Last year I wrote two columns on how to use game theory to improve your chances of winning March Madness bracket (see Using Game Theory to Win the Office Pool and Using Game Theory to Win Your Upset Picks)
. Last week I updated it with my upset pick rankings for the tournament (see How to Gain the Betting Edge
). In the top third of my upset picks (1 to 22) seven favorites won versus 10 second favorites and six games in which a team other than the first or second favorite won.
In the second-third (picks 23 to 44) 13 favorites won, five second favorites and four other teams. In the last third, the least likely upsets, 20 favorites won, two second favorites and one other team.
Another way to look at it is in my first-third, the winning team won by an average of 10 points (and when favorites won it was only by an average of eight), in my second-third the average margin was 12, and in the last third, the games I thought would be the most lopsided, the average winning margin was 16.
Of course, it's not hard to make better-than-random predictions. A No. 9 seed is more likely to upset a No. 8 seed than a No. 16 seed to upset a No. 1 seed. How do we evaluate my predictions? One way is to compare them to Matt Davio, who won the Minyanville March Madness bracket challenge
Matt picked 42 favorites, and won 32 of those games. As discussed in previous columns however, these are not very important. He picked 11 second-favorites and won nine of them, an extraordinary job. The chances are less than 1 in 40,000 of doing that by chance. With my top 11 games to pick the second-favorite, you would have won five, lost two and there were four games won by a team other than the first of second favorite.
Matt would have done better to stop there, but he also picked nine teams that were neither first nor second favorites for the slot. I generally don't recommend that. He won only three of these and lost six. Combining those to 12 wins and eight losses among upset picks still leaves an impressive excess of four upset wins, which will be five if Michigan wins tonight. In my top 20 second-favorite picks you get an excess of only two, which is fine for a small pool but probably not what you need for a reasonably large pool (105 people submitted choices for the Minyanville pool).
While I cannot be sure of this, I suspect some of Matt's success came from attending the University of Michigan, and letting sentiment influence his choices. While picking Michigan over Kansas was the percentage move, putting them in the Final Four, much less winning the championship, was hard to justify on the regular season performance. But maybe he knows something I don't; that's often the case.
The Minyanville bracket challenge
, like most versions, does not give the prize to the person with the most correct picks. It gives extra points for correct choices in later rounds, and also for picking teams with higher seeds. But as I explained in previous columns, the most important thing to focus on is the number of excess upset wins, and making those upset picks different from the common upset picks. While other things like good basketball judgment and figuring the specific rules of your bracket pool matter, 80% of winning is the game theory, which takes no skill at all. Almost everything else you read about March Madness brackets concentrates on the other 20%.
As expected, this was a year with a lot of upsets. That doesn't have much effect on your strategy, unless you think it will encourage other people in the competition to make extra upset picks. However, since most people pick too few upsets (I recommend picking 28 for a pool of 105, but I double count upset picks that are not the second favorite; Matt Davio picked only 20 but I count it as 29, which is a shrewd number) picking extra is likely to help them.
Of the 67 games, including the first four and final four, 40 were won by the lowest seed that could occupy the slot (if two equal seeds meet, I call the one with higher RPI the favorite). That's low since favorites win about two thirds of the slots historically. It's not the lowest ever, there were only 38 in 1990, 1999 and 2010. Seventeen slots were won by second favorites, which is about average. Twelve games were won by a team other than the first or second favorite. This has been surpassed only in one year, 13 in 2011, and maybe two, 12 in 2000. However this year is unique in that 2000 and 2011 had more favorites winning, the deep upsets came at the expense of second favorites. 2013 is the first year in which we had both so few favorites winning, and so many teams with seeds higher than the second favorite.
However, even had you known in advance that there would be an excess of deep upsets (and people did, I mentioned this in my column before the tournament), it's still a dangerous strategy to pick them. Look at Matt Davio who got only three winners against six losers on deep upset picks (and the wins were all Michigan). The problem is that while there is only one favorite and one second favorite, there are between two and 66 other teams that could win a slot. So even if you know that neither the favorite nor the second favorite will win a slot, you could easily guess the wrong team to win.
Anyway, win or lose, I hope you played the game. More important, I hope you watched some games with friends, either in person or on television. There's one more game, and if it's as good as the semi-finals, it will be worth watching.
No positions in stocks mentioned.
The information on this website solely reflects the analysis of or opinion about the performance of securities and financial markets by the writers whose articles appear on the site. The views expressed by the writers are not necessarily the views of Minyanville Media, Inc. or members of its management. Nothing contained on the website is intended to constitute a recommendation or advice addressed to an individual investor or category of investors to purchase, sell or hold any security, or to take any action with respect to the prospective movement of the securities markets or to solicit the purchase or sale of any security. Any investment decisions must be made by the reader either individually or in consultation with his or her investment professional. Minyanville writers and staff may trade or hold positions in securities that are discussed in articles appearing on the website. Writers of articles are required to disclose whether they have a position in any stock or fund discussed in an article, but are not permitted to disclose the size or direction of the position. Nothing on this website is intended to solicit business of any kind for a writer's business or fund. Minyanville management and staff as well as contributing writers will not respond to emails or other communications requesting investment advice.