clock menu more-arrow no yes mobile

Filed under:

Betting against popular college football teams seems like a good idea. Does it work?

If a team has lots of fans, surely a lot of people bet on it, right? Seems Vegas is even too smart for this, though.

If you buy something from an SB Nation link, Vox Media may earn a commission. See our ethics statement.

Not if it's coached by one of these guys
Not if it's coached by one of these guys
Getty Images

In gambling circles, one of the tried-and-true strategies for making money over the long haul is to fade the public. This is a simple approach: Figure out which team the public is favoring (especially if there is a large disparity), and then go the other way.

This is especially effective if one team is getting a significant majority of betting tickets, but not of the total money bet, as this typically indicates the public is on one side, but the sharps are on another.

One way of fading the public might be to bet against popular teams. The assumption here is that there are certain teams in every sport that have large fan bases full of die-hard fans, who will bet on their teams while using emotion instead of logic.

So, find the most popular teams and then bet against them? This seems like a logical way to make money, on the assumption that betting sites will make their lines favor popular teams, to balance out the fact that they get a lot of dumb money on those teams.

Would this strategy work in college football? Are the most popular teams the worst against the spread? With the obligatory caveat that this is college football and the sample sizes are smaller, let’s look at the last 10 years of data.

Here is every team that is at least 10 games over .500 against the spread over the past decade, according to Phil Steele.

Ohio State 78-52-1 0.599
Oregon 78-52-2 0.598
Oklahoma State 72-50-4 0.587
Duke 67-51-3 0.566
Stanford 72-56-2 0.562
Oklahoma 73-57-3 0.56
Kansas State 69-54-1 0.56
Ole Miss 67-53-3 0.557
BYU 70-57-1 0.551
Florida 69-57-3 0.547
TCU 68-56-3 0.547
Wisconsin 70-58-2 0.546
Utah 67-57-2 0.54

The list contains a few we would expect, like Oregon, Oklahoma State, Stanford and Duke. They have had a lot of success, but don't have traditionally massive football fan bases.

And then there's that team on the top: Ohio State. The Buckeyes might have the largest fan base in college football, full of loyal supporters likely willing to plunk down bets on their team. The fact that Ohio State has the best record against the spread creates a couple possibilities:

  • Betting lines do not move in anticipation or response to betting on popular teams; or
  • Betting lines do price in the popular team dynamic, but this pricing has not been enough to keep Ohio State from covering 60 percent of the time.

Regardless, the betting sites have likely taken a bath on the Bucks in the past decade and, to a lesser extent, Oklahoma and Florida.

And the surprising aspect about Ohio State's record is that it was largely acquired under Jim Tressel, who was 42-21 against the spread in his last five years, despite a reputation of playing close to the vest and only doing just enough to win a game. That might have been true in the first part of his career in Columbus, but it wasn't accurate when he had Troy Smith and Terrelle Pryor.

So how about the teams that are at least 10 games under .500 against the spread?

Is this going to be full of nationally popular teams who don't cover because their fans bid up the betting lines by placing emotional bets?

Colorado 53-66-2 0.442
Cal 55-69 0.444
West Virginia 55-68-3 0.448
Illinois 54-66 0.45
Michigan 55-67-3 0.452
Texas A&M 56-68-3 0.453
LSU 57-69-4 0.454
Kentucky 53-64-4 0.455
Auburn 59-69 0.46
USC 60-70-1

0.462

We have to get down to the fifth spot to find an exceptionally popular team, and then we have several SEC teams. The two Power 5 teams with the worst records against the spread over the past decade are Pac-12 schools with lukewarm fan support. There may be some money to be made in betting against certain popular teams, but the worst teams against the spread fall outside of that category.

And finally, let's take a look at popular teams in the aggregate.

To get a rough approximation of the category of popular teams, I'll use the valuation from The Wall Street Journal. That seems more reliable that Nate Silver's ill-fated effort in 2011, which concluded (among other things) that Georgia's fan base is only two-thirds the size of Georgia Tech's, an amusing statement for anyone who has attended when Georgia plays "at" Georgia Tech.

Ohio State 78-52-1 0.599
Oklahoma 73-57-3 0.56
Florida 69-57-3 0.547
Alabama 71-62-2 0.533
Texas 59-65-4 0.477
Georgia 59-65-4 0.477
Notre Dame 58-66-4 0.469
Auburn 59-69 0.46
LSU 57-69-4 0.454
Michigan 55-67-3 0.452

Four of our potential public teams are over .500, while six are under. In the aggregate, these 10 have posted a .503 winning percentage against the spread over the last decade. Based on that, there are a couple potential conclusions:

  • There is no such thing as a long-term public team. If betting sites are trying to get even money on both sides of their lines so that they make money, then we would expect that the most popular teams would have a record under .500, because the sites would shade the lines to account for the flood of loyal/emotional/dumb fan money. This hasn't happened, as the against-the-spread results for these teams are collectively close to coin flips.
  • Betting sites don't care about moving their lines based on the bets from popular-team fan bases. The second possibility is that betting sites are not trying to get even money on both sides of the line because they are wiling to stake a position on a game, confident that they know more than the general public and make a killing when most of the money is on the losing side. If that's the case, then they would not move the line based on a large number of bets coming in for a popular team.
  • Either way, the evidence shows that basing a betting strategy on going against popular teams won't work. The concept makes sense in theory, but it isn't borne out in practice.