Closing Line Value (CLV): The One Number That Proves You Can Beat the Bookies

Picture it. Monday morning, you spot a home win at 2.60 (8/5) and get stuck in. By Saturday the team news is out, the sharp money has landed, and the same selection kicks off at 2.40 (7/5). Then your lot concede in the 89th minute and draw. Bet lost.

Most punters would file that under "bad tip". We'd file it under "good bet, bad result". You bought something at 2.60 that the whole market, armed with every scrap of information going at kick-off, agreed was worth 2.40. That gap is closing line value, or CLV, and over a few hundred bets it tells you far more about your ability than your profit and loss ever will.

So let's get into it: why the closing price is the sharpest number in betting, how to calculate CLV properly (including the bit most people get wrong), how many bets it takes before your results mean anything, and why the bookmakers paying out on your winners watch it so closely.

Every decimal price is a probability wearing a disguise. Divide 1 by the odds and you get the chance the bookmaker is implying. Add those chances up across a market and they come to more than 100%. The excess is the margin, or overround, and it's how the book gets paid whatever happens on the pitch.

Take a hypothetical Premier League match with these closing prices:

Home 2.40   Draw 3.30   Away 3.10

Implied:  1/2.40 = 0.4167
          1/3.30 = 0.3030
          1/3.10 = 0.3226
Total             = 1.0423   ->  margin of 4.23%

To see what the market really thinks, strip the margin out. The simplest way is the proportional method: divide each implied probability by the total.

Home  0.4167 / 1.0423 = 0.3998   ->  fair odds 2.501
Draw  0.3030 / 1.0423 = 0.2907   ->  fair odds 3.440
Away  0.3226 / 1.0423 = 0.3095   ->  fair odds 3.231

Those "fair" odds are your benchmark. Remember the 2.501, because it comes back shortly. If you want implied probability and expected value from the ground up, our complete guide to probability in sports betting takes it step by step.

Why the margin isn't spread evenly

Punters have always overrated longshots and underrated favourites. That's the favourite-longshot bias, and a long-run pattern of losing around 5% backing favourites but 40% backing longshots isn't unusual. Bookmakers know it, so more of their margin tends to sit on the bigger prices.

That's why sharper bettors often reach for the power method instead. It finds a single exponent that makes the probabilities add up to exactly 1, which shaves a little more margin off the outsiders:

Power method: find k so that 0.4167^k + 0.3030^k + 0.3226^k = 1
k = 1.0396   ->  fair odds  Home 2.485 / Draw 3.460 / Away 3.242

On a tight 1X2 market the differences are tiny. Pick one method and stick with it. Consistency matters far more than perfection here.

A price isn't a fixed verdict. It's a running argument between the bookmaker and the money, and it gets smarter as the week goes on.

Pinnacle is the textbook case. Founded in 1998, it was the first sportsbook to bring in a reduced-margin pricing model: no sign-up bonuses, just around 1.95 to 1.96 on two-way US markets against the standard 1.91. Do the sums and a 1.91/1.91 market carries a 4.71% margin, while 1.95/1.95 carries roughly 2.56%. Pinnacle made its money on volume, not fat margins.

The clever bit is how it sets those lines. It opens early "overnight" markets with small stake limits, lets the sharps take their shots, and shapes the price around where that informed money goes. Once the number has been tested, the limits go up, to $50,000 or more on selected games. The book learns from its sharpest customers without leaving itself exposed to them.

The exchange version

Exchanges get to the same place by a different road. On Betfair, founded in June 2000 by Andrew Black and Edward Wray, nobody sets a margin at all. The price is simply where backers and layers agree to match, and the exchange takes commission on net winnings. When thousands of people with real money down settle on a number at kick-off, that number knows a lot. Our guide to bookmakers vs betting exchanges has the full comparison.

By the off, everything is baked in: the team sheets, the injury news, the weather, the model output of every syndicate that cared enough to bet. That's why the closing line is the sharpest number you'll ever see.

