Sports gambling is entering it’s first series of major crises since widespread legalization. While there is the typical handwringing around the intersection of vice and broad entertainment, there is also the added dimension of the role that insider information can and should play within any speculative market. Those arguments, conducted earnestly, are of course completely valid, but I think they are not giving enough attention to a key distinction in the online incarnation of sports gambling.
Speculating on sports outcomes produces the same elicitation and aggregation of information as a more traditional speculative market, such as commodities futures or stock equity markets. Information, acted upon through purchase, reveal each individual’s beliefs about the true value of a contract paid upon the conclusion of a sporting event or the price of agricultural commodity at a given date and time. The market exchange of these contracts aggregates these beliefs into a collective piece of information in the form of a market price. Some contract holders get richer, some poorer, and the broader world benefits from the distillation of private information into public prices. The problems within sports gambling stem from the second channel through which entertainment is provided and paid for: random outcome generation.
Sports match outcomes are something you speculate on. Random outcomes are something you gamble on. Yes, there is random chaos in sports the same way there is random weather in agriculture. There is no speculating on a roulette wheel, however, that’s a pure gamble. I believe the major sports leagues and the online gambling companies they partner with have made a grievous error allowing their sites to offer (nearly) pure gambles.
Think about how much casinos invest in the integrity of their games as pure and fair gambles. Dice are rigorously inspected and routinely replaced. Roulette wheels are engineered with astounding tolerances. Card games occur under multiple layers of scrutinous observation. Manipulation under such conditions is sufficiently costly such that it is almost never worth undertaking.
How do you go about making similar investments in monitoring 6 inches of horizontal manipulation of the first pitch of a baseball game? Of a marginal player taking himself out of a game injured a few minutes early? The answer is you largely can’t. So now you have human roulette wheels who can decide what number they land on. Which brings us to the second, closely related problem in the new regime of sports gambling: inframarginal game outcomes. Once a game is probabalistically decided before its official conclusion, teams will often play their substitutes to finish out the formality in order to rest their main players and protect them from injury. These players typically earn smaller salaries, often over far shorter careers, with less scrutiny over their quality of play. These are the exact players for whom a couple hundred thousand dollars may be worth incurring a small amount of risk. The product of their play in terms of success (i.e. scoring, hitting, etc) is still highly conditional on their ability relative to their opposition, but the play itself (i.e. shooting, swinging, pitching choices, fouling, etc) is entirely within their control. It may be less purely random, but it is nonetheless sold to gambling customers as fair.
Whether the outcomes in question are quasi-random outcomes or merely inframarginal, what matters is that they are not joint products of competition. To significantly manipulate these outcomes does not require the explicit or implicit coordination of multiple individuals across competing teams. Yes, one player can tilt the odds, but if you are looking to make significant money manipulating sports gambling, you can’t just tilt the odds a few percentage points. There’s a reason the Black Sox Scandal of 1919 involved eight players (seven if you consider Buck Weaver innocent, which I do).
As I love to point out, coordination across individuals is very difficult. Crimes involving coordination are, in turn, far easier to monitor. Online gambling massively reduced the transaction costs in sports gambling, opening the door for orders of magnitude increases in the number and variety of bets that could be taken. There’s obviously demand for pure gambling alongside outcome speculation, and that demand could now be met through random and inframarginal in-game player outcomes.
The danger, of course, is that few of these events are truly inframarginal. Every pitch and available player counts towards the outcome. Enough manipulation by enough players will graze away the integrity of the core product. The subsidy of lower end players through gambling will change how they approach their careers and how management approaches their employment. Fans will react accordingly as well, adjusting how they view outcomes. We’re already seemingly hardwired to view everything as causal and conspiratorial, overestimating bias in refereeing and player preferences. This will only stoke those fires further.
Organized crime famously offered a “numbers game” prior to state lotteries. Desperate for a credibly random outcome, a common mechanism was to use the middle three digits of the number of shares traded on the NYSE as the winning number. There are no shortage of lotteries now, but there obviously remains latent demand, and customers clearly enjoy bundling gambling with a product far more entertaining to consume than scratching off a ticket. Pro sports was unable to deny the profits from exactly such a bundling, but the cross contamination with their core product may prove to be of greater cost.
I’m not businessman, just a lowly economist and sports fan, but if I were running a $11.3 billion per year firm, I would be far more risk averse.
Economist Writing Every Day 