What is closing line value (CLV)?
Closing line value (CLV) compares the odds you took with the final odds just before the event starts. If you regularly beat the closing price, the market moved your way after you bet, which is widely treated as a sign of skill that shows up faster than profit does. It is a useful signal, not a guarantee: you still need a record of real results. 13 of the 122 AI tools we rate mention CLV or closing prices in their published record.
How CLV is worked out
- Implied chance at a price
p = 1 ÷ decimal odds - CLV (one common form)
CLV = odds taken ÷ closing odds − 1positive when you took a bigger price than the close
Illustration (not a real bet): you take decimal odds of 3.00 and the price closes at 2.50. CLV = 3.00 ÷ 2.50 − 1 = +20%. The market ended up rating your selection's chance at 40% (1 ÷ 2.50), against 33% (1 ÷ 3.00) when you bet.
CLV calculator
Enter decimal odds. Works in your browser; nothing is sent or stored.
The calculator uses the same formula as above. A positive CLV on one bet means little; it is the average over hundreds of bets that matters. It doesn't predict results.
Closing prices usually include the bookmaker's margin. Some analysts remove it first (“no-vig” closing odds); results differ slightly depending on the method, so a CLV figure should say how it was calculated.
Why people use it
- Speed. Profit swings a lot over a few hundred bets; whether you beat the close settles down faster.
- Market wisdom. Closing prices in big markets reflect a lot of information. Research that beat football bookmakers used the average market odds as its estimate of true chances (Kaunitz, Zhong and Kreiner, 2017).
- Price matters. A tip that beats the close only helps if you could get the early price. If the price is gone by the time you see the tip, the record may not be repeatable.
Limits
- CLV doesn't pay you; only results do. A positive CLV with a losing record over a large sample deserves a closer look.
- Thin markets (small leagues, early prices, minor races) can close at prices that aren't efficient.
- Self-reported CLV figures are only as good as the ledger behind them.
How we use closing prices
Our Transparency score gives points when a record states its price basis (SP, Betfair SP, exchange or closing prices). Our Performance score uses whatever price basis the record is settled at and flags records at advised odds, which may be hard to match. Badges and flags.
AI tools that publish CLV or closing prices (13)
Quoted from each tool's own published record or claims, as stated (self-reported unless the evidence badge says otherwise). We don't score CLV directly; it appears in the evidence notes and in the price basis of the Performance score.
Tennis Foresight
+6.20% ROI over 654 level-stakes bets (Live model picks, default rules, 7 Oct – 1 Nov 2026), self-reported. Price basis: Best regulated price available when each pick appeared. On CLV (as published): “Closing-line value as published on 26 Oct 2026: +0.0% (35% of 200 picks beat the closing price)”
• Self-reported⚑ Advised odds, may be hard to matchRecord period: Live model picks, default rules, 7 Oct – 1 Nov 2026Checked 2 Nov · next re-check 9 Nov53PerformanceProfitable record40TransparencyMedium transparencySportsPickLab
-1.30% ROI over 9,245 level-stakes bets (All time to 27 Oct 2026), self-reported. Price basis: Price recorded when each pick was published, flat 1 unit. On CLV (as published): “It also reports +1.9% average closing line value, with 67.6% of picks beating the closing price”
• Self-reported⚑ Advised odds, may be hard to matchRecord period: All time to 27 Oct 2026 (start date not shown)Checked 3 Nov · next re-check 10 Nov49PerformanceLosing record90TransparencyHigh transparencyKepler
-1.29% ROI over 1,353 level-stakes bets (Aug – 26 Oct 2026), self-reported. Price basis: Price posted with each pick (also shown at kick-off and close). On CLV (as published): “closing line value +0.59% on 668 picks priced at close”
• Self-reported⚑ Advised odds, may be hard to matchRecord period: Aug – 26 Oct 2026Checked 27 Oct · re-check overdue49PerformanceLosing record80TransparencyHigh transparencyKickscore AI
-3.90% ROI over 429 level-stakes bets (24 Sep – 31 Oct 2026), self-reported. Price basis: Odds recorded when the bet was named (CLV against close also shown). On CLV (as published): “Odds recorded when the bet was named (CLV against close also shown)”
• Self-reported⚑ Advised odds, may be hard to matchRecord period: 24 Sep – 31 Oct 2026 (AI consensus, our published-prediction line)Checked 31 Oct · next re-check 7 Nov49PerformanceLosing record25TransparencyLow transparencyOddsLens
