1-5 seeds perform similarly.

6) Seeds 1-3 have won all but three of the championships.


Jon said...

Again, this, combined with Strat-o-Matic, convinces me that you dedicate entirely too much time to hypothetical sports venues. Aren't you married now??

Nat said...

I haven't played strat for years!

Jon said...

But the many years you did racked up.

Dave said...

This blog appears to be in danger of turning into Nat's sports blog. Will someone get married or something so we can talk about something else? Or better yet, come visit Europe!