That line comes from a 2017 study with a gloriously cheeky premise. Rather than build a forecasting model to out-think the bookmakers, Kaunitz and his colleagues used the consensus of the market's own odds as their probability estimate. Whenever one bookmaker's price strayed too far from that consensus, they bet it.

It worked across a 10-year historical simulation using closing odds, a six-month simulation using minute-by-minute odds, and five months of staking real money with bookmakers. They published their code, data and models as well. The lesson for us is the starting assumption: the collective market price is the best probability estimate going, and beating it is where the profit lives.

That's CLV in a nutshell. If the closing price is the best available estimate of the true chance, then every time you get a better price than the close, you've bought a bet with positive expected value. Keep doing it and, given enough bets, the results follow.

Where the close is weaker

The close isn't gospel everywhere. It's sharpest in big, liquid markets: Premier League match odds, the major Asian handicap and goal lines. In thin markets such as lower-league football, player props and niche specials, far less money shapes the final price, so you'll see some odd closing numbers. The favourite-longshot bias also drags closing prices away from fair at the extremes.

Our view: treat CLV as the gold standard in the main markets and a useful guide everywhere else.

There are three sensible ways to express CLV, and all of them work. The mistake is mixing them in the same log.

o_taken       = the decimal odds you got
o_close_raw   = the closing odds as quoted (margin included)
o_close_fair  = the closing odds with the margin removed

CLV (vs raw close)        = o_taken / o_close_raw  - 1
CLV (vs fair close) = EV% = o_taken / o_close_fair - 1
CLV (probability points)  = 1/o_close_fair - 1/o_taken

Back to Monday's bet. You took the home side at 2.60. It closed at 2.40, which is 2.501 once the margin comes out.

vs raw close:        2.60 / 2.40  - 1 = +8.3%
vs fair close:       2.60 / 2.501 - 1 = +3.9%
probability points:  39.98% - 38.46%  = +1.5 points

The number to trust

That +8.3% looks lovely. It's also flattering you. Measuring against the raw close ignores the bookmaker's margin, so it overstates your edge. The +3.9% against the fair close is the honest figure: your expected return per pound staked, assuming the closing market had the probability right.

The probability-points version says the same thing a different way. You bought a 39.98% chance at a price that only needed it to be 38.46%. Do that a few hundred times and you're in business.

It works on any market. Say you take Over 2.5 goals at 1.95 in a hypothetical match, it closes at 1.85, and the fair price works out at about 1.90 once the margin's gone. That's roughly +2.6% CLV, and it stays +2.6% even if the game finishes 1-0.

Betting results are brutally noisy. A single bet at evens either wins or loses, and that swing dwarfs any edge you've got.

Take a bettor with a genuine 3% edge at average odds of 2.00, staking flat. The table shows how far their return on investment wobbles through luck alone, and how strongly their results would point to that real edge.

Bets placed	Typical luck swing in ROI (1 SD)	Strength of evidence for a real 3% edge (z-score)
100	±10.0%	0.3
500	±4.5%	0.7
1,000	±3.2%	0.95
5,000	±1.4%	2.1
10,000	±1.0%	3.0

Assumes flat stakes at odds around 2.00.

After 100 bets, a punter with a real 3% edge could easily be 7% down, while one with no edge whatsoever could just as easily be 10% up. On those assumptions it takes roughly 3,000 bets before results alone give you 95% confidence that a 3% edge is real. For most people, that's years of betting.

CLV cuts the wait. You're comparing the price you took with the price the market settled on, and that comparison carries nothing like the win-or-lose noise. A few hundred bets that consistently beat the close tell you far more than a few hundred bets in profit. It isn't proof, but it's the best early evidence you'll get. Our value betting guide shows how to go hunting for those prices in the first place.

No fancy software required. A spreadsheet does the job, as long as you record the right things every single time.

