How the 10.2% NBA Sportsbook Hold Is Built

The number that explains everything about your losses

The single most useful statistic I learned in my first decade of NBA betting was the industry-wide sportsbook hold. I had been losing slowly for years, and I could not explain why. My picks felt reasonable. My record looked acceptable. The bankroll kept drifting down. Then a friend showed me the maths. The major US sportsbook operators run an average hold of around 10.2 percent across all sports betting products. That number is what every recreational punter is fighting against without realising it. Understanding how it is built, where it concentrates, and how to navigate it is the foundation of every honest betting practice.

The 10.2 percent figure is a blended industry average across products. The actual hold on a specific NBA bet varies dramatically by market type. Spreads run thinner. Parlays run much thicker. Prop markets sit somewhere in between. The variation is what most punters miss when they think about “the sportsbook edge.” There is no single edge. There is a portfolio of edges, each priced for a specific market, and the operator’s overall profitability comes from the mix of activity their customers actually undertake. This article walks through how the hold gets built, where it hits hardest, and what UK punters can actually do about it.

What “hold” actually measures

Sportsbook hold is the operator’s gross gambling yield as a percentage of total wagering activity. Stake £100 across many bets, settle the results, and on average across the industry the punter gets back £89.80 while the operator keeps £10.20. The hold is the operator’s net take across all activity, before their own operating costs, tax obligations, and marketing spend.

This is different from the margin on any individual bet. Margin is the gap between fair odds and posted odds on a specific market. A 5 percent margin on a single bet means you are paying 5 percent extra to make that bet. The hold is what the margin actually produces in operator revenue across all customers, all bets, all products. The relationship between margin and hold is mediated by which products customers actually use and how skilfully they bet.

The crucial insight is that hold is higher than the margin on any individual bet, because the hold is built across a portfolio. A customer who bets only single moneylines at 4 percent margin will give the operator around 4 percent hold on their activity. A customer who bets multi-leg parlays at 25 percent compounded margin will give the operator around 25 percent hold on that activity. The industry average of 10.2 percent reflects the mix — and the mix is heavily influenced by which products operators promote and which products customers find appealing.

The product-by-product breakdown

The most useful way to understand the 10.2 percent number is to decompose it into product categories.

Point spreads on NBA games typically carry margins of 4 to 5 percent at major operators. This is the tightest market because the volume is high, the competition between operators is fierce, and the sharp money keeps the line honest. Punters who bet exclusively NBA spreads at competitive operators face the structural minimum cost of the betting industry. The effective hold on this category sits around 4 to 5 percent, and skilled punters can beat it.

Totals (over/unders) run similar margins to spreads, around 4 to 6 percent at major operators. The pricing is slightly less efficient than spreads because the underlying inputs (pace, efficiency, lineups) involve more forecasting uncertainty. Skilled punters can beat this category but the margin is slightly wider than spread betting.

Moneylines on NBA games carry slightly wider margins than spreads, typically 5 to 7 percent. The operator builds in extra margin partly because of the binary outcome (which raises variance) and partly because the public tends to favour favourites, allowing the operator to charge a premium on the favoured side.

Player props on major players carry margins of 6 to 10 percent. The pricing is materially less efficient than main markets because the inputs are more uncertain and the volume on any single prop is much smaller. Operators charge more because their own forecasting risk is higher.

Parlays carry compounded margins that depend on the number of legs and the leg pricing. A four-leg parlay at 5 percent per leg has a compounded margin of roughly 18.5 percent. An eight-leg parlay at the same per-leg margin has a compounded margin of around 33.7 percent. Same-game parlays add correlation premiums on top of the compounded margin, often pushing the effective hold to 25 to 40 percent depending on the correlation structure.

Why the industry blended hold sits at 10.2 percent

The 10.2 percent blended hold across the industry is the weighted average of these per-product margins, weighted by the actual volume of customer activity in each category. The arithmetic reveals where the operator’s profits actually come from.

If main markets — spreads, totals, moneylines — accounted for 100 percent of customer activity, the industry hold would sit around 5 percent. The 10.2 percent figure reflects the fact that a substantial share of activity is in higher-margin products. Player props, parlays, same-game parlays, and other recreational-friendly products generate a disproportionate share of operator profit even though they are a smaller share of total wagered volume.

