# NFL Line Cheatsheet: How to Understand an NFL Line

112 “Nothing’s for certain, it can always go wrong.”

– Robert Hunter

The quote above is correct, however, it can also be misleading. Simply because an event is uncertain, it does not follow that all possible outcomes are equally probable. The purpose of this argument is to demonstrate all of the information you should be able to pull from reading an NFL line. Armed with this information you should be able to make more informed decisions. In short, the book gives you a lot of information when it releases a line – you need to be aware of what it all means.

For purposes of this article, I am going to use the line in the Colts v 49ers game, and the Draftkings line:

1. Determining Draftkings projected final score

Determining the projected final score is straightforward. Simply subtract the spread from the total, divide by two (the number of teams) and then add the spread back onto the favorites score. In this case, and using the 10/20 line, that results in the following:

44.5 – 4.5 = 40

40/2 = 20

20+4.5 = 24.5

Draftkings projected final score = 49ers 24.5 – Colts 20.

1. Determine the implied win percentage from the projected final score

Determining the implied win percentage is also straightforward, though a calculator is useful. You simply multiply the 49ers’ projected score by the power of two. Then you divide that number by the sum of the 49ers’ result and the result of the Colts’ score to the power of two. In this case, that results in the following:

24.5^2 = 600.25

20^2 = 400

= 600.25/(600.25 + 400)

= 600.25/1000.25

= .60, 0r 60%

Consequently, based on the total and the spread, Draftkings has the 49ers as a 60% favorite, and the Colts as a 40% dog.

If you predict a final score for the games as part of your process, you can determine your own implied win percentage from the same method above.

1. Determining the implied win percentage from the moneyline

To convert the moneyline into a win percentage is easy and important. How else can you tell if you are getting a good price?

To convert a favorite’s moneyline into a percent, simply divide the amount you have to bet (-200) by the amount you have to bet and the amount you will win (-200-100):

= -200/(-200-100)

= -200/-300

= 66.66%

Thus, the book’s total and spread make the 49ers a 60% favorite (as seen in step two, above) but the book’s moneyline price makes the 49ers a 66.66% favorite. This results, if they are correct, in charging you an additional 6.66% tax for taking the 49ers on the moneyline.

To convert the underdog moneyline into a percent the process is as follows:

= 100 (the amount you will bet)/(100+170)

= 100/270

= 37.03%

Thus the book’s moneyline makes the Colts only 2.97% less likely to win the game (40%-37.03%) than the total and the spread indicated.

The other thing you should see immediately is that the 49ers’ moneyline win percentage, plus the Colts’ moneyline win percentage, combine to be more than 1 (66.66 + 37.03 = 103.69%). Now I guess it makes sense that athletes can give a 110% if books can get 103.69%! Of course, athletes cannot give 110%. That is stupid. The book’s line is over 100% because they are artificially moving the numbers to ensure profit.

Before moving on, you should also understand that the book wants dogs to win. Here is why. Suppose you and I were the only two people to bet the moneyline in this game. I took the 49ers, and you took the Colts. As is common on favorites, I bet to win \$100. As is common on dogs, you bet \$100. As a result, the book would be holding my \$200, and your \$100, for a total of \$300.

If you win, the book will pay you \$270 and make \$30. If I win, the book will pay me \$300 and keep nothing. The book roots for dogs. (This is a general rule of thumb. In practice, it can get far more complicated but that is beyond the scope of this article).

1. What you are really betting, when you bet on a spread

I am going to post the same line chart again here so that you will not have to scroll:

Suppose you took the 49ers -4.5, what are you betting? In the simplest of terms, you are betting that the 49ers will win by 5 points or more. However, that is not really what you are betting because the price to bet the spread is not even money, it is -110 (often referred to as a 10 cent line, or dime line).

So that means you have to win your bet more often than 50% of the time to be profitable. We can use the exact same math from above here:

= 110(Amount you bet to win 100)/(-110-100)

= 110/-210

= 52.38%

So to break even on this bet, the 49ers must win by five points or more, at least 52.4% of the time. That is what you are betting. As the vig changes for your bet increases, say to -115 or -120, so too does the win probability needed to be successful. Always, always, always, get the best number you can at the best price you can.

1. Understanding the value of key numbers in the NFL

The NFL has a unique scoring system. You can score a single point, but only after scoring six points. You can score two points by a try, a returned try, and a safety. You can score three points with a field goal. You cannot score four or five points on any play. You can score six points. These combinations result in certain numbers appearing more frequently in the margin of victory. The real question is how much more frequently. Here is a quick cheat sheet with key numbers and the percent of games that end by that margin.

So if you lay -3.5 points in a game, you are losing on 15.09% of games played. If you lay -3 instead, you now push on that 15.09% of games. Getting the right number matters.

1. Understanding the value of the number in a particular game

You should also understand that all three-point lines are not the same. For example, three points in a game with a total of 55 points, is far less valuable than three points in a game with a total of 40 points. The reason why is simple, variance.

Variance sounds complicated. For this article think of it as the name that describes outcomes that are not expected. For example, if you flip a fair coin ten times, the odds tell you to expect five heads and five tails. However, probabilities do not assemble in a neat queue and fall in line with one another. They are subject to variance.

In fact, if you flip a coin ten times, you are likely to get five heads and five tails only 24% of the time. The reason, variance. The more times you flip a coin, 1000 for example, the less of a role variance plays and the more likely you are to experience results that match the expectation – an even distribution of heads and tails.

Given that, it is important to understand the spread as a percent of the total expected points. If the total is 55, and the spread is 3, then the spread divided by the total (3/55) equals only 5.4% of the total scoring. A small amount, which is easily slain by variance. Conversely, if the total is 40 and the spread is 3, then the spread divided by the total (3/40) equals 7.5% of the total expected scoring – a much harder ask for a variance.

1. You Always Want To Beat The Closing Line

I understand that a lot of people like to bet on Sunday mornings. There are pros and cons which could fill an entire article on their own. Just be aware that you want to beat the closing line number. Betting on the NFL is a public marketplace not all that different from the stock market. The closing line represents the fair market price. You cannot consistently win by getting a fair price in this market, you have to get a better price, consistently. If you are constantly betting the closing line number, I am willing to bet you are losing on your bets overall.

Finally, before we go, let’s look at that line chart from Draftkings one last time:

The first thing you notice if you track the spread is that money came in early on the Colts. That money drove the line down to -3.5 at which point, it met resistance and the 49ers side got involved. We appear to have a stable market read on -4, or -4.5. No matter which side you take in this game, you better beat that number. It is not arbitrary, and every half a point and increased vig matters greatly.