Closing-line value

Beating the number

You bet a favorite at -2.5 on Tuesday. By kickoff the same book has it at -3.5, so you hold a better number than anyone betting at the close can get. That difference is closing-line value, CLV for short.

It works the same on a price. Take an underdog at +130, watch it close at +115, and you beat the close by 15 cents.

Why the close is the yardstick

The opening number is one shop's opinion, offered at small limits, and through the week the money comes in to correct it: the biggest bets tend to come from the people who know most, the limits go up as the book gets surer, and the price is moved each time. By the start of the game the line has taken in most of what there is to know (more on that in why odds move).

It is not perfect. It is the best estimate of the game that anybody has, and on most days it is better than yours or mine. I never thought an opening number, mine included, was as good as the one it turned into by kickoff.

So if your price is better than the close, you have bought something for less than the market later decided it was worth.

What 15 cents is worth

Stay with the underdog you took at +130. The game closes with the favorite at -135 and the underdog at +115. The -135 is 57.45 percent and the +115 is 46.51 percent, which together make 103.96. Take out the book's cut and the closing line says the underdog wins 44.74 percent of the time, while your +130 only needs 43.48 percent to break even.

You are being paid 130 on something the market rates at 44.74 percent. That bet is worth about $2.90 for every $100 staked, before the game has started. The no-vig calculator does the middle step for you.

Results lie for a long time

Results are slow and noisy. Give a coin-flipper 100 bets at -110, and about 18 times in 100 he finishes 55 and 45 or better. Nearly one bettor in five looks like a professional after 100 bets with no skill at all.

Closing-line value shows itself faster, because every bet gives you a measurement whether it wins or loses. That's why I'd write down, for every bet, the price you took and the price at the close, and judge it over a couple of hundred bets, not ten.

The book is keeping score too

Books keep score this way. In my experience a player who beats the close week after week has his limits cut long before his results would justify it, and a player who keeps taking worse than the close is welcome however well he is running. The book is not guessing. It is using the best predictor it has, and the page on limits covers what happens next.

Where the yardstick bends

On props and minor leagues the close is made by little money at low limits, so it is a softer yardstick there. News can fool you: if a quarterback is ruled out an hour after you bet against him, you beat the close by luck (it goes the other way just as often, and over many bets it washes out). Books don't all close at the same number either, so measure against the same one every time, ideally a book that takes big bets.

And you can beat the close and lose for months. It raises your chances. It does not settle any one bet.

If you are usually worse than the close, you are betting too late or betting the side the money is against, and no staking plan fixes that. If you are usually better, I still wouldn't bet bigger because of it. The stake stays at 1 or 2 percent of your bankroll, because the losing runs come anyway.

The scoreboard tells you what happened. The closing line tells you whether it was a good bet.