How we think about sportsbook pricing.

1. Why price matters

A betting decision is usually described as a choice between outcomes: this team or that one, over or under. But the outcome is only half of it. The price attached to that outcome determines how much you're risking to win a given amount, and that price is what determines whether the bet is favorable at all. Two bettors can pick the exact same winner and still make two very different decisions, because the price they got was different.

2. Implied probability

Sportsbook odds are a way of expressing a probability. A moneyline of -150 implies the sportsbook thinks that side wins somewhat more often than not; +130 implies the opposite. Converting odds into an implied probability puts every price, regardless of format, on the same scale, which makes it possible to compare prices and estimate value in a consistent way.

3. Vig and de-vigging

If you add up the implied probabilities of both sides of a typical sportsbook market, the total is usually a bit more than 100%. That extra percentage is the vig, or margin: the built-in cushion that lets a sportsbook operate profitably regardless of the outcome. It also means the raw implied probability on either side is not a clean read of the true chance of that outcome.

De-vigging is the process of removing that margin, rebalancing both sides back to a true 100%, so what's left is a cleaner estimate of the market's actual view. This concept is standard across the industry. It is not unique to FVB. What we build on top of it is.

4. Market consensus

One sportsbook's de-vigged price is still just one book's opinion, shaped by its own customer base, risk limits, and how quickly it updates relative to the rest of the market. Looking at several sportsbooks' prices together gives a broader read of where the market as a whole is settling.

Consensus is useful information, not a guarantee of correctness. Markets can be slow to react to new information, and a consensus built from a handful of books can still miss something a sharper, faster-moving book has already priced in.

5. The Fair Value Model

The Fair Value Model is FVB's proprietary framework for turning sportsbook pricing data into a usable estimate of fair value. At a high level, it gathers pricing across supported sportsbooks, removes each book's margin, and combines the results using FVB's own weighting and modeling approach into a single fair probability for each side of a market.

We don't publish the exact weighting formulas or internal implementation details. That's the proprietary part: not the underlying concepts of de-vigging or consensus, which are standard in the industry, but FVB's specific methodology for aggregating and weighting that data into a usable output.

6. Fair odds and fair probability

Fair probability is the Fair Value Model's estimate of how likely an outcome is, based on the market data it evaluates. Fair odds are that same estimate expressed as a price, as if it were being offered with no sportsbook margin attached. Neither is a prediction of what will happen. Both are a reference point: what a bet should roughly cost in a margin-free market, so you can judge whether a specific sportsbook's price is better or worse than that baseline.

7. Expected value

Expected value, or EV%, compares a specific sportsbook's price against the model's fair probability. A positive EV% means that price pays more than the fair baseline suggests it should. A negative EV% means the opposite.

Positive EV describes the price, not the outcome. A positive-EV bet can still lose, and a negative-EV bet can still win. Over many decisions made at favorable prices, the math favors better long-run results, but no individual bet is decided by expected value. It's decided by what actually happens in the game.

8. Sportsbook comparison

Because sportsbooks price the same market independently, the exact price available to you can differ meaningfully from book to book. Two people betting the identical outcome, at the same time, can end up with different expected value simply because one had access to a better number. This is why comparing prices across sportsbooks, not just picking a side, is part of evaluating a bet properly.

9. Research and context

The Fair Value Model evaluates price. It does not evaluate who's injured, which lineup is starting, or how a team has looked over its last five games. That's what matchup and player research are for. A favorable price and strong supporting context are different signals, and using both gives a fuller picture than either alone.

10. Limitations

• Model outputs are estimates built from available market data, not certainties.
• Probabilities describe likelihood, not guarantees. Even a well-supported estimate can be wrong on any single event.
• Markets change. Prices move as new information, including injuries and lineup news, reaches the market.
• A consensus built from current sportsbook pricing can lag real-world developments.
• Past performance, including historical model results, does not guarantee future outcomes.