Point Spreads Do Affect Outcomes of Sporting Events

Sport participants are motivated by lots of different things that interact in complicated and often unconscious ways.

Sports betting impacts outcomes of games.

Photographer: Ethan Miller

The University of California college football team trailed the University of Washington 38 to 7 this past weekend with one second left on the clock and facing a fourth-down-and-twenty-one. When there's no chance of winning, coaches usually either “take a knee,” 1 give seldom-used players a chance to gain some experience in real-life game conditions, or fight to the end by calling long-shot touchdown attempts.

California opted for the unusual choice of a field goal attempt. Washington was favored to triumph by 28.5 points. A field goal would be good for three points and reduce the margin to 28, so bettors wagering on California would win. Sports announcers are discouraged from mentioning gambling, so they often use coded expressions such as “this is meaningful to some” 2 and “people in the desert 3  are watching.” At 2 a.m. Sunday on the East Coast, with an outcome forgone by midnight, the remaining television audience was likely made up of bettors who understand the code.

Odd late game decisions that don't affect the outcome but change which bettors get paid are common in sports. To some people this proves that betting affects games, while other people laugh that nuts will build conspiracy theories on any unusual event. Academic economists have opinions. The 2005 best-seller "Freakonomicsalleged match fixing in Sumo wrestling and a classic 2006 paper by Justin Wolfers claimed that point-shaving was widespread in college basketball.

These and similar conclusions have been attacked and defended vigorously. It's clear that betting affects game outcomes because the statistical distribution of results is not consistent with everyone ignoring the spread. What's controversial among academics is whether the deviations can be explained by corrupt actions such as players shaving points for money or coaches trying to please local bettors.

No one has come up with a plausible story of rational actors profiting from systematic fixing. Plenty of people have been caught fixing matches, 4  but their capers are filled with confusion, double crosses and idiotic carelessness. Are there only occasional amateurish attempts to fix, or is fixing endemic and we catch only the most foolish attempts?

Bookmakers don't fix matches. Prior to World War II, bookmaking was a local business, and fixes were attempts to cheat bookmakers. 5 Then Charles McNeil 6  invented the point spread and organized crime took over gambling. From about 1945 to 1980 U.S. betting was centralized. The mob made a low-risk profit by balancing bets. It made the same profit whichever team won. It discouraged 7 attempts at fixing as that kind of fraud could attract investigations and hurt business. Since 1980, the betting industry has fragmented into retail operations that try to balance books, risk-taking bookmakers and syndicates, and assorted other players, none of whom have much incentive to fix. It's amateurs who have been caught fixing games.

Like kidnapping, the hard part of match-fixing is not the crime, but collecting the money. Retail betting sites protect themselves by cutting off winners. At matching sites like BetFair, you must win computer battles to get significant betting action at normal odds. Wholesale betting sites will continue to do business with winners, but they learn a lot about them, and also about suspicious matches. 8

Professional bettors typically rely on getting thousands of wagers per year at 5 percent to 10 percent edges. 9  Having a dozen or so bets with 20 percent edges from fixes would be nice, but not worth the expense or risk of fixing. And trying for hundreds of fixes or sure-thing fixes is how people get caught.

People in sports know the spread. Many of them bet despite rules against such activities, and nearly all know people who bet. A college coach who knows that most of his or her school's boosters have bet on his or her team might have a hard time making a decision that is meaningless in game terms but costs them money. 10 Knowledge of crowd desires by referees 11  creates large home-team advantages. Knowledge of the spread should have similar effects on everyone. That's why academic papers show betting affects games, but this does not mean large numbers of people are acting corruptly.

Sports are played by people -- not angels -- motivated by lots of things: the rules, team loyalty, personal benefit, friends' benefits, opinions of others. These and many other things interact in complicated and often unconscious ways. Individual actions interact for more complexity. You can rarely trace a specific game outcome to a specific motivation of a specific individual. But by averaging over thousands of games, you can demonstrate that point spreads affect outcomes.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
  1. Conceding gracefully and ensuring no one gets hurt in a meaningless play.

  2. The bettors.

  3. i.e. Las Vegas.

  4. And there are plenty more clumsy cover-ups because too many sports officials value avoiding scandal and protecting stars more than honest competition.

  5. Most famously, the “Black Sox” scandal when Arnold Rothstein fixed the 1919 World Series.

  6. The University of Chicago mathematician, and John F. Kennedy's elementary school math teacher.

  7. Severely.

  8. Generally, betting businesses have no interest in handing out net money to any customer. The exception is risk-taking organizations that are willing to pay winning bettors in order to get the information from their bets. But the amount they are willing to pay is limited to a fraction of the value of the information. Improving your winning percentage isn't useful if your constraint is the total amount people are willing to let you win. Moreover fixing is easiest for obscure and one-sided matches, for which there isn't much betting interest.

  9. See, for example, This is for big money lines like NFL football. Professionals can get larger edges for smaller bets in less popular areas; with larger edges fewer bets are needed for a stable living.

  10. Or, they might make that decision, to err on the other side. But either way, the spread affects the decision.

  11. Probably mostly unconscious.

To contact the author of this story:
Aaron Brown at

To contact the editor responsible for this story:
Robert Burgess at

Before it's here, it's on the Bloomberg Terminal.