Home > NCAA > Round 3 Complete. Update!

Round 3 Complete. Update!

Writer Method Round 2 Winners Round 2 Upsets Round 3 Winners Round 3 Upsets Score After Round 3 Best Possible Score
David SRS & Win Shares 23 2 / 6 8 1 / 6 39 143
Kai None 25 4 / 8 10 1 / 4 45 133
Marcus Team Names 24 1 / 3 8 0 / 2 40 124
Matt Bitterness 24 3 / 9 8 1 / 4 40 152
Sean 3PT/FT Percentage 21 3 / 10 11 3 / 6 43 167

Rounds 2 and 3 are now complete.  Ignoring the fact that they should still be called Round 1 and 2 (the play-ins don’t count in our opinion), we’ll give a brief summary of our writer’s score totals, and give some initial thoughts on the success of each methodology.  The leaders in each category are in bold, and the last-place writer is in italics.

To summarize, Kai is in the lead with 45 points, followed closely by Sean at 43 points.  What’s interesting is that Kai is in second-last place for potential points remaining, largely because of his nonsensical approach to the Southwest division.  By projecting a winner in a game without a participant, he actually hedged against himself, lowering his total possible points.  This approach will likely ensure that he doesn’t finish in last place, but will hinder his chances of winning the entire tournament, especially after already losing 2 Final Four teams.  The odds of him caring about this are about the same as a leprechaun becoming a UFC heavyweight world champion.

There are some fascinating lessons to be learned from this group approach to betting.  Taking a look at Round 2, you’ll notice that 4 of our 5 writers scored between 30-33% in upsets, with Kai being the lone exception at 50%.  Given that there were 7 total upsets in the first round (and if you think about it, one was an 8-9 upset which shouldn’t qualify as a real upset), all of us except Kai did worse than if we had picked a purely conservative “high-seed” bracket.

However, that wouldn’t be much fun to bet on.  And really, our writer average of 7.2 upsets predicted in the second round was almost bang-on with the 7 actual upsets, indicating we had a good feel for a reasonable number of total upsets.  By random chance, predicting 7 upsets out of the 32 games should have given us a 22% chance of calling the upsets correctly, meaning our 30-33% rate is above random chance (even if the majority of our upset picks were 8-9, 7-10, 6-11 or 5-12 upsets, which historically happen more frequently than high-seed/low-seed).

What other lessons can be learned?  Well, for one, Sean has the highest potential score despite the worst opening round of all the writers.  Given the point scoring structure we’re using that emphasizes the later rounds, it’s not a surprise, but its a reminder that you shouldn’t worry too much about the opening round if you’re just betting against a few friends on Facebook.  Yes, that Morehead or VCU upset would have been nice to predict if you’re trying to win the ESPN Bracket Challenge against a couple million other people (David was the only one to call those two upsets of the 5 writers, even though he missed other upsets that every other writer correctly anticipated), but in small groups, you’re much better off picking high-seeds to advance far, with maybe one or two Cinderella teams (5-8 seeds) making the sweet 16.

Fun statistic: Potential points remaining standings combined with total combined upsets predicted through Rounds 2-3.  Notice a trend?

Sean (167 points) – 16 upsets

Matt (152 points) – 13 upsets

David (143 points) – 12 upsets

Kai (133 points) – 12 upsets

Marcus (124 points) – 5 upsets

Funny coincidence, or could there be another reason?  We’ll do some digging and provide another update after Round 4 games.

  1. March 23, 2011 at 8:46 am

    Nice analysis!

    I am still bitter about calling it Round 2 & 3. The NCAA really dropped the ball there. The play-in games are not Round 1 to anybody!

  2. David
    March 23, 2011 at 6:02 pm

    Agreed…Round 1 and 2 are the four glorious days from Thursday-Sunday where I forget about everything and see more Greg Gumbel than his wife.

  1. No trackbacks yet.

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 )

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

%d bloggers like this: