NCAAB March Madness 2012

Filling out your March Madness brackets?  Please check out Sports Insights’ take on the popular NCAA bracket pools.  We will use historical results, probabilities — and apply several Sports Insights’ tools and models, including a contrarian approach,   implied probabilities, stochastic models, and game theory.  While Sports Insights typically focuses on value relative to point spreads in NCAAB (college basketball), the same approaches can be used for moneyline odds — which has practical applications for bracket pools.  

The Sports Insights philosophy of applying contrarian approaches to sports betting was presented as a paper to the 2012 MIT Sloan Sports Analytics Conference.  The information on this site is for entertainment and educational purposes only. Use of this information in violation of any federal, state, or local laws is prohibited.

Performance of Seeds in Round 1 of the Tournament

The chart below summarizes the results of the each numbered seed during Round 1 of the NCAA Tournament since 1985, when the tournament went to the 64-team format.  For the purposes of this article, we base the statistics on the 64-team field, notwithstanding the move towards additional teams in the draw (first to 65 teams, and now, to 68 teams).

Since 1985, no number 16 seed has ever beaten a number 1 seed in Round 1 of the tournament.  Number 15 seeds have gone 4-104 in Round 1 versus number 2 seeds, winning just 3.7% of the games.  Number 15 and number 16 seeds are very large underdogs to win their games outright — and may not be good picks for your bracket pool — but sports bettors may want to keep an eye on these seeds to beat the spread.

Table 1, below, along with Sports Insights’ live betting odds — should help you with your Round 1 selections of the tournament.

Table 1: Performance of Seeds in Round 1 of NCAA Men’s Tournaments

Seed Wins Losses Win %
1 108 0 100.0%
2 104 4 96.3%
3 92 16 85.2%
4 86 22 79.6%
5 72 36 66.7%
6 72 36 66.7%
7 65 43 60.2%
8 51 57 47.2%
9 57 51 52.8%
10 43 65 39.8%
11 36 72 33.3%
12 36 72 33.3%
13 22 86 20.4%
14 16 92 14.8%
15 4 104 3.7%
16 0 108 0%

Odds of Making it to the Final Four: A Statistical Markov Model

A large majority of March Madness pools have participants select their picks throughout the entire tournament.  The above information (Table 1) is specific to the first round.  To help you perform well in your NCAA Bracket Pool, our quantitative analysts wanted to take things a step further, so we took a look at which teams had the highest probability of advancing far into the tournament.

Based on point spreads, Sports Insights’ proprietary betting percentages, and other quantitative measures, our team developed a probabilistic Markov model to look at which teams have the highest probability of advancing to the Final Four.  Below are some of the standouts, sorted by probability.  In addition to a team’s actual strength, the team’s seed (and probable “draw”) plays a larger role than one would think.  For instance, the top two seeds play the 15th and 16th seeds in each region, and weaker teams, on average (than the rest of the field) — during each successive round.  Table 2 might be handy in helping you select teams to advance far in your brackets.

Table 2: Probability of Winning Region

Seed Team Region Final Four Prob
1 Kentucky S 42.4%
2 Ohio State E 41.4%
1 UNC MW 36.2%
2 Kansas MW 26.9%
1 Mich. State W 22.1%
2 Missouri W 17.7%
1 Syracuse E 15.7%
4 Wisconsin E 14.8%
8 Memphis W 10.4%
5 Wich. State S 9.6%
2 Duke S 8.2%


Dark Horses and Value

Based on an analysis of the brackets and potential roadblocks to the Final Four, these teams have the ability to win it all — and may be overlooked by others:

  • Ohio State – is just a #2 seed, but our model favors them to win the East Regional.
  • Kansas – another #2 seed, is almost even with #1 seed UNC to win the Midwest Region, partly because their potential first and second round opponents look to be as weak as North Carolina’s opponents.
  • Wisconsin – is just a #4 seed, but our model has them almost even with #1 seed Syracuse to win the East Regional.

Below is a list of teams that appear to be undervalued relative to their seeds, and may be able to make it through at least several rounds in the tournament. These dark horse teams might enable you to jump ahead of others early in the tournament:

  • Memphis – is only a #8 seed but made our list of favorites to win a region.
  • Wichita State is also a dark horse that might be able to grab you “value” in your bracket pool.
  • Indiana – is just a #4 seed, but is a force to be reckoned with.  Kentucky does not have a very easy road with potential of meeting Indiana or Wichita State in Round 3 of the NCAA Tournament.  These teams will be at least be as tough for UK — as Duke.
  • Although not on our list, New Mexico is another potential dark horse to make some noise, as a #5 seed.

Game Theory

There are several ways to perform well in a March Madness pool.  Some sports fans like going with entries that have the highest probability of winning each selection.  Other pool participants go for huge underdogs to advance far in the Tourney.  Your strategy may depend on the expected number of entries in your particular pool.  And, luck, of course, plays a large role — but there are ways to separate your bracket entry from others.

We believe that applying concepts of game theory and contrarian approaches — can help give you an edge, not only in your bracket pool, but also in betting against the spread.  You will want to separate yourself from the crowd and “follow the road less-travelled.”  For instance, although Kentucky has the highest probability of reaching the Final Four (Table 2), we did not highlight them in our write-up, because Kentucky is considered to be the consensus favorite entering the Big Dance.  We hope that the tables, probabilities, and statistics in this article will help you “pick your spots” — and do well in your office pool.  Let the Madness begin!

For Media Only: additional information, statistics, high-resolution screen shots, and interviews available –  please Contact Us.

 We do not guarantee that the trends and biases we’ve found will continue to exist. It is impossible to predict the future. Any serious academic research in the field of “market efficiency” recognizes that inefficiencies may disappear over time. Once inefficiencies are discovered, it is only a matter of time before the market corrects itself. We do not guarantee our data
is error-free. However, we’ve tried our best to make sure every score and percentage is correct.