Learn
The Clean Close: how to calculate closing line value without the margin
The Clean Close is a three-step way to calculate closing line value with the sportsbook's margin taken out of the closing price first, so your record grades the price you got and not the vig. Turn both closing prices into probabilities, divide each by their total, then compare that fair close with your entry price. 21+. Nothing is guaranteed.
This is the sheet from the Clean Close video. The video walks through one hypothetical bet. This page gives you the same steps in the same order, the formula for both kinds of CLV, a blank sheet to copy and the mistakes that make CLV look better than it is.
Who it is for: anyone who already writes down their bets and wants to know whether they are getting good prices, which you can see long before your win-loss record means anything. You need your entry odds and the closing odds on both sides of the market from one reference book.
The three steps
- Turn both closing prices into probabilities. Use the closing odds on both sides, not only yours.
- Divide each by their total. The total is over 100 percent because the margin is still in it. Dividing removes it.
- Compare the fair close with your entry. Positive means your price beat the close.
The whole point is step 2. Skip it and every number you write down is inflated.
Step 1, both sides of the close
Convert American odds to implied probability. For a favourite, divide the odds by the odds plus 100. For an underdog, divide 100 by the odds plus 100. The moneyline reference shows the same conversions and notes that the quoted probabilities add up to more than 100 percent on a balanced book, which is how the book makes money.
In the video's hypothetical, you bet a side at -110 and it closes at -125 on your side and +105 on the other. -125 is 125 divided by 225, about 55.56 percent. +105 is 100 divided by 205, about 48.78 percent. Together they make 104.34 percent.
Step 2, divide each by the total
Divide 55.56 by 104.34 and you get 53.25 percent. Divide 48.78 by 104.34 and you get 46.75 percent. Now the two sides add up to exactly 100, and 53.25 is the market's final fair estimate for your side.
This is the simple proportional method. Research comparing margin-removal methods found that the power method generally does as well as or better than the proportional one, and with the power method the same close comes out near 53.46 percent. Either is fine. What matters is that you pick one and use it for every bet, or your numbers will not be comparable.
Step 3, compare with your entry
Your -110 implies 52.38 percent. The clean close is 53.25. The difference is +0.87 points, which is probability CLV.
There is also price CLV, a percentage, and one documented version from SharpAPI's CLV reference is: CLV percent = (closing no-vig probability x your entry decimal odds - 1) x 100. Your -110 is decimal 1.9091, so 0.5325 x 1.9091 - 1 gives about +1.65 percent. The same docs say that positive CLV means the price beat the close, that the close should be devigged, and that the reference book is typically Pinnacle.
Which CLV to track
Track both if you can, but if you pick one, pick price CLV. Probability points treat every bet the same, while a one-point move on a long underdog is a much bigger change in what you are paid than a one-point move on a heavy favourite. Price CLV uses the payout you actually got, so bets at very different odds sit on the same scale. Probability points are easier to explain to yourself, which is why the video uses them.
Why the raw number fools you
Without step 2 you would compare your 52.38 with the raw 55.56 and write down 3.18 points. That is more than three times the clean 0.87. It feels like a great bet, but most of that gap is the book's margin on the closing line, not anything you did.
| Raw close (wrong) | Clean close | |
|---|---|---|
| Closing probability, your side | 55.56% | 53.25% |
| Your entry | 52.38% | 52.38% |
| Probability CLV | +3.18 pts | +0.87 pts |
The example is hypothetical. The point is the size of the error, which you can check with the moneyline conversion on any market you like.
The sheet
Copy this and fill one row per bet. The first row is the hypothetical from the video.
| Date | Bet | Your odds | Your implied | Close (your side) | Close (other side) | Total | Fair close | CLV (pts) |
|---|---|---|---|---|---|---|---|---|
| demo | Side A | -110 | 52.38% | -125 (55.56%) | +105 (48.78%) | 104.34% | 53.25% | +0.87 |
If you would rather not do the division by hand, the CLV calculator does the same steps.
What to watch out for
- Use one reference book. A sharp book's close is the usual reference. A fallback book gives a different close, and SharpAPI warns that fallback references can show negative CLV that should not be read as edge.
- Use the real close. The last price before the game starts, not a price from the morning.
- Same market. Compare a moneyline with a moneyline, a spread with the same spread number.
- Do not expect CLV to win every bet. A bet that beat the close still loses often. CLV grades the price, not the result.
How to tell it worked
After a few hundred rows, look at your average CLV and how often it is positive. Researchers who built a strategy from closing-odds probabilities showed that the market's closing prices carry real information, and a study of forecasting found betting odds often beat mathematical models as predictors. That is why beating the close over many bets is a sign of good prices. One week of rows is not. For more background, see closing line value explained and the fair odds calculator.
21+. For informational purposes only. Nothing is guaranteed. If you or someone you know has a gambling problem, call 1-800-GAMBLER.
Questions people ask
How do you calculate closing line value?
Convert both closing prices to probabilities, divide each by their total to remove the margin, and compare the fair close for your side with the probability your entry odds implied. Positive means you beat the close.
Why remove the margin from the closing line?
Because both sides of a close add up to more than 100 percent, as the moneyline reference explains. Without removing it, CLV is overstated, in the hypothetical above by more than three times.
Which closing line should I use?
One sharp reference book, used consistently. SharpAPI notes the reference is typically Pinnacle and that fallback books can mislead.
Does positive CLV mean I will profit?
No. It is a sign of good prices over many bets, not a promise. Any single bet can lose, and nothing is guaranteed.
Sources
- Closing Line Value (CLV), SharpAPI docs
- Moneyline odds, Wikipedia
- Adjusting bookmaker's odds to allow for overround, American Journal of Sports Science
- Beating the bookies with their own numbers, arXiv
- The Betting Odds Rating System, PLOS ONE
