What the vig really costs you

Ask a casual bettor what the bookmaker's cut is and you will usually hear "about 5%". Ask what they need to hit to break even and you will usually hear "50-something percent". Both answers are wrong in the same direction, and the gap between the real numbers and the assumed ones is roughly where a losing year comes from.

Where the vig hides

A fair coin-flip market would be priced at even money on both sides — +100 and +100, or 50c and 50c. What you actually get is −110 and −110, or 52c and 52c. Convert those to implied probabilities and they add up to more than 100%:

That extra 4.76% is the overround — the book's margin on the market. It is not a fee you pay separately; it is built into every price you see, on both sides, whichever one you take.

Stripping it out: the no-vig price

To recover what the book actually thinks, divide each implied probability by the total. In the example above: 52.38 / 104.76 = 50.0% each, which is the honest number. This is the *no-vig* or *fair* price, and it is the only version worth comparing your own estimate against.

On an uneven market it works the same way. A −250 / +200 moneyline implies 71.4% and 33.3%, totalling 104.7%. De-vigged, that is 68.2% and 31.8% — so the book's real view of the favourite is 68%, not the 71% the price appears to be quoting.

What it does to break-even

At −110, you risk $110 to win $100. Break-even is 110 / 210 = 52.38%, not 50%. That 2.38 points is what you are giving away before you have had a single opinion, and it compounds over a season:

Look at the first row for a moment. Being *exactly right* — genuinely picking a 50/50 correctly half the time — loses nearly five thousand dollars. This is the entire business model, and it is why "I win about half my bets" is a description of losing money.

Prediction markets are cheaper, not free

Polymarket and Kalshi do not charge vig the way a sportsbook does, which is a genuine structural advantage. But there is still a cost, and it still comes off your edge:

• The spread. If the book is 54c bid / 56c ask, taking the ask costs you a point against mid immediately.
• Slippage. The top of the book might only be a few hundred dollars deep. Size beyond it and your average fill is worse than the price you clicked.
• Fees. Check the current schedule for the venue you are on — they change, and they apply to winnings on some venues rather than to stake.
• Getting out. If you might close before resolution, you pay the spread twice. Budget for the round trip, not the entry.

A realistic all-in round-trip cost on a liquid prediction market is often under a point, versus four-and-a-bit on a −110 sportsbook line. That difference is real and it is worth caring about — it roughly halves the hit rate you need to be above water. It is not zero.

The practical rule

Never compare your estimate to a raw price, and never call a gap an edge until you have taken the cost out of it. A 2-point disagreement with a sportsbook is not a bet — after vig it is a loss. The same 2 points on a tight prediction market might be a small one.