# What Baseball Betting Teaches Us About How Markets Deal With Profit Opportunities | Professor Matt Ryan

Wouldn’t you love to know how the stock market is going to perform today? After all, if you knew with certainty how the market would move, you could effectively write yourself a check for any amount.

Of course, knowing exactly what the market will do today is impossible. But it’s not for lack of trying. Enormous sums are poured into the effort to find a profitable investment strategy — and you can bet that if anyone discovers one, they aren’t in a hurry to publicize it. Which is why the publication of a profitable investment strategy in 2012 offers a unique opportunity to see how markets deal with profit opportunities.

But this investment strategy didn’t deal with price-to-earnings ratios, short-term market fluctuations, or initial public offerings. In fact, it didn’t deal with the stock market at all. It dealt with the Major League Baseball wagering market.

### Profitable Wagering on Baseball Underdog Teams

Understanding wagering on baseball requires understanding what’s known as a *moneyline*. Football or basketball wagering typically involves *point spreads* — a forecast of the number of points by which a stronger team is expected to defeat a weaker one. You bet on whether or not a team will “cover” the spread, or win by more than the forecasted point differential. Moneylines do not forecast point differentials but instead adjust payouts on winning bets. Take the April 18th moneyline between the New York Yankees (NYY) and the Chicago White Sox (CWS):

CWS +170

NYY -185

The numbers denote payout rates for successful wagers. A winning wager on the White Sox would pay $170 for each $100 wagered, or any fraction thereof, while a winning wager on the Yankees would pay $100 for each $185 wagered.

The profitable strategy dealt with wagering on underdogs — the team with the “+” number in the moneyline. In short, wagering on underdog teams early in the baseball season, which runs from April through October, was persistently profitable from 1999 through 2009 — and the earlier in the season and the heavier the underdog (i.e., the larger the “+” number), the better. (For readers interested in more details, table 4 in the paper, on page 714, gives a quick summary.) Blindly wagering on every underdog during the first month of the season yielded an average return of about 3% per wager. Doing the same during the first half of April raised the average return to 7%; wagering on particularly heavy underdogs (those with a line at +160 or greater) during the first half of April increased the average return to over 28%. These last two strategies are statistically significant well beyond the accepted thresholds for chance occurrence.

The question, then: How did the Major League Baseball wagering market respond to the publication of profitable wagering strategies?

Before moving forward, a few words about wagering markets.

### Why We Study Wagering Markets

Wagering markets are attractive financial markets to study for a number of reasons. First, data are extensive and readily available. Empirical analysis, obviously, would be a bit more difficult without data.

Second, wagers have fixed and certain termination points where their value becomes known. Empirically, this characteristic is very convenient. Stocks do not have this feature; in fact, some dividend pricing models have infinite time horizons.

Third, the particular structure of a wagering market — its size, the fact that bettors know immediately whether they have won or lost, and individuals taking part repeatedly — has led some to believe that wagering markets have a better chance at becoming efficient (or reaching equilibrium) as compared to other markets. In other words, there should exist no simple, persistently profitable wagering strategies.

That last point is particularly important for the issue at hand. We would expect bettors to act on any information that generates positive returns and claim profits for themselves. In response, sports books, which take bets and pay out winnings, would adjust accordingly to eliminate the hemorrhage of funds from their accounts, thereby eliminating the profitable wagering opportunity. (Then again, we might not expect profitable wagering opportunities to exist in the first place. Indeed, most studies show wagering markets to be remarkably efficient, leaving little opportunity for bettors to profit.)

So, what actually happened?

### The Consequences of Revealing Baseball Wagering Secrets

After this strategy was published, the profitable opportunities more than went away. In fact, they mostly turned into mirror-image negative returns. In the ensuing five seasons after that article appeared in print, the all-underdogs, all-April wagering returns dropped from 3% to –2.5%; the all-underdogs, early April wagering returns dropped from 7% to –4%; and the heavy-underdog, early April wagering returns plummeted from 28% to –29%.

These results would seem bizarre if we were attributing changes in returns only to posted moneylines; wouldn’t a return around 0% seem appropriate? While we explained above why we wouldn’t expect to see persistently *positive* returns to wagering strategies, we similarly wouldn’t see persistent, excessively *negative* returns to wagering strategies because unfavorable moneylines wouldn’t attract bettors in the first place, requiring sports books to make the terms more favorable for potential wagerers.

There are two factors at play in determining the change in returns to wagering strategies: the moneylines offered by sports books (which impacts the return on a winning bet), and the rate at which underdogs actually win baseball games (which determines how many underdog bets pay off). And to understand the true impact of publishing the profitable wagering strategies, we need to look at how both changed.

From 1999 to 2009, underdogs won about 44% of all April games and about 45% of early April games, with heavy underdogs winning about 35% of all April games and 40% of the early April games. All of these rates subsequently dropped for all games played from 2012 to 2016, from 44% of all April games to 43%, from 45% of early April games to 42.5%, from 35% of heavy underdogs across April to 31%, and from 40% of heavy underdogs in early April to 26%. Presumably, bettors and academic papers have no direct control over the outcome of a game — the underdogs were simply less successful after the strategy’s publication.

But the posted moneylines changed as well. For example, the average underdog line across all underdogs in April from 1999 to 2009 was +137; it subsequently dropped to +128 from 2012 to 2016. Heavy underdogs in particular averaged +192 across April and +193 in early April, but fell to +187 and +181, respectively. As it turns out, not only did underdogs win less often, but wagers on underdogs that *did* win paid out less money.

### How the Market Changed Its Payouts for Winning Underdogs

What we’re interested in is the magnitude of the latter effect — how much of the decrease in return was due to wagering lines decreasing payouts for winning underdogs? If we keep the winning percentage of underdogs the same across all time periods and only allow the lines to vary — distilling out the changes in underdog success and isolating the impact of line movement — then we can get a sense for the relative impact of the market response.

It’s a rough measure — it assumes some distributional properties of both underdog victories and wagering lines that aren’t true — but it’s insightful nevertheless. For the wagers involving heavy underdogs, the impact of the change in winning percentage dwarfs the impact of the change in lines — at most, 15% of the change can be attributed to books adjusting lines.

For all underdogs, however, the impact of the market response is much larger; the lowest estimate puts the impact around 35%, with the largest being just a shade under 50%. That’s a substantial result: as much as half of the discrepancy in returns to betting across all underdogs between periods comes from books adjusting their moneylines. In other words, we see the full cycle of profits being claimed by entrepreneurs and prices adjusting to dissipate these opportunities moving forward.

Of course, the underlying assumption here is that the article’s publication and the subsequent change in returns aren’t simply spuriously correlated. Nevertheless, it’s still interesting to get a view of the entrepreneurial actions of market participants in light of the ever-changing nature of the market and how exactly markets get themselves to their present states.

*Article by Matt E. Ryan is an Associate Professor of Economics at Duquesne University.* *Originally published at **www.learnliberty.org**.*