How Closing Line Value Is Calculated: The 4 CLV Modes

Closing line value is calculated as the decimal odds you took multiplied by the closing win probability, minus 1. Take +120 on a game that closes at an implied 47.86% and your CLV is +5.30%. One formula, no ambiguity.

The ambiguity is in the second number. A closing market is not one price, it is one price per sportsbook, and reducing those into a single closing probability is a choice. In SmartStake that choice is a setting called your CLV Mode, and it means two bettors who took the identical price on the identical game can hold different CLV numbers.

Scoring One Bet Four Ways

Type a price, pick the book you took it at, and watch all four modes score the same bet against the same closing market at once. Change the book and notice how many of the four rows move.

Those numbers are not a reimplementation. The widget calls the same function the scoring job writes your stored CLV with, so what it shows is what the bet tracker would store.

The One CLV Formula

Every mode ends at the same arithmetic. In decimal odds:

$CLV = o_{taken} \times p_{close} - 1$

Here o is the decimal price you took and p is the win probability implied by the closing line. The result is an expected value percentage: how much better or worse your price was than the market's verdict at kickoff, per unit staked.

Working the widget's default through by hand, +120 is 2.20 in decimal (the odds converter does the conversion, and the odds conversion guide explains it). The de-vigged close implies 47.86%:

2.20 × 0.4786 − 1 = 0.0530 = +5.30%

This is the same expression as expected value, because that is exactly what it is. CLV is the expected value of the price you took, evaluated at the closing probability instead of the current one. If you want the concept rather than the calculation, what closing line value means is the place to start.

The Four Closing Lines

All four modes produce a p for the formula above. They differ in which prices they read and whether they take the sportsbook's margin out first.

Only two of the four remove vig. That single fact explains most of the spread between them, and it is why the modes are not four opinions of equal weight.

Under the widget's defaults, one bet at +120 scores like this. These figures are an illustrative example built on made-up closing prices for a fictional game, not live odds and not a typical result:

Same bet, same closing market, and 5.24 points of CLV between the highest and the lowest reading.

Why the Numbers Disagree

Two things separate the four: whether the vig comes out, and how many books get a vote.

A sportsbook's posted price carries its margin. Both sides of a two-way market add up to more than 100% of probability, so the price on your side alone implies a win chance slightly above the market's honest estimate. Scoring against that inflated probability counts the book's margin as if it were your edge. The de-vigging guide walks the arithmetic, and the de-vigging calculator runs it on any two-sided market.

That is why Respective Sportsbook reads highest here at +10.00%. It compares your +120 to DraftKings' own close of +100, vig included, at one book. It is a real question, just a narrow one.

Best Available is the harshest in this example at +4.76%, because it holds your price against the best number anyone could still get at kickoff, which was Caesars at +110. That is not a universal rule. When the best closing price on the board is worse than the fair price, Best Available reads higher than No-Vig instead.

The reference set matters too. Custom Book Selection over Pinnacle and Circa returns +5.65%, a little above the full-consensus +5.30%, because those two books had shaded slightly further toward your side than the field. Narrowing to the sharpest books is a defensible choice, and it is also a smaller sample.

The clearest demonstration is to change the book you bet at and watch what happens. Only the Respective row moves:

The other three modes hold steady, because they read the market rather than your account. And notice that betting at Caesars, the book that ended up with the best close, collapses Respective Sportsbook and Best Available into the same number. If your book is the best price, the two questions are the same question.

Choosing Your Mode

There is no mode that is right for everyone, but there is a reason the default is the default.

No-Vig Fair Odds is the honest baseline. It removes the margin and reads the whole market, so a positive number means you beat what the market genuinely thought, not what a book was charging. If you are using CLV as a leading indicator of whether your process holds up, this is the mode to leave it on.

Respective Sportsbook answers a narrower and more personal question: did you get in before your own book moved? That is useful for timing, and useless for comparing yourself to anyone else. Read it as a measure of when you bet, not of how good the price was.

Best Available is a self-audit on line shopping. A persistently negative reading here, alongside positive No-Vig CLV, is a specific and fixable diagnosis: your bet selection is doing its job and you are conceding value by not holding enough accounts to catch the best number.

Custom Book Selection exists for bettors whose market is not the default market. If Pinnacle is not licensed where you bet, as is the case in regulated Ontario, a consensus built from the books you can actually reach is a truer benchmark than one anchored on a book you cannot. The Canadian tools guide covers that market in more detail. The tradeoff is real: fewer books means a noisier fair line.

