February 22, 2012 - by Gregory Matthews
[This guest post was conceived of and written by Gregory Matthews. Gregory received his Ph.D. in statistics from the University of Connecticut in 2011 and blogs at Stats In The Wild. His idols are Derek Kellogg and John Tukey. If you have an idea for a guest post that you think would fit right in here on our blog, please email us.]
With March Madness rapidly approaching, I’ve been poring over the the 2012 NCAA Tournament Bracket Odds, which provide, for each team, an estimated probability of reaching any given round of the tournament, based on recent bracket selection and seeding projections.
Providing this data for each team projected to make the tournament and for each round is certainly interesting, but it’s a lot of information for one brain to process. So I’ve decided to try to visualize the data and make it easier to consume all at once. To do this, I’ve made a so-called balloon plot:
(click to enlarge)
The balloon plot here consists of 68 columns, one for each team projected to make it into the 2012 NCAA tournament, and 6 rows, each representing a round of the tournament (2nd round, Sweet Sixteen, Elite Eight, Final Four, championship game, and champion), which form a grid.
In each of the rectangles within the grid, a circle represents the probability of the team in that column advancing to the round in that row. The larger the circle, the larger the probability that that team advances to the given round.
The columns are then grouped by conference, and the conferences are ordered by the number of teams that are projected to qualify for the tournament.
Within a conference, teams are ordered by projected seed, with the highest seed on the left and lowest seed on the right. I’ve grouped all of the one team conferences into one group and ordered them by projected seed in the same way of the other conferences.
Many interesting things jump out at me from this chart, especially:
Printed from TeamRankings.com - © 2005-2021 Team Rankings, LLC. All Rights Reserved.