March Madness: Opening Round Review

Well, it appears that my bracket can be safely torn to shreds. I got just 23 of the 32 first-round games, which would be a lot more impressive if not for the lack of first-round upsets. This means that pretty much everyone got more, including 60% of the ESPN Tourney Challenge brackets. There are almost two million people ahead of me in that competition, but my confidence is not shaken. Let’s see how we got to this point.

First round upsets:

9 Xavier over 8 BYU
9 Purdue over 8 Arizona
9 Michigan State over 8 Marquette
11 Winthrop over 6 Notre Dame
11 VCU over 6 Duke

And that’s it. Five upsets, so most people are probably in the high-20s in points. More people were probably picking VCU and Winthrop than they should have, and the other games are basically not upsets at all (9-seeds lead 8-seeds all-time).

My first-round upset picks, and the round I eventually had them reaching:

14 Oral Roberts over 3 Washington State (2nd round)
13 Holy Cross over 4 Southern Illinois (2nd round)
12 Illinois over 5 Virginia Tech (Sweet 16)
12 Arkansas over 5 USC (2nd round)
10 Georgia Tech over 7 UNLV (Elite 8)
10 Creighton over 7 Nevada (2nd round)
9 Purdue over 8 Arizona (2nd round)
9 Michigan State over 8 Marquette (2nd round)
9 Villanova over 8 Kentucky (2nd round)
9 Xavier over 8 BYU (2nd round)

Only three of these (Purdue, Xavier, and Mich. St.) actually happened, though several were close. Creighton, GT, and Illinois all lost in the final minute. That doesn’t make them wins on my bracket, though.

Other first-round losses for me, with the round I eventually had them reaching:

6 Notre Dame (Sweet 16)
6 Duke (Elite 8)

In a 1-2-3-4-5-6 scoring system, I have 23 points, but I can only get 97 of a possible 120 because of later-round losses.  I would also have 23 in a 1-2-4-8-16-32 system, with 165 possible out of 192.  Using ESPN’s system (10-20-40-80-120-160), I have 230, with 1200 possible.  That’s not very good, at least not right now.

Now, let’s take a look at the 32 remaining teams and their chances to win the tournament.  These numbers are based on a log5 analysis using Ken Pomeroy’s pythagorean win percentages.  KenPom has the numbers for the opening round on his blog. The teams below are ranked by probability of winning the championship.

There weren’t a lot of changes in the championship probability column, since none of the real favorites lost (as much as I would like to believe Duke was a favorite, I suppose they were not).  UNC still has the best odds to win it all, mainly because they’re the best-rated team on KenPom’s site.  My pick, Kansas, was a close second.  I also have A&M winning it all in several of my Tourney Challenge brackets, so we’ll see how that goes.

Now, go watch some second-round games.  Hopefully A&M can pull out a win against a semi-home Louisville crowd.