March Madness Visualization

This year's NCAA Men's Basketball Tournament has made history by something that has never happened before. For the first time in the Tournament's history, a 15 seed (Florida Gulf Coast) has made the Sweet 16.    It is generally a given that your 1 and 2 seeds will do well in the tournament, regardless of the fact that they are generally playing a 20+ win team.  The 64 teams in the first round of the tournament were awarded a bid by the NCAA committee as a result of their successful season.  Even though we often see teams like Florida Gulf Coast, or Harvard, who did not receive that much exposure throughout the season, why are they generally expected to be inferior to the top seeds?  To get a better idea of this, we collected data from allbrackets.com and created a motion chart to determine if there was any possible correlation between the amount of regular season wins a team (or seed in the tournament) and the amount of wins they have in the March Madness Tournament.  To accomplish this we first compared regular season wins to seeding. Does the amount of regular season wins affect the seeding a team is awarded?  We concluded that regular season wins do not necessarily result in a higher seeding, and attributed that to strength of schedule differences.  We then tried to determine if seedings had an affect on tournament wins by analyzing the motion chart and looking for patterns or similarities.  We also ruled out that, in general, the seeding does not reflect the consistent success in the tournament. Even though first and second seeds often see multiple combined wins in the tournament, they were not consistent enough to say with confidence how many wins to expect for each seed outside of 4 combined wins (By analyzing the 16 seeds, we see they have never won a march madness game, leading us to conclude that the first seeds will all win their opening round game.).

Visualization by Carter Astin and Russell Champion
Feel free to comment below