How Does Sportsbook Hold Percentage Work?
Every sportsbook prices every market to win over time — and the mechanism it uses is hold (also called vig, juice, or overround). It's the cut the book builds into both sides of a line, baked in so the book earns revenue regardless of which side wins. This guide walks through what hold actually is, why books price to a house edge in the first place, and what a real-world two-book comparison looks like — so you can read hold the way sharp bettors do.
Definition: what hold, vig, and juice mean
Hold is the sportsbook's edge, baked into both sides of a market. The math is straightforward: convert American odds on each side to implied probability, sum them, and the total always exceeds 100 %. The gap between that sum and a fair 100 % is the hold on the market.
Take a simple moneyline. A favorite priced at –150 implies a 60 % win probability (100 ÷ 250), and an underdog at +130 implies a 43.5 % win probability (100 ÷ 230). Together the two sides sum to 103.5 % — that 3.5 % gap is the hold.
On NFL spreads, hold shows up as the price (or “juice”) next to the spread. A typical spread prices both sides at –110, meaning a bettor risks $110 to win $100 on either side. Each side carries an implied probability of 52.38 %, summing to 104.76 % — roughly a 4.76 % hold baked into the market.
The names are interchangeable: hold is the industry term, vigorish (vig) is the older term still common among bookmakers, juice is recreational shorthand, and overround is the technical term for the implied-probability gap. They all describe the same idea — the book's edge, loaded into the price on both sides.
Why every sportsbook prices to a house edge
Sportsbooks aren't in the business of picking winners — they're in the business of balancing liability and collecting the built-in edge on every wagered dollar. A perfectly balanced book earns the hold on the loser's stake no matter which side wins the event. Hold is the sportsbook's revenue model, the same way a casino's house edge is its revenue model on blackjack.
The size of the hold varies for two reasons. First, market efficiency: highly efficient markets — NFL sides and totals, NBA moneylines, major soccer leagues — price tight (typically 2–4 % hold), because sharp money is plentiful and books compete on price to attract it. Less efficient markets — niche props, smaller leagues, novelty markets — run wider holds (6–10 %+), because sharp price discovery is rarer and books have more pricing power.
Second, book positioning: sharp-focused books like Pinnacle and Circa deliberately run low hold (~2–3 %) because the sharp action they attract helps them price the rest of the market. Recreational books like FanDuel and BetMGM price wider (~4.5–5 %) because they're optimizing for revenue from casual bettors who are less likely to shop line to line.
For bettors, hold is the structural drag on every dollar wagered. Even with perfect information about which team wins, you'd lose the hold to the book over the long run. That's why every edge you build has to clear the hold before it becomes a real profit.
Worked example: NFL Week 6, Chiefs at Bengals, two books
Both books list the same spread — Chiefs –3 — but they price it differently:
Pinnacle: Chiefs –3 (–105) / Bengals +3 (–105)
FanDuel: Chiefs –3 (–115) / Bengals +3 (–105)
Implied hold: Pinnacle ≈ 2.4 % · FanDuel ≈ 4.7 %
Convert to implied probability. Pinnacle's –105 on either side implies 51.22 % on each; both sides sum to 102.44 %, a ~2.4 % hold. FanDuel's Chiefs –115 implies 53.49 %; dog side +3 (–105) implies 51.22 %; total 104.71 %, a ~4.7 % hold, with the extra cost loaded onto the favorite side.
On a $100 payout target, the per-side stake shows the gap:
Pinnacle Chiefs –3 (–105): stake $105 to win $100
FanDuel Chiefs –3 (–115): stake $115 to win $100
Required win rate to break even: 51.2 % vs. 53.5 %
Same number on the board, different price, different hurdle. A bettor picking the Chiefs at Pinnacle needs to win ~51.2 % of the time to break even on the spread; at FanDuel, the breakeven rate is ~53.5 %. The ~9.5 % gap in required stake is entirely a hold difference, with no change in the underlying line.
The pattern repeats at every book, every day: spread, total, moneyline. Comparing prices for the same number across books is what turns line-shopping from a slogan into a measurable edge.
Odds above are illustrative and not from a live feed. Used for educational demonstration only.
What hold means for bettors
Hold is what makes the rest of sports-betting math worth doing. Closing Line Value is meaningful only because holding a lower-hold line is cheaper to clear; line shopping — comparing prices across books for the same event — is meaningful only because holds differ from book to book. The simplest version of line shopping is exactly the two-book comparison above: same number, different price, different long-run cost.
Sharp books' lower holds compound over hundreds of bets. A consistent 2 % hold gap between your baseline book and a recreational alternative is often the difference between an edge that shows up in your P&L and one that gets quietly absorbed by the book. Bettors who aren't already using a low-hold book as their baseline aren't seeing the full value of whatever edge they think they have.
For a single bet, the gap is small. For a season's worth of wagers, it's the difference between tracking your edge accurately and bleeding it slowly to the book.