A winning ticket can still be a bad bet—and a losing ticket can be well judged.
A bettor takes Team A at -120; by kickoff, the same wager is widely available at -105. Team A wins, so the ticket cashes—but the bettor paid more than the market ultimately required. Another bet loses after moving from +110 to -110. The result hurts, yet the earlier price was valuable.
That split matters because one game reveals little about decision quality. Injuries, turnovers, officiating, and ordinary randomness can decide a result without making the original wager wise or foolish. Profit measures what happened; closing line value helps assess the price taken. Over many bets, consistently beating the closing market is stronger evidence of sound selection than a brief winning streak. It does not guarantee profit, but it separates repeatable pricing decisions from short-term luck.
What closing line value measures
- Closing line value (CLV)
The difference between the odds or line originally taken and the market price near the event’s start. It measures whether a bettor secured better terms than late market participants.
- Price taken
The wager’s recorded odds or spread when placed. For spread bets, a basic understanding of point spreads helps distinguish a meaningful half-point move from a change in quoted odds.
- Closing line
The market’s final widely available price before betting stops. A representative market average is usually more reliable than one sportsbook’s last quote.
- Positive CLV
Terms that outperform the close—for example, taking -3 before the market moves to -4, or +120 before it falls to +105.
Consistently positive CLV suggests that a betting process identifies useful information before the broader market fully reflects it. That can validate research, timing, and price shopping across a meaningful sample.
It does not prove that any individual wager was correct. The closing market can be wrong, and even an excellent price can lose. CLV evaluates the quality of the terms secured—not the result of one event.
Choosing a reliable closing benchmark
A market rarely has one universal closing line. Sportsbooks may shade prices for their customer base, manage exposure differently, or update at different speeds. A low-limit book can also display a stale number after higher-liquidity markets have already moved.
Late information widens those differences. An injury report, lineup change, weather update, or unexpected delay may reach books at different times. Data providers also vary: one may record the scheduled start time, while another captures the final quote before betting is actually suspended.
For repeatable CLV tracking, use either one liquid, market-leading sportsbook or a consensus close built from several reputable books. A sharp-book benchmark is simple and stable; a consensus reduces the influence of one unusual or stale price. Switching benchmarks according to which result looks better defeats the measurement.
Set the timing rule in advance and apply it to every wager:
- Record both the bet timestamp and benchmark timestamp.
- Define the close as the last available price before the actual event start.
- Exclude or flag bets placed within a chosen buffer—such as one minute of the start—when timestamp precision is uncertain.
- Keep post-delay or reopened-market bets in a separate category.
Three ways to express CLV
Decimal-odds ratio
For decimal odds, a simple payout-based calculation is:
CLV = (odds taken ÷ closing odds) − 1
A wager placed at 2.10 that closes at 1.95 has CLV of 7.7%: (2.10 ÷ 1.95) − 1. Positive values mean the earlier ticket pays more than the closing price for the same stake.
Implied-probability change
Decimal odds convert to raw implied probability with 1 ÷ odds. Thus, 2.10 implies 47.62%, while 1.95 implies 51.28%—an increase of 3.66 percentage points, not 3.66%.
Raw probabilities include the sportsbook’s margin. For a two-outcome market, remove it by dividing each outcome’s raw probability by the sum of both raw probabilities:
Fair probability of A = (1 ÷ odds A) ÷ [(1 ÷ odds A) + (1 ÷ odds B)]
Comparing fair probability at bet time with fair probability at close gives a margin-aware measure. It requires prices for every outcome at both timestamps.
Spreads and totals
Point-based markets can be recorded directly. A favorite taken at −3 that closes −4 gained one point of CLV; over 47 closing at 49 gained two points. Translating that move into win probability requires a market- and sport-specific spread-to-probability conversion for CLV.
A one-point move has no universal percentage value. Its importance depends on scoring patterns, the line’s location, and key numbers; moving from −2.5 to −3 can matter more than moving from −5 to −5.5.
Reading CLV in real tickets
A moneyline that shortens
A $100 bet at +120 returns $220, including stake, so its decimal price is 2.20. If the market closes at +105, the closing decimal price is 2.05.
CLV = (2.20 ÷ 2.05) − 1 = 7.3%
The corresponding implied probabilities are 45.45% and 48.78%. The bettor secured a materially larger payout than the closing market offered. That is positive CLV whether the selection wins comfortably or never threatens.
