Does the Play-In Game Create Parity?

From 1985 to 2011, two seeds had a 104-4 record against 15 seeds in NCAA tournament games. In 2012 and 2013, they were just 5-3. 

Last year, every seed from 11 to 15 had at least one tournament win, and on the first day of the 2014 tournament, Dayton, Harvard, and North Dakota State posted upset wins. 

Everything from scholarship limits to television exposure to crooked television executives has been floated as a possible reason for parity. But ask any gymnast what’s causing it, and she’ll tell you the cause:  The play-in games.

In gymnastics, the low score for a routine is discarded (along with the high score, but that doesn’t come into play in basketball talk). That’s essentially what the play-in game does. The four worst teams in the field take up just two lines in the 64-team bracket. In other words, two of the worst teams are “thrown out”. 

Take this year’s tournament: Albany, Mt. St. Mary’s, Texas Southern and Cal Poly played early in the week as co-16 seeds. If there were no play-in games, those four teams would have been the 16 seeds in the four regions. Instead, Coastal Carolina and Weber State join Albany and Cal Poly as the 16’s. 

Why does that matter? Because instead of playing Texas Southern, ranked 239 in the RPI, a one-seed has to play Coastal Carolina, who is 40 spots higher. Replacing Mount St. Mary’s with Weber State means a 39 spot jump in RPI. 

The impact on the 2-15 game is even larger. Without a play-in, Coastal and Weber would be 15. Instead, Eastern Kentucky and American (who would have been 14s without a play in) drop to the 15 line. That means that the two seeds are playing teams 81 RPI spots and 54 RPI spots better, all because of the play-in games. 

That’s held up over time. From 2001 to 2010, 28 of the 40 15 seeds (70%) were 100 RPI or worse. Since the second play-in was added, only six of 12 (50%) have been that bad.

A look at the history of tournament upsets supports the idea that the play-in games give high seeds tougher early games. In the first 16 years of the 64-team field, only three 15 seeds won games. In 2001, the first year of a 65-team field–and a single play-in–Hampton took out two seed Iowa State. The second play-in game was added in 2011, and the three upsets in two years started the following season. 

If there had been no play-in game from 2001 to 2010, 16 seeds would have had an average RPI of 175. Instead, it was 151, meaning the play-in game elminating the “low score” gave top seeds an opponent 24 RPI spots better. Since the second play-in was added, the average 16 seed improved from 198 to 143–55 spots better.

The single play-in game improved the average 15 seed seven RPI spots. Two play-ins improved it 18 spots.