This is why operator marketing focuses so heavily on parlays and prop products. The maths is direct. A customer who bets £1,000 in spreads through a season gives the operator roughly £50 in revenue. A customer who bets £1,000 in four-leg parlays gives the operator roughly £185 in revenue. The customer behaviour that produces the higher-margin activity is what operators incentivise through promotional structures, free bet rewards, and product positioning.

The 2025 trend in the US market has been a measurable rise in industry hold, driven by faster growth in parlay and prop volume than in main-market volume. The shift is structural — operators have built better parlay products, more prop options, and more sophisticated cross-selling between sports. The customer mix is moving toward higher-margin products. The blended hold has crept up year on year as a result.

Where the hold hides in NBA pricing

Knowing the industry hold figure is useful only if you can read it in the prices on your screen. The mechanics of how hold gets hidden are worth understanding directly.

The most obvious hiding place is in the juice on close lines. A spread bet at 1.91/1.91 has an implied probability of 52.4 percent on each side, totalling 104.8 percent. The 4.8 percent excess is the operator’s margin. A spread bet at 1.85/1.95 has implied probabilities of 54.1 percent and 51.3 percent, totalling 105.4 percent. The 5.4 percent excess is the operator’s margin on that asymmetric line. The price differences between operators are partly competitive positioning and partly inventory management, but the underlying maths is always the same — both sides’ implied probabilities sum to more than 100 percent.

A second hiding place is in long-tailed prop markets. A prop market on a relatively niche stat (made three-pointers for a role player, for example) might be priced at 1.80/1.90 or even wider. The implied probabilities sum to over 108 percent on these markets, meaning the operator margin is over 8 percent. Customers who casually bet niche props are paying significantly more per bet than customers who bet main markets, often without realising it.

A third hiding place is in parlay correlation pricing. Same-game parlays explicitly price the correlation between their legs into the combined price. The “boosted” parlay prices that operators promote are often boosted from a baseline that already includes the correlation premium plus the compounded margin. The headline boost looks generous but the underlying price is still substantially above fair value. Reading parlay prices honestly means decomposing them into their legs, identifying the implied correlation, and comparing to the leg-independent fair price.

The disciplined punter’s response

The practical takeaway from understanding sportsbook hold is structural, not tactical. You cannot eliminate the hold. It is the price of admission to the betting market. What you can do is bet predominantly in the categories with the thinnest margins and avoid the categories with the thickest ones.

The cleanest betting practice for a UK NBA punter is built around main markets — spreads and totals on individual games, with the occasional moneyline. These markets carry the structural minimum hold and allow skilled punters to have positive expected value in the long run. Adding modest player prop activity is reasonable if you have specific analytical edges. Heavy parlay activity is structurally a losing strategy regardless of skill level. Same-game parlays should be treated as entertainment betting, sized accordingly, and never relied on as a profit centre.

The CLV measurement framework is the long-run feedback loop that tells you whether your activity is beating the hold. Tracking closing line value across a season reveals whether your prices on average are better than the market’s eventual settled price. Positive CLV on main markets, sustained over hundreds of bets, demonstrates that you are extracting value from the operator faster than the hold extracts it from you. Without that measurement discipline, you cannot tell whether you are beating the hold or being beaten by it.

The deeper truth is that the 10.2 percent industry hold figure is the structural baseline you and every other punter are fighting against. The operators are not your adversary in any moral sense — they are running a business with a structural margin, just like a casino, a stockbroker, or an insurance company. The skill is to participate in their market on terms that work for both sides. Bet the markets where the hold is thinnest. Track your CLV honestly. Size your stakes to bankroll preservation rather than to dopamine. The maths is unforgiving but it is also knowable. Knowing it is the start of every betting practice that survives more than a few seasons. The closing line value workflow is the operational discipline that turns knowledge of the hold into measurement of your own performance against it — and the two together are the only honest scoreboard a serious NBA punter has.

Created by the "NBA Stats For Betting" editorial team.