The Efficient Market Hypothesis and Sports Betting

Making money betting on sports is not easy – you either need to know more about sports, more about statistical modeling, or have access to more information than the oddsmakers. That is, sports betting is a very efficient market. Without any evidence to the contrary, one might guess that Vegas has some of the best sports analysts and modelers working for them to set the lines the public bets against.

But that is not necessarily the case. In fact, the role of the oddsmakers in sports betting is much, much simpler. In this article we’re going to explore what the oddsmakers do, how they determine the opening lines, how they move the lines, and why an oddsmaker doesn’t necessarily need to be as accurate as you might otherwise think. We’ll also take an economic/financial market perspective on sports betting to help us understand why the Vegas lines are always so hard to beat. Let’s start by covering the language and vocabulary commonly used in this world.

The Vocabulary of Sports Betting

Moneyline – The moneyline or ML is a type of bet which will look like either “Team A +300” or “Team B -200”. The “+300” and the “-200” are the moneylines. A moneyline with a “+” in front indicates the profit on a $100 bet. For example, a successful bet of $100 on a +300 ML will return the original $100 wager to the bettor and an additional $300 of profit. On the other hand, a prepended minus in front of a number indicates the amount bet in order to profit $100. A successful $100 bet on a -200 ML will return the bettor’s original $100 plus an additional $50 of profit (because a $200 bet would have returned $100 in profit and the ratio remains constant no matter the amount bet).

Moneylines let Vegas offer “unequal” lines where they think it is more likely one team or another will win. The underdog will typically have the “+XYZ” odds while the favorite will typically have the “-XYZ” odds. Moneylines are also used in bets involving more than two teams, for example in betting on the champion before a season starts.

Point Spread – In an unequal matchup, an alternative to offering ML odds is to offer point spreads. For example, the phrase “Team A (+3.5) v. Team B” means that Vegas thinks that taking Team A’s score and adding 3.5 points to it should be a fair matchup against team B. Betting on Team A (+3.5) will be successful if the team you bet on loses by 3 points or less or wins because they won the handicapped matchup.

Vig or Juice – Because the sportsbook is offering a service to the public in taking bets and setting lines, they usually take a cut of all the money bet for themselves. This cut shows up in what is called the vig or the juice. One way this shows up is that point spread bets are usually paid out with -110 odds even though it is a supposed 50/50 matchup.

If one better bets $110 on team A and a second bets $110 on team B, both against the spread, then one of them will receive their original $110 bet + $100 profit while the other bettor receives nothing. $220 was bet but only $210 was paid out (no matter who wins! More on this later) so the book retains $10. This -110 number on point spreads is fairly common, though some more aggressive books may offer vigs as low as -107 or -105 to entice more players and more money to be spent.

Opening Line – Because situations change (players get hurt, teams over/under perform, etc.), the odds will change over time. In fact, the movement of the line is vital for oddsmakers to ensure that they can make a profit. However, the line has to start somewhere. The oddsmakers first guess as to what they think is fair is called the opening line.

Efficient Market – While not a sports betting term, we’ll talk a lot here about efficient markets. The phrase efficient market refers to a system of buying assets where the price of an asset reflects its value perfectly. Though usually used in a stock market context, we can study sports betting as an efficient market or an inefficient market by asking whether or not the asset we’re buying is a fair value. In sports betting, the thing you’re buying is the right to collect more money than you bet originally if your bet is successful.

The Roll of Oddsmakers and How Sports Betting Works by Moving the Lines

The movement of the line is precisely how oddsmakers guarantee themselves a profit. Let’s just look at a simple point spread example to show how updating lines can guarantee that the book profits no matter what happens. Suppose an oddsmaker sets an opening line of Team A (+3.5) v Team B. They let this line run for a day and $1100 is bet all on Team A’s side of the spread.

This is bad for the book. Now if Team A covers the spread, the book will lose $1000 while if Team B covers the book will win $1100. It might be a fair tradeoff, but it’s a big risk for the oddsmakers.

What if – on the other hand – after taking the first $110 on team A the sports book moves its line down to Team A (+0.5) instead of +3.5? That means that team A is being given fewer points and thus bettors will be more likely to bet on Team B.

If the next $220 is bet on team B, that means that the line moved too far towards team B because now everyone is betting on them. At this point the oddsmakers would adjust the line again to either Team A (+1.5) or (+2.5). The closing lines reflect the numeric value where an equal amount of money is bet on both teams.

If that happens, now in our example of $1100 total being bet, $550 is bet on each team. No matter who wins, the winner receives their $550 wager back plus $500 of winnings. This results in $1100-$1050=$50 of profit for the book because they moved the lines. This is good practice for the sportsbook and oddsmakers and is the reason that sports markets are efficient markets.

Bad Opening Lines Can Hurt the Book

Oddsmakers don’t need to perfectly nail the accuracy of the opening line in order to profit because they can move the line as we saw in the last section. However, good models and a good understanding of the underlying sports actually are important. A bad opening line can hurt the book. Let’s see how.

Suppose that everyone thinks that team A and team B are perfectly evenly matched. But, suppose the smart guy at the sportsbook goes out on holiday and they set the opening line at Team A (+9.5). $1100 comes in on team A immediately before the line can move because adding +9.5 points to an otherwise fair matchup means that Team A should win almost every time.

