Profit Answers the Wrong Question
Every bettor asks whether they are good at this, and almost everyone answers in the wrong way by looking at their account balance.
Profit is a statement about outcomes, and outcomes are dominated by variance in the short run. Take an edge of two per cent per bet and the distribution around it is wide: a run of eleven losses inside a thirty-bet block has a probability of a few per cent. A skilled player regularly looks terrible for a month and an unskilled player regularly looks excellent. Our bankroll management guide covers the stake sizing that stops a variance streak from becoming a bankroll event.
Closing line value asks a different question. It asks not what happened but what you were paid relative to the market's final assessment of the same bet. That is a property of the decision, not of the result, and it is the only widely available measurement that separates skill from luck.
The Four Facts You Need to Record
Four columns. The recording burden is small and it is the whole method.
| Column | What goes in it | Why it matters |
|---|---|---|
| Selection and market | The exact market, not the fixture name. | Closing prices only exist for identical markets. |
| Price taken | Your decimal odds at the moment of placement. | The numerator of the comparison. |
| Closing price | Where the market settled, from an unbiased source. | The denominator, and the thing being measured. |
| Stake | Amount staked, and the book used. | Converts edge into money and reveals restriction. |
How the Measurement Works
Closing line value is a simple ratio with one important subtlety in how you choose the closing price.
The comparison is the price you took against the price the market finished at. Back a selection at 2.05 and it closes at 1.95, and the market moved against you after you filled, which is the signature of having taken value. Take 2.05 and it closes at 2.20, and the market moved towards your position.
The subtlety is the source. Recording the closing price from the book you bet with measures that book's final position, not the market's. Recording it at one sharp book while betting a soft one measures the gap between the two, which is real but a different thing. The honest choice is a consistent, liquid, low-margin reference book, named in every row. Our the margin guide covers why the reference book's margin changes the number.
What a Positive Number Actually Means
A positive average closing line value says you are systematically taking better prices than the market. That is a real edge, and it is genuinely hard to fake.
One limitation needs stating plainly. If you only bet when the current price sits above the eventual closing price, you have built a positive average by selection rather than by skill, and that selection is not predictive. This is the difference between a value bet and a bet on movement, and only one is an edge. The test is whether you would have placed it before seeing any movement.
The second limitation is sample size. Closing line value per bet has a standard deviation of four to six percentage points on mainstream markets, so the standard error on a hundred bets is roughly half a point per bet. At that noise level a player running a genuine plus 0.4 per cent edge looks identical to a player running nothing. You need several hundred bets, which is the least convenient truth in this article.
The Number That Moves First Is Not the Number That Pays
Track both, and expect them to disagree for long stretches.
The Recording Habit That Makes It Possible
The measurement is only useful if it is cheap enough to maintain for two years, so the record has to be minimal.
Most people fail by recording too much: every price movement, every line shopped, every near miss. A journal with thirty columns is abandoned inside a month, and an abandoned journal is worth nothing. Four columns is enough: market, price taken, closing price, stake.
Set the closing price capture at a fixed moment and write that moment down when you set up the process. The last price before kickoff is conventional and easy to reproduce. Either that or the last price at a fixed reference book is defensible; capturing it retrospectively from memory is not.
Our how we rate bookmakers covers the operational testing, and sportsbook limits covers what restriction looks like in the data.
Seven Rules for Benchmarking Honestly
These are the rules that keep the measurement from flattering you.
- Record at the moment of placement, not afterwards, because your memory of a price is optimistic.
- Use one reference book for closing prices and record which one, every single time.
- Capture the closing price at a fixed time and stick to it even when a different convention would suit you better.
- Log the bets you did not take, at least occasionally, because only logging winners is the easiest way to fabricate an edge.
- Never filter the sample by outcome, because a benchmark drawn from winning bets measures nothing.
- Review at 100-bet intervals, not daily, because daily closing line value is noise with a decimal point.
- Keep betting after the first good number, because the noise band is wide enough that one good stretch proves nothing.
How to Read the Result Once You Have Hundreds of Bets
At that point the number is interpretable, and there are three readings.
Consistently positive. You are taking better prices than the market clears at. The next question is size: a 0.3 per cent edge with ten bets a month is not an income. Convert the edge into expected monthly profit, because a positive number on a spreadsheet is not a plan.
Around zero. You are taking roughly the market price, so results are determined by variance rather than by you. Most recreational bettors sit here: the profit line is noise around a downward drift, and your stakes decide the outcome.
Consistently negative. The commonest reading and the most useful, because it is diagnosable. You are systematically taking prices the market has moved away from: chasing, betting on movement rather than value, or staking into markets where the book has a sharper feed.
Where the Method Breaks Down
Three honest limitations, none of which are reasons to abandon it.
Markets with no meaningful closing price. Correct score, some player props and many exotic markets have a closing price that is itself unreliable, because the market is thin and the last price is a single trader's position rather than a consensus. Comparing your price to that is measuring noise.
Live betting. There is no useful closing line on a live market, because the final in-play price is reached immediately and the bet has already been settled against it. The comparable discipline for live is to record the price at the moment of decision and compare it against the price on the same market a fixed number of seconds later, which is a different measurement with the same logic.
Promotions. A free bet or enhanced odds changes the effective price in a way the closing price does not capture. Record the effective price after the promotion, not the headline, or your benchmark will flatter every promotional bet.
Our the live betting guide covers the in-play variant, and free bets explained covers how to price a promotion properly.
The Honest Position
Closing line benchmarking is the only measurement in betting that tells you the truth about skill, and it tells you very slowly.
It works because it isolates the decision from the outcome. A bettor who takes better prices than the market will, over enough bets, finish up. A bettor who does not will not, however many lucky accumulators they have along the way. Knowing which one you are is the difference between gambling and evaluating a decision.
The slowness is the cost. Hundreds of bets is a year of record keeping, and the noise band is wide enough that most people read a false signal somewhere in the first hundred. Anyone who has found their edge in a fortnight has found variance, not skill. The value betting mathematics covers the sample size argument properly, and the arbitrage guide covers the one part of betting that produces a measurable return without an edge at all.
✓ What We Like
- Covers the payment problem that decides whether the product is usable for you
- Explains the tax structure, which is the part that sets the price
- Uses named regulators, laws and figures rather than general description
✕ What Could Improve
- A limit or ceiling constrains how much you can actually stake
