Upset frequency right on schedule

Using the ESPN National Bracket to identify “favorites” in each of the games played yesterday, there were four upsets out of the 18 results.  That’s right in line with what I found when I looked at the issue a couple of years ago. 

A couple of years ago, I counted upsets in about five separate NCAA tournaments and found that the first round average is 8.  That is also in line with the results I got when I tracked NBA upsets.  Those occur in about 29% of the games.

That’s also exactly the number of wrong picks in each of my brackets.  One bracket was based on the results from the previous post, the other was based on straight Adjusted Marginal Win Score.  The misses were different, but the results were the same.

That’s why I stopped claiming I could provide any secret inside knowledge two seasons ago.  I can’t.  No one can.  If you pick based on strength, you will almost certainly lose.

In order to win, you have to either follow the roulette wheel strategy or the hedge fund strategy.  The roulette wheel strategy involves analyzing the stronger team in each matchup, and then going against that information on 25% of your picks in each round.  Basically, you have to get lucky.  (Bill Simmons annually follows a roulette wheel strategy, but he goes way too far with it.  He picks way too many upsets and normally finishes with terrible results.) 

The second, better strategy, is the hedge fund strategy, outlined on  Using this strategy, you pick according to strength up until the final rounds, at which time you analyze the pool of strong teams and pick the one that would be the lowest “consensus” champion. 

I accidentally followed the hedge fund strategy with great success a few years ago when Kansas won the championship.  My numbers showed Kansas, but the consensus picks that season were North Carolina and Memphis.  So when Kansas won, it set me apart. 

This season my brackets have no chance of succeeding for the same reasons.  But I knew that.

3 Responses to “Upset frequency right on schedule”

  1. J.D. Mo Says:

    Man on man, my bracket looks like Bill Simmons’ bracket!?! Figures.
    How did BS miss the Richmond “upset” of Vandy, though? That one was begging to happen, given the darn near fraudulent ranking of the Atlantic 10 tournament champ as a 12 seed.
    Oakland and Belmont were just too trendy for me, I guess, and I did two picks better than BS did, but a lot of it looks eerily familiar.

  2. Fred Bush Says:

    For your “roulette wheel” strategy I I think you’d be better off doing power ratings, and then randomly determining upsets based on the actual perceived percent win chance rather than just 25%.

    Trivial example: if you pick 25% of 16s to beat 1s your bracket is toast.

    • tywill33 Says:

      That’s actually genius! Obama appears to have used “selective intuition”. His big pick was the Vandy game. I’m sure he remembered Vandy’s past failures.

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 )

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

%d bloggers like this: