NCAAF Betting Trends: Fading AP Top 25 Teams After a Loss

NCAAF Betting Trends: Fading AP Top 25 Teams After a Loss

Last season we published an article that detailed how sharp bettors could take advantage of the Associated Press (AP) Top 25 rankings. Square bettors will frequently put too much stock into these rankings despite the fact that many voters are not watching every game and are actually far less informed than the oddsmakers. With this knowledge in mind, our determination was to find other profitable betting systems that capitalized on the artificial value creating by these AP rankings.

Although most teams are actually undervalued following a loss, I thought that this trend would flip-flop when dealing with teams ranked in the Top 25. On many occasions I have observed a ranked team falling in devastating fashion, only to see AP voters react minimally by dropping them just a few spots in their subsequent rankings.

Using our Bet Labs software, we sought to determine whether we could find any historical value by fading (betting against) ranked teams following a loss. Our first step was to utilize the “Rank” filter to focus on teams ranked between 1 and 25. Our second step was to use the “Previous Game Margin” filter to focus on team’s who had lost their previous game (margin between -1 and -47). We then checked off the “Fade System” box to make sure we were betting against these teams. My original hypothesis was confirmed by the results as fading ranked teams had gone 221-189 ATS for +20 units and a 4.9% return on investment (ROI).

While these results were encouraging, we knew that with such a large sample size on a basic system it would be easy to improve our expected return on investment. We’ve previously discovered that teams are undervalued following a blowout loss in the NBA and NFL, and wanted to use that information to improve our results.

The table below displays the performance of ranked teams following a loss as we gradually increase the margin of defeat.

Previous Game MarginRecordWinning PercentageUnits EarnedROI
-1 or more221-18953.9%+20.004.9%
-10 or more110-8157.6%+22.9812.0%
-21 or more53-3957.6%+11.3012.3%
-28 or more29-1369.0%+14.5334.6%

Although the highest return on investment comes when we look at ranked teams following a loss of at least 28 points, the sample size is so small (42) that it will not produce many matches this season.

The optimal range would appear to be fading ranked teams following a double digit loss as teams fitting that criteria have gone 110-81 ATS (+22.98, 12% ROI) since 2005. That includes a 33-14 record (+17.15 units) over the past two seasons and an 81-49 mark (+27.81 units) since the start of the 2008 season.

We’ve recently introduced a number of new filters to Bet Labs, including many dealing with the performance of teams in the previous year. By utilizing the “Previous Season Made Postseason” filter we’re able to focus solely on teams you played in a Bowl Game during the previous season.

This simple addition improves our system to 95-65 (59.4%) for +24.79 units won and a 15.5% ROI:

NCAAF Fade System

It’s interesting to note that the best ranked teams to fade following a loss have been: California (7-1), Michigan State (6-1), Texas Tech (6-1) and Alabama (8-3).

Bettors can view this system within our Think Tank, and can view up to the minute odds and trends on our new and improved free College Football odds page.

Looking to create your own data-driven betting systems this football season? We offer a 6-day trial of our unlimited Bet Labs subscription for $50 and or you can purchased the unlimited monthly subscription for just $199 — a 33% discount off our previous rates.

11/28/14 Update: This system has gone 8-7 ATS during the 2014 season boosting the all-time record to 103-72 (+25.42 units, 15% ROI).

[adrotate group=”3″]

David Solar

David was the Content Manager at Sports Insights. He has since moved on to greener pastures.

  • Keith
    09/05/2014 at 7:38 pm

    Hello David,

    I noticed during this article you referenced the AP top 25 poll.

    On Aug. 26 you wrote an article referencing the coaches poll. Which poll is the one used for your bet labs ‘rank’ filter?

    Thanks, and thank you guys for the work you do.

  • Jeff
    09/12/2014 at 4:09 pm

    Thanks for the article. Utilizing this method, the play for week 3 would be Kent St? Is this correct?

    I noticed that you mentioned that the plays are 1-1 this year….I know that East Carolina was the winner, I could not determine who the losing play was.

Post a Comment