Those who follow college football year-to-year know that bowl season historically represents one of the best periods for backing the underdogs across the board. Since 1997, bowl underdogs have gone 148-106 against the spread, for a very respectable 57% winning percentage. A $100 wager on every underdog the past 10 years would have put over 3 grand in your pocket, and would have turned a profit every season except 1997 and 2003. There has historically been about a 2 point bias towards the favorite in bowl games, and it has shown no sign of disappearing in recent years.
Underdog overperformance is not just reflected in against-the-spread results - underdogs in bowl games have won 42% of their games outright over the past decade, resulting in even greater profits to those who have backed them straight up. A $100 wager on every bowl underdog on the moneyline would have returned over $5000 in the past decade. Throw out 1997, when underdogs went a brutal 1-19 outright, and this simple strategy gave you 45% winners and about $7000 in extra holiday cash in the past 9 years.
While the historical results are impressive, "past performance is not necessarily indicative of future results" is a mantra worth following when it comes to sports handicapping. There are three caveats to withdrawing your 401(k) and splitting it amongst the underdogs this bowl season. First, this underdog bias is fairly well known in the football gambling community. As with any situational angle, oddsmakers and sharp bettors can be expected to eventually catch onto this trend, causing lines to adjust to compensate for the situation. In fact, a 3-year moving average of the return from this angle has shown a fairly strong downward trend since 2000 including a losing season last year. Only time will tell if that means this angle is beginning to lose (or has already lost) its usefulness.
A second issue, fairly common in football handicapping, is the limited sample size. While the historical results seem solid on the surface, they could be a result of random chance. Although we are looking at 10 years of data, that only represents just over 250 games. With such a limited amount of data, it is hard to say with confidence whether these results are significant and predictive.
A final caveat to consider for the 2007-08 bowl season is the performance of underdogs during this past regular season. 2007 NCAA football was marked by an apparent increase in parity, as highly-regarded teams seem to fall to inferior opponents week after week, and every Sunday brought a completely reshuffled Top 25. The public at-large knows that this has been an "any given Saturday" type of season, and that will likely be represented to some degree in the betting lines, quite possibly deflating these lines below historical levels and leading to a weakening of this angle's profitablility.
Past results show that a successful regular season for underdogs generally results in a sub-par bowl showing. In the past decade, the five worst regular season for underdogs against the spread represented 5 of the 6 best seasons in that span to back the bowl underdog on the moneyline. The 5 years in which underdogs outperformed in the regular season actually showed a loss in this betting system. This trend also holds for against-the-spread results, with underdogs in "outperforming" seasons actually ending up less than 50% ATS in the subsequent bowl games. An analysis of average margin-of-victory gives the same results. This predicts a below-average year for underdogs in the 2007-08 bowls.
While 2007 was good for underdogs ATS, it was the 3rd worst season in the past 10 years to bet underdogs straight up (on the moneyline). This is another negative indicator, as years in which the underdogs won outright at a rate less than the 10-year average shaved about 10 points off their outright bowl wins percentage. While this subset still returned a historical profit, it is another reason for some caution when backing the underdogs this year.
Notably, the 2007 season was the second best year for across-the-board underdogs in the last decade. Using the available (limited) data, this suggests that the 2007-2008 underdogs will underperform in the bowl games, both straight up and against the spread. This is not surprising, as bowl season tends to favor "contrarian" handicapping which goes against public opinion. With the recent strong performance of underdogs (particularly against big "public" favorites), we can expect the value side of the bowl lines might lie with the favorites this year. Overall, the data at hand predicts an unusually tepid bowl season for underdog bettors. While it is difficult to bypass an angle holding such a great recent track record, those who have followed this very profitable moneyline strategy in the past may want to cut back a little this year and err on the side of caution in what looks to be a down season.
If you would like to make or read comments about this article, you may do so by visiting the Mess Hall forum at MajorWager.com where a thread has been started. Please click HERE