What to log for every bet

• The bet itself. Date and time taken, event, market, selection, bookmaker, odds taken and stake.
• The close. The closing price for your selection and for every other outcome in that market, so you can strip out the margin.
• The verdict. Fair closing odds, your CLV percentage and the result.

Closing prices are easier to find than you'd think. For football, football-data.co.uk publishes free historical results and odds files where closing prices carry an extra "C" in the column name. So B365CH is Bet365's closing home price and PSCH is Pinnacle's, with market-average columns alongside. Better still, you can test any idea on past seasons before risking a penny.

The calculation is short enough to write yourself:

implied  = [1/h, 1/d, 1/a]                # closing 1X2 prices
total    = sum(implied)
fair     = [total / p for p in implied]   # fair odds = 1 / (p / total)
clv      = odds_taken / fair[selection] - 1

Then track your stake-weighted average CLV, the share of bets that beat the close, and CLV split by league, market, bookmaker and when you placed the bet. Our rule of thumb: give it at least 200 to 300 bets before drawing conclusions. Decide up front which close you're measuring against, whether that's Pinnacle, an exchange or the market average, and never switch halfway through.

CLV is the best yardstick going, but it has limits, and knowing them makes you a better judge of your own record.

Timing isn't the same as a model. If you simply bet early on news everyone will know by Saturday, prices will tend to move your way. That's still genuine value, and you're being paid for being quick, but it's a different edge from out-thinking the market on the football itself. Split your log into pre-team-news and post-team-news bets and you'll soon see which one you've actually got.

Variance cuts both ways, too. You can beat the close over hundreds of bets and still be down, and you can lose to the close and be up for a while. Over time the two converge. In the short run, they don't have to.

And don't forget the thin-market problem. A brilliant CLV figure built on fourth-tier football and obscure player props is less convincing than a modest one on Premier League match odds.

Here's the twist. The very number that shows you're good is the one that gets you noticed.

Recreational bookmakers make their money from margin on recreational action. A customer who keeps taking prices that later shorten is the clearest sign going that someone is beating the book, and the obvious response is to cut their stakes or close the account. Kaunitz's team flagged exactly this, calling out the betting industry's "discriminatory practices against successful clients". Consistently beating the closing line is about as loud a signal as you can send. Our guide to why bookmakers limit winning accounts goes deeper.

What to do about it

• Use sharper venues for the bulk of your betting. Exchanges and low-margin books built on high turnover are designed around informed money, not against it.
• Shop around. The more prices you compare, the more often you take the best one available, and that alone lifts your CLV. Our line shopping guide shows how.
• Accept it as the cost of winning. A restriction is the market telling you that you were right. Wear it as a badge, not a crisis.

• Judge every bet by the price, not the result. A loser at 2.60 that closed 2.40 was a good bet. A winner at 2.20 that closed 2.50 was a bad one.
• Always measure against the fair close. Raw-close CLV flatters you; the margin-free number is the one that predicts your long-run return.
• Log everything from day one. Two hundred bets of honest CLV data are worth more than a year of remembering your best winners.
• Use the big markets as your proof. Premier League match odds and main goal lines give you the cleanest CLV signal; treat thin markets as a bonus, not your evidence.
• Bet early when you have a genuine view. The biggest CLV sits in the gap between Monday's price and Saturday's team news.
• Test before you stake. Free historical closing odds mean you can check any system against the close on past seasons first.
• If you're getting restricted, you're probably doing something right. Move your volume to sharper venues and keep going.

Whatever your CLV says, keep your stakes flat and sensible. Even a genuine edge needs a bankroll that survives the bad runs.

• Bookmaker margin and fair odds: how to remove the vig
• Probability in sports betting: the complete guide to odds, EV and value
• Value betting in football: how to find EV bets
• Why do bookmaker odds differ? A line shopping guide
• Why bookmakers limit winning accounts

18+. Please gamble responsibly. For free, confidential support visit BeGambleAware.org.