To Bowl or Not to Bowl?

To bowl or not to bowl, that is the question;
Whether 'tis nobler in the mind to suffer
The indignity of a losing season,
Or perchance a victory many miles away?
And thus delivering 7 wins to thee.

My humblest apologies to William 'Bill' Shakespeare.

As a resident of Columbia, South Carolina, I have had the opportunity to witness the weeping and gnashing of teeth that has accompanied the Gamecocks' 5-game skid to end their once-promising season. A bowl bid that once seemed certain, now has only a snowball's chance of coming to fruition. Around talk radio and the water cooler, the sentiment is that the Gamecocks need a bowl bid in order to improve heading into next season. Conventional wisdom says that the extra practices and extra game will give the Gamecocks a better chance at improving in the 2008 season. Being the inquisitive type, I wanted to find out if this is indeed true. I looked at every BCS conference team that finished the regular season with either 5 or 6 wins from 2000-2005. I divided those teams into two categories: 6-win teams that went bowling, and 5/6-win teams that stayed home. I think its safe to assume that BCS teams that finish with either 5 or 6 wins are pretty even (not outstanding and not terrible), whether or not they go bowling. The two tables below compare how those teams fared the following season.


As you can see, both sets of teams improved marginally the following season. The teams that participated in bowl games had a better overall record. However, the difference in winning percentage equates to less than half a win over the course of an 11 or 12-game regular season. In terms of improvement, the non-bowl teams fared better. Their winning percentage improved by .065 and the bowl teams' winning percentage improved by .024. Still, as before, the difference over the course of an 11 or 12-game season is less than half a win.

After compiling those numbers, I decided to further test the notion that bowl games aid improvement by running two sets of regression analysis. Both sets included the same sample of 5 and 6 win regular season BCS teams from 2000-2005. They were coded either as '0' if they didn't play in a bowl game and '1' if they did. I ran the regression to see how well playing in a bowl correlated with 1) the next season's winning percentage and 2) whether they played in a bowl game the next year.

Both those R squared values are about as close to zero as you can get. Bowl game participation, in and of itself, has basically no correlation with success the following year. The 6-year window I observed has examples of 6-win bowl teams that improved substantially (Southern Cal from 2001 to 2002), 6-win bowl teams that imploded (West Virginia from 2000 to 2001), 6-win non-bowl teams that improved substantially (Syracuse 2000 to 2001), 6-win non-bowl teams that imploded (Washington 2003) and everything in between. So Gamecock fans, don't worry about a possible bowl snub. While the pageantry and trip may make memories for players, fans, and coaches, the effect it has on the next season's success or failure is negligible at best. Gamecock fans can take heart in the case of the Kansas Jayhawks, a team not included in this study because their 6-6 bowl-less season occurred in 2006. The Jayhawks you may have heard, are currently sitting at 11-1 and closing in on a BCS bowl bid.

All the world 's a stage, and all the men and women merely players. They have their exits and their entrances; And one man in his time plays many parts.