The "39th" ranked team (I know, I cheated...I'm using the "others receiving votes category...) at the end of the 2007 season had 9 wins. For the 2004 season, the "38th" ranked team, had 8 wins. Weis wrapped it up with 3 wins, Willingham finished up with 6. I'm not sure if that satisfies the criterion or not given that I'm not going to find and crunch the numbers to be rigorous, but it definitely appears to be a statistically significant underperformance. Perhaps the 2 standard deviation suggestion is too loose; maybe that would have to be adjusted. I don't know, but the point is that I'd like to see coaches hit in the wallet for statistically significant underperformance. The golden parachute thing is bullshit.