Re-Scoring Your History

The part that surprises people is what happens when you change the setting.

Capture and compute are separate. The closing market is captured once per market, as one immutable snapshot at kickoff, and everyone who bet that market shares it. Your CLV, by contrast, is derived from that snapshot under your mode and stored as one number on each bet, which is what lets the performance dashboard average thousands of bets quickly.

So changing your CLV Mode does not reinterpret a number on the fly. It clears the stored digit on every affected bet and re-scores your entire history against the new benchmark in the background. Your average CLV, your share of bets that beat the close, and the positive and negative split are all recomputed. Nothing is frozen at the value it had the day it was scored.

That is a deliberate choice rather than a limitation. Because the close is stored permanently instead of being read back from a short retention window, there is no need to lock a digit in place to preserve it, so your history can always be re-read under a new lens.

The mode is a per-user default, not something recorded on each bet. Free accounts are pinned to No-Vig Fair Odds; choosing another mode or a custom book set is a premium setting, and a saved choice survives a lapse and comes back on re-upgrade.

Coverage and the Blank Digit

A CLV number needs a captured close. When a bet's market has no closing snapshot, the digit renders as a dash rather than a zero, and the bet is left out of the average instead of dragging it down. That is a coverage boundary, not an error.

Some modes are pickier than others. Respective Sportsbook needs the close at your specific book, so a market that closed everywhere except there scores as uncovered. Custom Book Selection needs at least one of your chosen books to have priced it, which you can see in the widget above by clearing every book from the selection: the Custom row goes blank while the other three keep scoring.

De-vigged modes are the most forgiving, because any pair of books that priced both sides is enough for a fair line.

Finding the Setting

CLV Mode lives with the rest of your CLV settings and applies to your whole tracked history. The bet tracker captures the closes as your bets settle, and the performance dashboard reports the result. The CLV tracking update covers what shipped and when.

If your CLV is consistently negative under No-Vig Fair Odds, that is a bet selection signal rather than a settings problem, and the positive EV tool attacks it directly by surfacing prices that beat the sharp line while the market is still open. Beating the fair line now is what tends to show up as positive CLV later.

Track your closing line value free.

Frequently Asked Questions

How is closing line value calculated?

Closing line value is the decimal odds you took multiplied by the closing win probability, minus 1, expressed as a percentage. If you took +120 (2.20 in decimal) and the closing line implies a 47.86% chance, your CLV is 2.20 times 0.4786 minus 1, which is +5.30%. The formula never changes. What changes is where the closing win probability comes from, because a closing market has one price per sportsbook rather than a single number, and the way those prices are reduced to one probability is a setting called your CLV Mode.

What is a no vig closing line?

A no vig closing line is the closing price with the sportsbook's margin removed, worked out by de-vigging both sides of the market at once. A book's posted price on your side implies a win probability slightly higher than the market's honest estimate, because the margin is built into it. Removing the vig strips that padding out and leaves the fair probability, which is a stricter benchmark than the raw number on the board. SmartStake scores CLV against the de-vigged consensus by default for that reason.

Why does my CLV change when I change the CLV Mode?

Because the mode decides which closing line your price is scored against, and each mode resolves to a different closing probability. Your stored CLV is one number per bet derived under your current mode, so changing the mode clears those numbers and re-scores your whole history in the background against the new benchmark. The closing market itself never changes. It is captured once per market and shared by everyone who bet it.

Which CLV Mode is the most honest?

No-Vig Fair Odds is the most honest default, because it is the only mode that both removes the sportsbook margin and reads the whole market rather than one book. Respective Sportsbook usually returns the friendliest number, since it scores you against a single price with the vig still in it, so part of what it reports as edge is really the book's margin. Best Available is the strictest realistic benchmark, because it holds you to the best price anyone could still get at kickoff.

The Bottom Line

CLV is one formula and four possible benchmarks. The formula is your decimal odds times the closing win probability, minus 1. The benchmark is your CLV Mode, and it moves the answer by more than most bettors expect: the same +120 in the example above scores anywhere from +4.76% to +10.00% depending on which closing line it is measured against.

Leave it on No-Vig Fair Odds unless you have a specific reason not to, because it is the only setting that asks whether you beat the market's honest price rather than a marked-up one. Then read the other modes as diagnostics: Respective for your timing, Best Available for your line shopping, Custom when the default market is not your market.

CLV measures the quality of the price you got, not what you will make from it. A bet that beats the close can still lose, and a good CLV record is evidence that a process is sound rather than a forecast of profit.