How to Read a De-Vig Line and Spot a Soft Number

Every line a sportsbook posts has two things baked into it: an estimate of what’s actually going to happen, and a built-in margin that guarantees the book makes money regardless of the outcome. That margin is the vig. If you look at a line without stripping the vig back out, you’re not looking at the market’s real opinion – you’re looking at the market’s opinion plus a fee. De-vigging is the process of removing that fee so you can see the number underneath. Once you can do that, spotting a soft line – a price one book is offering that doesn’t match where the fair number actually sits – gets a lot easier.

What “de-vig” actually means

Start with the vig itself. On a standard two-way market, both sides’ implied probabilities add up to more than 100%. That extra percentage is the book’s cut – it’s what makes -110/-110 lines add up to roughly 4.55% over 100% instead of exactly 100%. A market that summed to exactly 100% would have no house edge at all.

De-vigging takes those two inflated implied probabilities and scales them back down so they sum to exactly 100%, proportionally. What’s left is the market’s best estimate of the true probability of each outcome, with the house’s cut removed. That de-vigged number is the closest thing to a “fair” price the market is telling you exists.

The formula

For a two-outcome market (moneyline, spread, or total), the process is three steps.

Step 1 – Convert each side’s American odds to implied probability.
For negative odds: -odds / (-odds + 100)
For positive odds: 100 / (odds + 100)

Step 2 – Add the two implied probabilities together. This sum will be greater than 100% – the amount over 100% is the vig.

Step 3 – Divide each individual implied probability by that sum. The two results now add up to exactly 100%, and each one is the de-vigged fair probability for that side.

Worked example

Take a moneyline: Lakers -150, Celtics +130.

Lakers implied probability: 150 / (150 + 100) = 60.0%
Celtics implied probability: 100 / (130 + 100) = 43.5%

Add them: 60.0% + 43.5% = 103.5%. That 3.5% over 100% is the vig on this specific line.

Now divide each side by 103.5%:
Lakers de-vigged: 60.0 / 103.5 = 57.97%
Celtics de-vigged: 43.5 / 103.5 = 42.03%

Those two numbers – 57.97% and 42.03% – are the market’s real, fee-free estimate of each team’s win probability. Everything above that on the Lakers side, or below it on the Celtics side, was the book’s margin, not its opinion.

Why the de-vigged number is the one that matters

Raw odds tell you what price you’d get. De-vigged odds tell you what the market actually believes. Those are two different things, and confusing them is one of the most common mistakes bettors make when they try to eyeball value.

A -150 favorite “feels” like a 60% shot. It isn’t – it’s a 57.97% shot once you remove the vig. If you’re comparing a book’s number against your own model, or against a fair-value benchmark like Kalshi pricing, you need to compare against the de-vigged number, not the raw one. Comparing your model’s probability to a vig-inflated line will make almost every bet look worse than it actually is, because you’re subtracting the house’s fee twice – once in the price you’d pay, and once again in your head when you eyeball it as “too short.”

This is also why win rate alone tells you so little and closing line value matters more – CLV is, at its core, a comparison between two de-vigged numbers at different points in time.

How to spot a soft line

A “soft” line is a price at one book that hasn’t caught up to where the rest of the market – or a sharp reference point – has already moved. Once you know how to de-vig, spotting one is a matter of comparison, not guesswork.

Pull the same game from three or four books. De-vig each one using the steps above. Most of the time, the de-vigged numbers will cluster tightly – within a point or two of each other, because books are all watching the same sharp money and adjusting in step. When one book’s de-vigged number sits noticeably apart from that cluster, you’ve found a soft number. That book hasn’t moved yet, or moved the wrong amount, or is carrying a stale line because their limits haven’t attracted enough action to force a correction.

The same comparison works even better against a market that isn’t a sportsbook at all. Kalshi prices sports outcomes through a real two-sided exchange with buyers and sellers setting the price directly, which means its implied probabilities carry a much smaller structural margin than a book’s -110 line built to guarantee house profit. That makes a de-vigged Kalshi price a cleaner fair-value anchor than averaging four books that are all quietly copying each other’s numbers. A sportsbook line that’s noticeably out of step with de-vigged Kalshi pricing is a stronger soft-line signal than a sportsbook line that’s merely out of step with other sportsbooks.

A quick gut check before you trust a “soft” number

Not every gap is a real opportunity. Check three things before you treat a de-vigged discrepancy as an edge:

Line freshness – has this book actually taken meaningful action on this game yet, or did the number just post? Early lines move fast and shouldn’t be judged the same way as a line that’s been live for hours.

Market depth – a soft number at a book with a $50 max isn’t the same opportunity as the same gap at a book that will take real size. Limits tell you how much the book itself trusts its own price.

Injury and lineup timing – a gap that appeared in the last ten minutes before lineups drop is usually news, not inefficiency. Confirm there isn’t a reason the number should be moving before you assume it’s simply behind.

The takeaway

Vig hides the market’s real opinion inside a price that’s built to guarantee the house a profit. De-vigging strips that fee back out with basic arithmetic, and once you’re comparing de-vigged numbers instead of raw odds, soft lines stop being a feeling and start being a calculation – a book’s number sitting apart from a cluster of de-vigged prices, or apart from a cleaner benchmark like de-vigged Kalshi pricing. That’s the same math running under CLV, under fair-value comparisons, and under every +EV alert worth trusting.

Automatehive Edge runs this exact de-vigging process against Kalshi-anchored fair value on every game, all day, so you don’t have to pull four books and do the arithmetic by hand: see it in practice at automatehive.net/edge.

Not betting advice. Bettors must be 21+. Bet responsibly — only wager what you can afford to lose. Gambling problem? Call 1-800-GAMBLER.