Understanding the Vig: What Bookmakers Don't Tell You

Walk into almost any sportsbook and you'll see a market like this: Team A at -110, Team B at -110. It looks symmetric, even fair. It isn't. That symmetric-looking pair of numbers is actually the bookmaker's fee, built directly into the price.

Where the vig comes from

If a coin flip were priced with true 50/50 odds, both sides would be +100 — bet $100, win $100. Instead, a standard two-way market prices both sides at -110, meaning you have to risk $110 to win $100 on either side. The bookmaker isn't taking a real position on the outcome; they're collecting a toll on the transaction itself.

Calculating it directly

Convert each side to implied probability (see our EV walkthrough for the formula), then add them together.

For a -110 / -110 market:

• Each side implies 110/210 ≈ 52.38%
• Total: 104.76%

That extra 4.76% over 100% is the vig on this specific market. It's also sometimes called the "overround." A wider gap between the two prices — say -120 / +100 — usually means a larger built-in margin, though the exact split between the two sides also shifts based on which way the market expects money to flow.

Why the number is usually understated in people's heads

Bettors often think of a -110 line as "roughly even money with a small fee." The actual math says something sharper: to break even at -110 on both sides of a coin-flip market, you need to win 52.38% of your bets, not 50%. That 2.38-point gap compounds fast over a large sample — it's the difference between a long-run winning and losing record even if your actual read on games is genuinely a coin flip.

Why this matters for evaluating any "edge"

Any claimed statistical edge has to clear the vig before it's real. A model that's right 51% of the time on -110 lines isn't profitable — it's still below the 52.38% breakeven point. This is the first sanity check worth applying to any anomaly or "value" claim, including the ones on this site: does the calculated edge account for the actual priced-in margin, or is it being measured against a naive 50/50 baseline? Our EV engine strips the vig out explicitly for exactly this reason — comparing a model's probability against the implied probability, not an assumed even line.

The takeaway

The vig isn't hidden, exactly — it's sitting right there in the odds. But almost nobody calculates it directly, which is precisely why it's worth doing once by hand. Once you've seen the arithmetic, you'll never look at a -110 line the same way again.