A half-point on the spread
Consider an underdog taken at +3 (-110) that closes +2.5 (-110). Because the price remains -110, the comparison is straightforward: the ticket holds an extra half-point.
That half-point changes a three-point loss from a closing-line loss into a push. The ticket therefore has positive point-spread CLV, even if the team loses by ten and the wager loses.
Better number, worse juice
Suppose +3 (-110) later becomes +3.5 (-125). The spread improves, but the required break-even rate also rises:
- -110: 110 ÷ 210 = 52.38%
- -125: 125 ÷ 225 = 55.56%
Ignoring pushes for a simplified comparison, the extra half-point must increase win probability by more than 3.18 percentage points to justify the higher price. If it adds only two points, +3 at -110 is economically better; if it adds four, +3.5 at -125 is better.
Because key numbers and pushes affect that conversion, point movement alone cannot settle the comparison. A de-vigged probability estimate or same-book alternate spread is needed. The score after kickoff contributes nothing to that judgment.
Build a CLV ledger that stays comparable
- Capture the ticket immediately
Record the placement timestamp with time zone, market, selection or side, quoted line, odds, stake, and sportsbook. Use one row per distinct wager or leg.
- Save the market snapshots
Keep the raw opening price and the chosen benchmark close exactly as published, including line and odds. Also record the benchmark source and cutoff timestamp.
- Normalize without overwriting
Convert odds to one standard format—usually decimal odds—and calculate implied probability consistently. Store these values in separate columns so the original quotes remain available for audits.
- Calculate both aggregate views
Per-bet CLV gives every qualifying ticket equal weight. Stake-weighted CLV weights each ticket by the amount risked, preventing a small wager from counting as much as a large one.
- Apply fixed exception rules
Pushes may retain CLV because entry quality can still be compared. Voids should be excluded; partial cash-outs should be split by remaining exposure or clearly flagged; missing closing data should remain missing rather than estimated.
Use the same odds convention, benchmark source, and exception policy across the full sample.
A win, loss, or push does not change the price originally taken. Keep ticket outcome in a separate field, and preserve all raw values even when corrected or normalized columns are added. If a rule changes, document the effective date rather than silently recalculating older records.
Reading the pattern, not the streak
A CLV ledger becomes more informative when its rows are summarized several ways. No single statistic tells the full story:
- Beat-the-close rate: the percentage of bets with positive CLV. This shows consistency but treats tiny and substantial edges equally.
- Average CLV: the mean across all bets, using one normalized method. It captures magnitude but can be pulled upward by a few extreme values.
- Stake-weighted CLV: each result is weighted by amount risked. This reveals whether larger wagers received better prices, although one unusually large bet may dominate.
- Segmented results: splits by sport, market type, sportsbook, bet timing, or odds range. Persistent strength in one segment may identify a genuine process advantage—or a benchmark mismatch.
For example, a 56% beat-the-close rate with modest positive average CLV across several hundred bets is generally more persuasive than a spectacular average driven by three long shots. Review the median, sample size, and largest observations alongside every average.
Comparisons also need context. Main markets often close more efficiently than props or lower-liquidity events, while point spreads and moneylines require different CLV measures. Small samples should be treated as preliminary, especially after slicing the ledger into narrow segments.
Closing markets can reflect sharp information, but they still contain vig, limits, book-specific bias, and occasional late overreaction. Positive CLV is evidence of a sound process—not proof that every closing price is correct.
Three CLV myths that distort the numbers
A repeatable CLV review routine
- Freeze the rules
Use the same closing source, cutoff time, and treatment of pushes, voids, and partial settlements.
- Normalize every price
Convert odds to one common measure—preferably de-vigged implied probability—and preserve the raw ticket and close.
- Build a comparable sample
Separate sports, market types, bet types, and time periods; avoid conclusions from a small or heavily clustered set.
- Review several views
Compare median and average CLV, stake-weighted CLV, positive-CLV rate, and results by segment.
- Audit the surrounding process
Check benchmark quality, selection rules, available limits, execution timing, and realized profitability before changing strategy.
Schedule reviews at fixed sample or time intervals rather than after unusually good or bad runs.
CLV is most useful as a diagnostic, not a target to maximize at any cost. Persistent positive readings strengthen confidence in the betting process, but only when the benchmark is credible, the sample is comparable, and the bets remain practical and profitable.