Securities Market Efficiency and the Reigning Super Bowl Champions.

Inefficient prices in the securities market can be challenged because a precise cutoff date for measuring returns is impossible to establish. No such ambiguity exists when betting on football games since holding-periods end upon completion of the game. When wagering against the reigning NFL Super Bowl champions from 1987-97, a wins-to-bets ratio was profitable and a sign that bettors inflate point spreads to a level of inefficiency. Thus market participants over react to important events on and off Wall Street.

The vulnerability of stock prices has long intrigued investors and researchers. Beating the market has an inescapable appeal. The overwhelming evidence that regular above average returns are denied to all but those with inside information has not slowed efforts to find market errors or tap into profitable trends. One reason for hope is that past studies have never truly resolved how long securities must be held before a particular trading strategy can be measured. Is a six-month holding period sufficient? Is a one-year wait better? Is a two-year period too long?

Pankoff has proposed that the market for bets on National Football League games can serve as a proxy for the securities market [5]. Because betting outcomes are usually known within one to seven days after wagers are placed, there is no question about an appropriate holding period. Pankoff drew an analogy between investing and betting when he noted that bettors are no less numerous, knowledgeable, competitive or profit maximizing than investors. Like the price of a stock, the points that Las Vegas odds makers award a game's weaker team discount all available information and reduce predictions of future outcomes to random chance.

Recent studies have used Pankoffs analogy to test whether streaking teams, clubs winning or losing successive games against the point spread, constitute an anomaly to the efficient market hypothesis. Camerer argued that bettors bid up point spreads on streaking teams in the National Basketball Association beyond what their true abilities warrant [1]. He proved that betting against hot clubs to take advantage of inflated spreads produced higher wins-to-bets ratios than betting on them. However, the profits from his contrarian strategy were too slim to challenge market efficiency. For example, if the New York Knicks win three consecutive games by eight points each but were favored to win only by five points, Camerer would expect bettors to regard the Knicks as a hot team and over estimate their winning margin in the next game. By betting against the Knicks in the fourth game, Camerer would, in a sense, be going long on an undervalued asset (the Knicks' opponent) or shorting an overvalued one (the Knicks).

Brown and Sauer confirmed Camerer's hypothesis that streaking teams cause otherwise rational bettors to overreact and even identified the margins by which they overreact 0.25 point and 0.67 point for clubs with two-game and four-game point-wise streaks, respectively [2]. While Brown and Sauer were no more successful than Camerer in turning overstated spreads into abnormal profits, both studies revealed that changes in speculative prices could be traced in part to the misperceptions due to over reacting by market participants.

Gray and Gray applied Camerer's contrarian rule for streaking teams to the NFL. They reported that the strategy of betting on NFL clubs that (1) were playing at home, (2) were rated as underdogs and (3) had performed poorly against the spread in four previous games produced non-random profitable wins-to-bets ratios [31. They discovered that the market tended to undervalue home underdogs. By combining that error with Camerer's insight that point spreads over react to streaking teams, Gray and Gray exposed a vulnerable market. If the football-betting market over reacts to streaking teams and yields above average returns, does it also over react to a seemingly more impressive achievement -- winning the Super Bowl -- and bare more vulnerability?

Study Methodology

Do bettors over estimate the performance of the reigning Super Bowl champion and thus inflate point spreads to an exploitable extent? By wagering on the Super Bowl winners' opponents, wins-to-bets ratios might experience an upward bias. If the New York Giants win the Super Bowl in year t, then for some span of games in year t+1 the market will yield abnormal profit opportunities for bettors who wager against the Giants. Because the market will eventually correct any over reacting behavior, the most vulnerable time should be early in season t+1 Hence, the rule for beating the market is to bet against the previous Super Bowl champions in the first three to five games of the current season.

The contrarian strategy is applied to the 1987-97 NFL seasons, and the results are subjected to the test of profitability by Tryfos et al. [6]. Lawrence was used to access the Las Vegas point spreads and game outcomes [4].

Study Results

The wins-to-bets-ratios for reigning Super Bowl winners for the years 1987-97 are shown in Exhibit 1. The 33% wins-to-bets ratio in the first five games of the 1987-97 NFL seasons by the Super Bowl champion from the preceding season strongly suggests that the strategy of betting against such teams during the early part of the season is very profitable. AZ-value of -3.11 is significant at p [less than].001.

Z = W/B-1.1(L/B) / [{1/B[(W/B+1.21(L/B))

-[(W/B-1.1(L/B)).sup.2]]}.sup..5]

where: B = total number of bets

W = number of winning bets

L = number of losing bets.

Z = 18/55-1.1(37/55) / {1/55 [(18/55+1.21(37/55))

-[(18/55-1.1(37/55)).sup.2]]}.sup..5]

Z = -3.11

The results support Camerer's assertion that betting against streaking teams is more successful (if not profitable) than betting on them.

Conclusion

It can be argued that an adoring public accords Super Bowl champions special treatment. Television executives scramble to schedule appearances for last season's Super Bowl winners. Team members are awarded bigger contracts and lucrative endorsements. Therefore, it would be surprising if bettors are completely rational when analyzing games where the reigning champion is competing.

Although bettors who remain objective may earn abnormal profits by betting against the recently crowned Super Bowl champion, a more significant implication is that participants in well functioning markets exhibit irrational behavior in response to important or unexpected events. Even more fundamental, market participants misperceive what is important or unexpected.

References

(1.) Camerer, C. "Does the Basketball Market Believe in the Hot Hand?" American Economic Review, Dec. 1989, 1257-1261.

(2.) Brown, W. and R. Sauer. "Does the Basketball Market Believe in the Hot Hand? Comment" American Economic Review, Dec. 1993, 1377-1386.

(3.) Gray, P. and S. Gray. "Testing Market Efficiency: Evidence from the NFL Sports Betting Market." Journal of Finance, Sept. 1997, 1725-1737.

(4.) Lawrence, M. Playbook. Cleveland, OH: Preferred Publications, Inc., 1997, 32-93.

(5.) Pankoff, L. "Market Efficiency and Football Betting." Journal of Business, Apr. 1968, 203-214.

(6.) Tryfos, P., S. Casey, S. Cook, G. Leger and B. Pylypiak. "The Profitability of Wagering on NFL Games." Management Science, Jan. 1984, 123-132.

WINS-TO-BETS RATIOS

FIRST FIVE GAMES OF NEXT SEASON Year Super Bowl Winner Wins-to-Bets 1987 NY Giants 0/5 1988 Washington 1/5 1989 San Francisco 3/5 1990 San Francisco 2/5 1991 NY Giants 1/5 1992 Washington 1/5 1993 Dallas 3/5 1994 Dallas 3/5 1995 San Francisco 2/5 1996 Dallas 2/5 1997 Green Bay 0/5 Total 18/55

33%