A Tuesday bet at +150. By tip-off on Thursday the same side is +130, with the other side at -150. The bet has not won or lost anything yet, and it may still lose. But the market has already said something about it: the price taken on Tuesday is better than the price the market settled on. That gap is closing line value, and it is one of the few measures of betting skill that does not have to wait for the final score.
What the closing line represents
The closing line is the last price offered before an event starts. It matters because it is the line that has absorbed the most information: injury reports, lineups, weather, and the bets of everyone who chose to act before the start, including the bettors the sportsbook respects most.
In efficient markets, the final price is the best available estimate of the true chance, because every participant’s information has had a chance to move it. Betting markets are not perfectly efficient, but major closing lines are hard to beat consistently. So the closing line works as a reference. A bettor who regularly gets prices better than the close is, by that reference, regularly getting prices better than the market’s best estimate.
Every bet sits somewhere on the path between the opening price and the close. The line movement guide covers why lines travel along that path, from injury news to respected money to model updates. CLV takes that whole path and asks one question of each bet: when the price stopped moving, was the ticket on the right side of it?
A useful mental model is a stock bought before good news. The purchase does not become profitable until the shares are sold, but the moment the price rises, it is clear the buyer paid less than the market now thinks the shares are worth. A betting ticket with positive CLV is in the same position, except that it can never be sold at the closing price. It can only be held until the game decides it.
Measuring CLV with the no-vig price
The calculation has three steps. The example uses the Tuesday bet at +150 and the closing prices of +130 and -150.
| Step | Calculation | Result |
|---|---|---|
| Implied, bet side at close | 100 / 230 | 43.48% |
| Implied, other side at close | 150 / 250 | 60.00% |
| Market total | 43.48 + 60.00 | 103.48% |
| No-vig closing probability | 43.48 / 103.48 | 42.02% |
| Fair closing price | 1 / 0.4202 | 2.38 decimal, +138 |
| Decimal odds taken | +150 | 2.50 |
| CLV | 2.50 × 0.4202 - 1 | +5.04% |
A CLV of +5.04% means that, if the fair closing probability is right, the bet was worth about 5 cents of profit per dollar staked at the moment it was placed. On a $40 bet that is about $2.02 of expected value.
Why not compare +150 with the raw +130 close? Because the raw closing price includes the margin. The raw implied 43.48% overstates the fair chance, and measuring against it inflates the figure to 2.50 × 0.4348 - 1 = 8.70%. The bettor in this example is ahead of the fair price by a smaller amount than the raw numbers suggest, and only the no-vig version shows how much. The no-vig calculator returns the fair closing probability directly.
Two more examples: a favorite and a bet the market went against
CLV works the same way on a favorite. A bet at -120 that closes at -140, with the other side at +120, looks like this: the closing sides imply 58.33% and 45.45%, a total of 103.79%, so the no-vig chance of the bet side is 58.33 / 103.79 = 56.20%. The decimal odds taken are 1.8333. CLV is 1.8333 × 0.5620 - 1 = +3.04%. That is smaller than the underdog example, partly because a short price returns less per dollar, so each point of probability gained is worth less in return.
Negative CLV is just as easy to read. Take the original +150 bet, but suppose the market moved the other way and closed at +170, with the other side at -200. The closing sides imply 37.04% and 66.67%, a total of 103.70%, and the no-vig chance of the bet side is 35.71%. CLV is 2.50 × 0.3571 - 1 = -10.71%. The market decided the Tuesday price was too short, and the fair closing price of +180 says the bet was overpaid for by a wide margin.
| Bet taken | Close, bet side | Close, other side | No-vig chance | CLV |
|---|---|---|---|---|
| +150 | +130 | -150 | 42.02% | +5.04% |
| -120 | -140 | +120 | 56.20% | +3.04% |
| +150 | +170 | -200 | 35.71% | -10.71% |
The third row is what a bet can look like when it is placed just before the market moves the other way, and it is exactly the row that a CLV log is designed to expose.
