How Do Markets Function? An Empirical Analysis of Gambling on the NFL

NFL Betting Trends
Steven Levitt, Prof University of Chicago
December 2002

One of the more interesting studies of wagering market efficiency comes from Steven Levitt, a noted and distinguished economics professor at the University of Chicago. His paper entitled, “How Do Markets Function? An Empirical Analysis of Gambling on the NFL”, looks at over 20,000 bets placed at an online sportsbook during the 2001-2002 NFL season. He records the betting habits of over 200 players in a “handicapping tournament”. This is one of the few studies that looks at actual bets placed. Note, however, that we feel some of his data is flawed. He only looks at bets within a “tournament”, and relies on info found on Wagerline, which is in itself flawed. We still find some interesting results.

Levitt makes the bold assumption that bookmakers are well aware of the market’s inefficiency, and they are happy to exploit it. Conventional wisdom has it that bookmakers try to balance their action. Their goal being to make money simply on the transaction cost, vig or juice. Levitt looks closely at bettor tendencies. He finds, not surprisingly that, the public tends to bet favorites regardless if they are home and away. Here is quick summary of his results.

NFL Bettor Tendencies in Tournament 2001-2002 (19,201 Bet)

# of Bets Win %
Home Favorite 6,741 49.1%
Visiting Underdog 5,270 50.4%
Visiting Favorite 4,904 47.8%
Home Underdog 2,286 57.7%

Levitt does admit that his study focuses on only one season. He goes on further to show that the win percentages he observes are consistent with the last 20 years of NFL games results ATS. The paper also suggests reasons for this bias. His overall position is that bookmakers fully understand this bias and create a betting line that exploits it. has long advocated a strategy that utilizes this bias called “Bet Against the Public.”

This paper is an excellent introduction to the stats and philosophy developed by, which suggests betting against the public on lopsided games. Unlike Levitt’s study’s stats come directly from the line boards of multiple online sportsbooks. Their betting percentages represent actual bets placed at sportsbooks.

“How Do Markets Function? An Empirical Analysis of Gambling on the NFL”

Published: Levitt, Steven D. “Why Are Gambling Markets Organized So Differently From Financial Markets?,” Economic Journal, 2004, v114(495,Apr), 223-246.