I am the greatest college football totals handicapper ever is the title of this post, but this article is not so much about college football totals as it is about adjusting your college football handicapping methods from year to year (or month to month if necessary).
If there is one thing I have learned, it is you can’t stand pat on the same old handicapping methods every year because college football changes so much from season to season, maybe more than any other major team sport.
Unfortunately I had to learn this lesson the hard way. From 2004 thru 2009 I had a fantastic record betting on college football totals. I may not have been the greatest college football totals handicapper ever, but I thought I was pretty darn close.
2005: 35-21 62.50%
2006: 41-29 58.57%
2007: 49-38 56.32%
2008: 80-48 62.50%
2009: 101-61 62.35%
Total: 356-236 60.14%
For you math wizards out there, please correct these numbers if I am wrong, but I believe my Z score was 3.7776.
Approximate results in Excel:
Z =(356/1.1-236)/SQRT(592*10/11) ≈ 3.7776
p(z < Z) = NORMSDIST(3.7776) ≈ 99.99208%
Exact results in Excel using the binomial distribution:
P = BINOMDIST(356-1,592,52.381%,1) ≈ 99.9912%
Which means the odds of me being on a 6-year, 600 game sample size college football totals lucky streak were pretty much 0%. I knew what I was doing when it came to betting on college football totals, and I was on a roll.
Sometime in early 2004 I had read an article about how the public loves to bet on the over in the NFL, yet overall the games went under around 51.5% of the time (I never confirmed this number, but I thought the premise that the public preferred to wager on the over was sound).
My theory was that if the NFL betting public preferred to wager on the over, then that was probably also true when betting on college football totals. And starting at 51.5%, you’re already less than 1% from break even, so a few tweaks here and there and I should be able to produce a profit.
My initial formulas were pretty simplistic. I concentrated on teams that liked to run the football a lot and played good defense.
Before the 2008 season I added some running and passing efficiency numbers to my overall college football totals formula. My confidence was soaring and as the above table indicated, I really started to ramp up the number of college football totals I was wagering each year. The next two years produced an overall record of 181-109 = 62.41%
Another thing I had noticed was that in the first half of the season I was hitting nearly 66% of my college football totals. As the season progressed, my win percentage slowly dropped into the mid-50 percentile.
I went into the 2010 season with one thing in mind; find as many college football totals as possible in the first 8 weeks of the season to wager on. Any game that fit into my proven college football totals formula was to be considered. And that is exactly what I did, with devastating results:
Week # 1: 3-10 (23.08%)
How was that possible? I was using the exact same college football totals formula that had been producing winners for the past 6 years. Not only that, the most amazing thing was that I had beaten the closing line in all but 2 games, and some closing numbers I was destroying.
Pitt-Utah U 50.5 47.5 L
FAU – UAB U 55 48.5 L
Purdue – ND U 55 53.5 W
NM – OR U 59 55 L
UCLA – KST U 45 44.5 L
SMU – TT U 62.5 57 W
NW – Vandy U 45 41 W
BG – Troy U 56.5 53.5 L
Mem – MSST U 52.5 47 L
USC – Hawaii U 55 50.5 L
WSU – OKST U 50 50.5 L
AZ – Toledo O 60 58.5 L
Cin – Fresno St O 56 57 L
Week # 2: 2-5 (28.57%) And I again beat the closing line in 5 of 7 games.
To make a long, and painful, story short:
2010: 54-60 47.37%
And those numbers were made more “respectable” after going 11-4 on college football totals in the bowl season.
What had happened? Had the game changed and I simply hadn’t adjusted? Was it just a freak season of bad variance? After all I continued to beat the closing line at nearly a 65% clip.
To tell you the truth I don’t really know what happened. Last year (2011), I made some minor adjustments to my college football totals formula and ended the season 62-56 (52.54%), barely above the 52.38% break even point.
I cannot say for sure what has happened over the past 2 seasons. Both the 2010 & 2011 college football season produced more overall unders than overs, though not at the 51.5% clip I had estimated back in 2004.
50.6% games went under in 2010** **
Based on this article I read, AN UNDER BIAS IN THE FOOTBALL BETTING MARKET: FACT OR FICTION?: A NOTE, these numbers are more accurate than the 51.5% I had originally read about.
I was still beating the closing line on a consistent basis, but my results were nowhere near what they had been during that 6 year run.
What I do know is that after the 2009 season I stopped tweaking and adjusting my formula because I thought I already knew how to consistently beat college football totals. And it was at that point that I started to lose my edge.
So bottom line, no matter how much success you have had in the past with a formula, situation, trend or angle, don’t stop working on ways to improve your handicapping or you may end up losing your edge. College football is constantly changing and as handicappers we also need to change and adjust our handicapping methods in order to stay ahead of the curve.
Does anyone have any similar handicapping stories they can share with us? I believe we all would appreciate learning something new. Thanks – Pez.