Does Uncertainty of Outcome Define Sport?

A common argument against match fixing, doping and other corrupting influences in sport is that such activities detract from uncertainty in outcomes, which is taken to be a defining feature of sporting contests. Here I argue that the notion of "uncertainty of outcome" as a criterion to be used in identifying corruption is highly problematic both empirically and theoretically.

McClaren (2008, here in PDF) offers a standard view of the role of uncertainty in sport and its relationship to corruption:

Corruption attempts to alter this equation and make sport more of an entertainment event with a greater certainty of outcome. This equation is altered when corruption is centered on match fixing or gambling; biased refereeing; and, to a similar but different degree, when sporting results are affected by the use of performance enhancing drugs. Corruption, in any of the foregoing forms, robs sport of its essential feature of uncertainty of the outcome and accelerates its spin into the forum of entertainment, and thus it no longer is sport. Corruption gnaws away at the fundamental foundations of sport and therefore of sporting integrity. It becomes essential to protect that integrity to ensure that sport is free from any corrupt influence that might cast doubt over the authenticity and unpredictability of the sporting result.

Let's take a step back and ask, what is uncertainty anyway?

In The Honest Broker I define uncertainty as follows:

Uncertainty means that in a particular situation more than one outcome is consistent with our expectations. An “outcome” simply refers to an actual situation (i.e., some realized or true condition) in the past, present, or future, such as the number of whales in the ocean, the current temperature you feel on your skin as you read this sentence, or the roll of a die.

In sport, an “outcome” might refer to any number of things. It could be victory in a contest, the result of individual events within a contest (a serve, a pitch to a batter, a corner kick, a field goal attempt, etc.), the result of a series of contests (a league winner or tournament champion) or something else. From this perspective, sport is indeed a competition over uncertainty in an effort to reduce the possible scope of outcomes in a particular direction. A pitcher wants to make an out certain, a batter wants to make getting on base certain.

From this perspective on uncertainty, can we use the notion of "uncertainty of outcome" as the basis for identifying corrupting influences in sport?

Let's first ask an empirical question. Do spectators of sports value uncertainty as the defining characteristic of sport? If this value is strong, one could make a utilitarian argument against efforts to reduce uncertainty, based on the subjective perceptions of stakeholders in sporting events.

On the face of it one might expect the evidence to be mixed. After all, audiences routinely turn out to see performances such as Romeo and Juliet and Zero Dark Thirty, despite knowing that there is no uncertainty in the outcome or the events within the story. Perhaps, however, sports are different?

In a 2003 review Stefan Szymanski found the evidence was indeed mixed (here in PDF): 

Overall, of the 22 cases cited here, ten offer clear support for the uncertainty of outcome hypothesis, seven offer weak support, and five contradict it. Given that even supportive studies on the issue of match uncertainty seem to imply that attendance is maximized when the home team is about twice as likely to win as the visiting team, the empirical evidence in this area seems far from unambiguous. This is remarkable given the weight that is placed on this argument in policy making and in antitrust cases. Given that even quite unbalanced matches, championships, and leagues can be attractive to consumers, a more nuanced approach is called for.

In addition to the lack of strong evidence in support of the "uncertainty of outcome" hypothesis, there are also fundamental theoretical problems with the uncertainty of outcome hypothesis.

At the start of a fair match, opposing sides wish to reduce uncertainty of outcome such that certain outcomes are ruled out (e..g, their opponent proving victorious) and others are inevitable (e.g., a victory for themselves). Right away this shows that a definition of corruption as an effort to rob sport of uncertainty has the unhelpful effect of defining the actions of would-be corruptors in the exact same manner that we define the actions of fair competitors:  

Both would-be corruptors of and fair-minded participants in sport seek to reduce uncertainty of outcome. Consequently, it is not the effort to reduce uncertainty or its consequential reduction that is problematic, it is the means employed.

Ryan Rodenberg, a smart professor of sports law at Florida State University finds it problematic that existing laws (in the United States) cover only gambling-related efforts to influence uncertainty of outcome. Last week he wrote over at The Atlantic:

Uncertainty of outcome is also the reason American sports fans should take a moment to pause when commercialized sports are juxtaposed with the Quiz Show Scandal from more than 50 years ago. The web of laws applicable to sports, such as the Sports Bribery Act of 1964, only prohibits gambling-related corruption. There is no federal law explicitly preventing the clandestine manipulation of sporting events to enhance suspense. This gap is problematic. As with certain televised quiz shows decades ago, the in-game action of sporting events can be contrived in profit-maximizing ways. The federal law passed in the wake of the Quiz Show Scandal does not explicitly include televised sports; it only forbids deception of the public in connection with contests of an "intellectual" nature. Wheel of Fortune, Jeopardy!, and Who Wants to Be A Millionaire are covered by the law. Football, basketball, and baseball are not.

I'm not so sure. I get what he is arguing, however, I would argue that the existence of manipulation is not necessarily problematic, much less an indication of a corruptive influence. Sport after all are nothing if not a contrived activity.

The only reason that match-fixing and doping are prohibited -- sometimes by the constitutive rules of sporting contests and sometimes by domestic or international law -- is because people have decided to put rules in place proscribing certain actions. Corruption in sport is whatever is decided to be corruption in the constitutive rules of the contest.

Unfortunately, efforts to reduce uncertainty in sporting outcomes do not provide a clear basis for identifying corruption. Ultimately it is the means employed to reduce uncertainty that matter, and will vary from sport to sport, jurisdiction by jurisdiction. This of course is one important reason why rooting out corruption in sport proves so difficult. I will follow up with discussion of specific cases in future posts.