CLV on spreads and totals
On spreads and totals, CLV is often quoted in points: a bet at -2.5 that closes at -3.5 has a point of CLV. That shorthand is useful but incomplete. Points are not equal in value, and the price attached to the line can change too.
The more precise approach converts the bet and the close into probabilities at the same number. That requires an estimate of how much each half-point is worth, which depends heavily on the sport and on where the line sits. In football, a half-point that crosses 3 or 7 is worth far more than a half-point at 11. Without that conversion, point-based CLV is best treated as a rough sign of direction, not a measurement.
What positive CLV proves, and what it does not
The usual argument for CLV goes like this. Results over a few hundred bets are dominated by luck. CLV is not, because it measures the price taken against the market’s estimate, without waiting for outcomes. A bettor with steady positive CLV is beating the best available estimate, so profit should follow over time.
The argument has a weak point, and it has been tested. In an essay published in August 2026, Karl Whelan studied NBA moneylines across 3,670 games from the 2022/23 to 2024/25 seasons, using quotes from licensed American sportsbooks and taking the median final price across books as the closing line. Bets with positive CLV did better on average than bets with negative CLV. But only the top tenth of bets by CLV turned that into meaningful profit; most of the positive-CLV range produced little or no profit on average. His summary is blunt: “Clearly, getting positive CLV does not guarantee winning money.”
The reason is arithmetic. Small positive CLV can still sit inside the margin. If the price taken was slightly better than the raw close but the fair chance is still below the break-even chance at that price, the bet loses money on average. CLV is a better signal than short-term results. It is still a signal, not a proof.
The live-betting problem
For in-play bets, CLV becomes awkward. A pregame bet has one obvious benchmark: the last price before the start. A live bet placed in the second quarter has no closing line in the same sense, because the market keeps moving until the game ends and its last price is effectively the result.
Some bettors substitute a short-horizon benchmark: the price at the next stoppage, or the price a fixed number of minutes later. That measures whether a live bet was timed ahead of the market’s next update, which is genuinely useful. It is not the same thing as pregame CLV, because the next price is set by a model reacting to a single play, not by a full week of information and money. There is no agreed standard here, and any number labelled “live CLV” should say how it was measured.
Why sportsbooks watch the same number
The closing line is public, so the comparison that a bettor runs on their own bets is one a sportsbook can run on every customer. An account that regularly gets prices better than the close is, from the book’s side, an account that knows something or moves faster than the market. Sportsbooks can reduce the maximum stake on individual accounts, and steady positive CLV is the kind of pattern that draws that attention.
This creates an odd tension. The measure that best indicates skill is also the measure most likely to shrink the stakes a skilled bettor can place. The timing trade-off is real as well: betting early captures more of the move toward the close, but early lines carry lower limits and less information, so an early bet is placed with a weaker picture of the game.
Keeping a CLV log that means something
If you want to track CLV, a few rules keep the log honest. Record the decimal odds taken, the time, and the closing prices on both sides from a consensus of books, not one. Convert both to fair probabilities. Track the average CLV and also the share of bets with positive CLV, because a high average driven by a few lucky long shots is less convincing than a steady small edge across most bets.
A spreadsheet is enough. Six columns cover it: date, market, odds taken, closing price on the bet side, closing price on the other side, and the CLV figure the formula returns. Sorting by CLV after a month often shows which markets or which times of the week produce the best prices, which is more actionable than a win-loss record.
Then compare CLV with results over time. They should point the same way across hundreds of bets. If CLV is consistently positive but results are consistently negative after several hundred bets, the CLV measurement is probably flawed, or the edge is too small to clear the margin. Whelan’s results show that the second possibility is a real one.
The Tuesday bet at +150 is a useful marker: about 5% of CLV on a side the market closed at a fair +138, worth $2.02 in expected value on a $40 stake, and still a ticket that loses more than half the time.