-1.40% ROI over 7,443 level-stakes bets (9 Apr 2026 – 4 Nov 2026), self-reported. Price basis: Best price available when each pick was logged before the off, £1 level stakes. On CLV (as published): “Average closing line value +10.4% against SP, with 60.5% of picks logged at a bigger price than the horse started”
• Self-reported⚑ Advised odds, may be hard to matchRecord period: 9 Apr 2026 – 4 Nov 2026 (all time)Checked 4 Nov · next re-check 11 Nov48PerformanceLosing record100TransparencyHigh transparencyAlphr
-4.70% ROI over 533 level-stakes bets (2026 seasons: AFL, NRL, Super League), self-reported. Price basis: Real closing head-to-head prices (simulated settlement). On CLV (as published): “Head-to-head tips, 1 unit on every priced tip, settled at closing prices: AFL 217 bets −12.3u (−5.7%), NRL 213 −10.8u (−5.0%), Super League 103 −1.9u (−1.9%)”
• Self-reportedRecord period: 2026 seasons: AFL, NRL, Super LeagueChecked 3 Nov · next re-check 10 Nov48PerformanceLosing record65TransparencyHigh transparencyNoGuessBets
-4.77% ROI over 13,192 level-stakes bets (Ledger to 25 Oct 2026), self-reported. Price basis: Posted American odds frozen at publication (sportsbook named per pick). On CLV (as published): “CLV coverage 86.0%”
• Self-reported⚑ Advised odds, may be hard to matchRecord period: Ledger to 25 Oct 2026 (all graded picks; start date not stated)Checked 2 Nov · next re-check 9 Nov44PerformanceLosing record100TransparencyHigh transparencyPostTime (Neigh I)
Stakes vary per pick (1–2 units), so there is no level-stakes record to score. On CLV (as published): “every pick, loser and non-runner is kept on a public record page with CLV against Betfair SP”
Record period: 25 Sep 2026 – page read 2 Nov 2026Checked 2 Nov · next re-check 9 NovLow dataPerformanceInsufficient data75TransparencyHigh transparencyMVP Capping
The headline record and the downloadable file disagree, so we can’t say which figures are complete. On CLV (as published): “Current model (since 24 Jul 2026): 4528-3252-13, −200.91u, −2.6% ROI, average CLV +0.9%”
Checked 1 Nov · next re-check 8 NovLow dataPerformanceInsufficient data65TransparencyHigh transparencyMysports.AI
Headline 79% is an out-of-sample backtest hit rate on high-confidence picks (N=2,780); the live figure shown is closing-line value on 262 bets. No live profit and loss at stated stakes found. On CLV (as published): “+6.62% CLV vs closing line”
Checked 3 Nov · next re-check 10 NovLow dataPerformanceInsufficient data20TransparencyLow transparencyPick1
Says every pick is logged with a timestamp and closing line, but we could not find totals (bet count, profit or ROI) on its public pages. On CLV (as published): “+3.8% Avg CLV”
Checked 1 Nov · next re-check 8 NovLow dataPerformanceInsufficient data10TransparencyLow transparencyAI Sports Tips
Its public track record page shows totals only (7 settled signals in the last 60 days when we checked) and lists no individual picks, odds or results without a login, so no profit or return can be worked out. On CLV (as published): “Home page: '6,593 settled signals' and '55.0% positive CLV'”
Checked 3 Nov · next re-check 10 NovLow dataPerformanceInsufficient data0TransparencyLow transparencyUnabated
Research and odds tool with no published picks or results record. On CLV (as published): “Odds screen across sportsbooks, exchanges and prediction markets, a vig-free 'Unabated Line', player-prop projections, simulators and calculators (CLV, hedge, no-vig, teaser), plus betting education and an odds API”
Checked 4 Nov · next re-check 11 NovLow dataPerformanceInsufficient data0TransparencyLow transparency
Questions and answers
What is a good closing line value?
Any consistent positive CLV over a large number of bets suggests you are taking better prices than the market settles at. How large it needs to be depends on the market and the margin; it should always be read alongside actual results.
Is CLV better than ROI?
They answer different questions. ROI is what you actually won or lost. CLV is an early signal of whether your prices were good. A credible record shows both.
Does a positive CLV mean a tipster is profitable?
No. It means the market moved towards their selections after they were tipped. Profit also depends on results, the prices you can actually get, and any fees.
How we check the records
The transparency checklist sets out what counts as evidence for each Transparency criterion, the same for every listing. The AI tipping transparency report 2026 shows how many of the AI tools and tipsters we rate meet each criterion.