Closing line value: the metric that survives small samples
CLV compares the price you took against the price at kickoff. Because the closing line is the market's most accurate estimate, beating it is evidence of skill long before profit becomes statistically meaningful. Formulas, benchmarks and honest limits.
Closing line value measures whether you took a better price than the market settled on. If you back a team at 2.20 and it closes at 2.00, you have positive CLV: you bought at a price the market later judged too generous. Because the closing line aggregates all available information, consistently beating it is the fastest credible evidence that a method has an edge.
Why is the closing line the benchmark?
A market opens with the bookmaker's estimate and then absorbs everything that follows: team news, injuries, weather, professional money, and the accumulated opinion of everyone willing to back it. By the time it closes, the price reflects the most information it will ever contain. Studies of betting markets consistently find the closing line to be the single best available predictor of outcome.
That makes it a useful yardstick. If your price was better than the closing price, you were ahead of the information — not lucky about the result, which is a separate question entirely.
How do you compute closing line value?
The simplest form compares the two decimal prices directly:
CLV % = (odds taken / closing odds − 1) × 100Backing 2.20 that closes at 2.00: (2.20 / 2.00 − 1) × 100 = +10%.The rigorous form compares de-vigged probabilities rather than raw prices, because the closing price still contains margin. Take the closing odds for the full market, de-vig them (see de-vigging methods), and compare the fair closing probability against the fair probability implied by your price.
CLV (probability) = p_fair(closing) − p_fair(taken)The distinction matters more than it looks. Raw-price CLV flatters you at bookmakers with wide margins, because you are comparing your price against a number that includes a large deduction. Two bettors with identical skill will show different raw CLV purely from where they bet.
What counts as good CLV?
| Sustained CLV | Interpretation |
|---|---|
| Below 0% | The market disagreed with you and was probably right. Profit here is likely variance. |
| 0% to +1% | Marginal. Could be noise, could be a thin edge; needs a large sample to separate. |
| +1% to +2% | A genuine edge, and enough to be profitable at low margins over a long run. |
| +3% and above | Exceptional, and the first thing to check is whether the measurement is right. |
Two cautions on those bands. They assume CLV measured against a sharp closing line — a market maker's price, not a recreational book that closes soft. And they assume the account has not been limited, which is the usual fate of anyone who beats the close consistently.
Where CLV misleads
- Steam-chasing. Following a line move produces positive CLV by construction without any independent insight. You are measuring your speed, not your judgement.
- Thin markets. In a market with little money, your own bet can move the line, and you then measure yourself against a price you created.
- Wrong reference book. CLV against a soft book measures the softness of that book. Use the sharpest closing price available.
- Non-monotone markets. In markets with a push zone or half-win outcomes, a better price is not automatically better value; the settlement rules change the mapping from price to expectation. See expected value.
- Survivorship in reporting. CLV averaged over bets that were placed, when placement was itself conditional on the line already moving, is not an unbiased estimate.
CLV is evidence, not proof. It is a leading indicator of skill that becomes reliable much sooner than profit does. It is not a promise of profit, and a service quoting CLV without stating the reference book and the de-vig method is quoting an unfalsifiable number.
Why does CONSENSUS not publish a CLV figure yet?
Computing CLV honestly requires capturing the closing price for every published signal, from a consistent sharp reference, and de-vigging the closing market with a stated method. We currently fix and publish the price at the moment of assessment, but we do not yet capture the closing price for every settled signal.
So we do not have a CLV number, and we are not going to estimate one. Every service in this category could publish CLV if it wanted to; none of the ones we surveyed do. When we start capturing closing prices, the figure will appear with its reference book, its de-vig method and its sample size attached — or it will not appear at all.
Check any of this against our record
Every signal CONSENSUS publishes carries the bookmaker odds fixed before the event starts and the settled result afterwards — including the drawdowns and the losing runs. The running total is on the front page and every entry is in the log.