Now, suppose the smart guy comes back from their holiday and helps the lines to settle at Team A (-1.5) to get an additional $5500 bet on Team B and an additional $4400 on Team A (as an exercise: why would more money be bet on Team A with the (-1.5) spread?).

Now, $5500 dollars were bet on both team A and team B. But, because the sportsbook has to honor the $1000 of bets on team A at (+9.5), we have the following amounts bet at each of the following odds.

Then, the following three things can happen.

Notice that in the third case the book loses money. The bad initial line means that there is a case where the book loses money when two of the groups hit. Coming up with a good initial line will minimize the impact in this case. Though the closing line reflected an efficient market, the damage was done because the assets were incorrectly priced relative to their value at the beginning of the betting period.

Efficient Markets and Betting Lines (Why Sports Betting is Hard)

One of the reasons that consistently making money sports betting is difficult is precisely because money can be made doing it. I am not an economist by any stretch of the imagination, but there is an explanation for this fact in the efficient market hypothesis.

The efficient market hypothesis states that the price of a liquid asset will accurately reflect its value based on the available information. If a line in sports is incorrectly set, then people will bet on the side that is expected to make money. Because lines move in response to betting, an incorrectly set line will move towards the equilibrium point until an equal number of bets are coming in on either side.

So how can an efficient market be beaten? The only way to beat an efficient market is to have access to information that the rest of the populace involved in the market does not. That’s right, we’re talking about insider information. There is a reason insider trading is illegal. Being privy to information about which players are sick or hungover can swing the projected outcome of the game enough to change the likely outcome of the game. This extra information let’s the bettors have a better idea of the true value of the asset.

Information is a tricky thing, though. I’m not suggesting that anyone making money sports betting has the cell phone number of a team trainer. A new statistical model that nobody else has developed might be able to generate more information that is not publicly available. For example, maybe my purified possessions model is able to give me enough information that I can have an edge in the market. If my mathematics or my model is able to generate enough new information that the general public doesn’t have, then I can make money as a professional gambler.

Self Interest Breaking the Efficiency

Some people – including Warren Buffet – don’t believe in the efficient market hypothesis. However, I wouldn’t trust Warren Buffet on such a topic because the product he is selling at his trust fund requires the purchaser to think that the markets are incorrectly priced and that an actively managed account can make money in the long term. That is, if it was widely known that the stock market was a perfectly efficient market, there would be no desire to spend money on hedge fund management. This is why I don’t put any stock in what hedge fund managers think about the validity of the efficient market hypothesis.

But that is the stock market. There are some considerations that may indicate that the sports betting market is not efficient. The most prominent of which is that sports bettors do not always act in their own self interest. For example, there are certain individuals that will bet on their favorite team to win the championship every year regardless of the offered odds. Abstractly, these people are buying an asset at any price regardless of its true value. Because the price of this asset will adjust in response to this person purchasing the asset, perhaps the asset will not be fairly priced.

In more concrete terms, maybe because the Cowboys have the most fans they will also always have the most money bet on them to win the championship. This means that Vegas might systematically overestimate the Cowboys’ odds to win games and championships. As a result, perhaps betting against the Cowboys is a long-term moneymaking strategy.

On the other hand, maybe others have realized that this is the case and have already invested large sums betting against the Cowboys because of this perceived benefit. If this is the case, then the lines would shift back away from the Cowboys back towards their efficient market equilibrium.

Sports Market Efficiency from a Machine Learning Perspective

There is a neat parallel that can be drawn between efficient markets and ensemble learning. I’ve actually talked before about how many things in sports can be viewed as examples of ensemble learning.

Ensemble learning is a concept in the field of machine learning which, informally, suggests that consensus opinions from sources that reach their conclusions in independent ways are more accurate than the opinions of individuals. For example, the college basketball top 25 is an ensemble (average) of the rankings of many different pollsters. Each pollster may have their own methodology or biases, but these will tend to even out when everyone’s opinions are brought together and averaged out.

We can relate this idea back to efficient markets and the concept of information. Different investors and sports bettors will use different models and criteria in making their bets. These independent bettors will be acting on different information but all the available information that is valuable will certainly be used by one party or another.

As a result, even though each individual sports bettor has their own methodologies and biases and may not be making very smart bets, the closing Vegas line is actually an ensemble of all the individual predictions of the individual bettors. So, in some sense Vegas lines are a crowd sourced ensemble learning machine learning algorithm.

Machine learning research has shown the power of ensembles of weak models. It is hard to beat them. Maybe this is one of the reasons that sports betting is so hard and wisdom of the masses ends up being a pretty powerful thing.

Final Thoughts

Anything involving money and ways to make money is going to be complicated because of the potential payoffs. Sports betting is nearly identical to the stock market in that successful money-making methods are nearly infinitely scalable and require nothing other than existing capital.

In sports betting, the role of the oddsmaker is not to predict the outcome of the game, but to run a market for the buying of assets. With that perspective, it might be a bit funny to say, but sportsbooks are themselves not betting on sports. It really doesn’t matter what is being bet on, just that there is money changing hands and enough interest from the general populace to sustain the market.

To subscribe for email updates when new posts come online, use the subscription form below!