Closing line value, explained

Sportsbook lines move as money and information arrive, and the closing line — the number just before kickoff — is the market's most informed price. Closing line value asks: did the market move toward your pick after you made it? If you took Over 249.5 passing yards at −110 and the game closed Over 249.5 at −125, the market ended up agreeing the Over was more likely than the price you paid. That is positive CLV, and you had it before the game settled anything.

A season of NFL cards is roughly a hundred-something graded picks. At that sample size, win rate is mostly noise: a coin-flip bettor lands anywhere between 42% and 58% often enough that a “60% season” proves little. CLV converges much faster, because every single pick produces a measurement — how far the line moved — rather than a binary win/loss. Consistently beating the close over even 40–50 picks is hard to fake; a 60% win rate over 50 picks is not.

This is also why serious bettors treat sustained positive CLV as the working definition of edge: books limit players who beat the close, not players on hot streaks.

Prices contain the book's margin. Over 4.5 receptions at −115 with the Under at −105 does not imply a 53.5% chance of the Over — part of that price is vig. To compare entry and closing prices honestly, both sides get de-vigged first: convert both prices to implied probabilities, then scale them so they sum to 100%. CLV is the difference between the de-vigged closing probability and the de-vigged entry probability, in percentage points.

Worked example: entry Over 249.5 at −110/−110 de-vigs to 50.0%. The close at −125/+105 de-vigs to roughly 54.0% on the Over. CLV = +4.0 points. If the closing line lands on a different number than the entry, the comparison is modeled back to the entry number using a standard distribution for that market — the math is on the methodology page.

Market-beating props bettors sustain average CLV in the low single digits of probability — one to four points. Anyone advertising double-digit average CLV over a real sample is describing a measurement error or a fiction. Expect most individual picks to have small CLV either way; the average and the share of picks that beat the close are the numbers that matter.

Every graded pick on the ledger shows its entry number and price, closing number and price, and de-vigged CLV — including the losers. The sitewide average is published alongside the win rate, so you can judge the record by the stricter of the two numbers. That is the point of the site.