The legs you can actually combine on an NFL builder
Most UK operators publish a list of permitted leg types for their builder, and the lists are similar in structure but differ in detail. The standard menu includes the main spread, main total, main moneyline, halftime spread and halftime total, alt-spread lines (typically in three-point increments), alt-total lines, race-to-X-points markets (10, 15, 20), team totals, player passing yards over/under at multiple lines, player rushing yards over/under, player receiving yards over/under, anytime touchdown scorer for any eligible scorer, first scoring play type, and a handful of game-flow specials.
Around 76 percent of UK gamblers aged 18-24 use mobile phones as their primary gambling access method – the highest of any age group – and the design of the builder product reflects that. The mobile flow on Sky Bet and bet365 lets you tap legs onto the slip as you scroll the fixture page, with the combined price refreshing in the slip footer in real time. Add a leg, the price updates. Remove a leg, the price updates. The friction is minimal, which is precisely why the product has scaled so well in the UK market.
The restrictions are where the operators differ. Some firms permit the main moneyline plus the main spread on the same ticket; others restrict the combination because the two outcomes are almost perfectly correlated and the discount required to price it fairly would be unattractive on the slip. Some firms permit two different player props from the same player; others restrict the combination because the underlying correlation is high enough that pricing it cleanly would expose the bookmaker to too much risk. The deeper menu lives under the prop pages – my guide to NFL prop bets in the UK covers the broader specials catalogue if you want the full picture of what the menu actually contains.
Correlation pricing, and why it matters more than the headline number
The correlation adjustment is the heart of the bet builder product, and getting it right or wrong determines whether the punter has value. The trouble is that the bookmaker’s correlation model is invisible – you cannot read off the screen what discount has been applied. You see the combined price, and you have to work backwards.
The way I check is straight multiplication. Take the individual decimal prices of every leg, multiply them, and compare the result to the offered combined price. If the offered price is meaningfully shorter than straight multiplication, the bookmaker has applied a substantial correlation discount – usually because the legs are positively correlated. If the offered price is roughly equal to straight multiplication, the bookmaker is treating the legs as independent, which often means the legs are weakly correlated or the bookmaker has not built a correlation model for that specific combination. If the offered price is meaningfully longer than straight multiplication, the bookmaker has applied a negative correlation adjustment.
The cases where punters find value are the legs that the operator’s correlation model misses. Two player props on the same QB – passing yards over and completions over – are positively correlated, but some operators do not link the two legs in their model and price the combination as independent. The straight multiplication price is sometimes longer than the true correlated value, which leaves the punter with a structural edge. These mispricing opportunities are rare and shrink as the operators improve their models, but they exist.
The trap is the opposite case. Two legs that the operator’s model assumes are positively correlated but which are actually weakly correlated in reality. The discount applied to the builder is generous to the bookmaker, the offered price is shorter than fair, and the punter overpays for the correlation that does not actually exist. These cases are hard to spot from outside, but they are common in builder structures involving moneyline plus first-scorer markets, where the correlation assumption is much weaker than the model treats it.
Builder versus the equivalent straight accumulator
The same set of legs can sometimes be expressed as a straight cross-game accumulator and a same-game builder, and the comparison is instructive. A four-leg builder on a single Sunday fixture might combine the Bengals spread, the total under, an anytime touchdown for Joe Mixon and a passing yards over for Joe Burrow. Each of those legs could, in theory, be picked from four different games and stacked as a straight accumulator instead.
The straight cross-game accumulator at four legs carries roughly the same overround as a four-leg straight accumulator generally – compounded individual leg vig, no correlation adjustment, total margin around 18 to 22 percent. The same legs as a same-game builder carry an overround structure that depends on the bookmaker’s correlation model. If the legs are positively correlated, the discount the operator applies reduces the combined price relative to straight multiplication, but it almost never reduces it to the level where the effective overround drops below the cross-game equivalent.
The reason punters often prefer builders is psychological rather than mathematical. Watching a single game with four positions riding on the same fixture is more engaging than watching four games with one position in each. The shared narrative – same teams, same broadcast, same forty-minute window – concentrates attention and amplifies the experience. The bookmaker prices the product accordingly, charging a meaningful margin premium for the experience.
For value-focused punters, the implication is that builders should be used selectively rather than reflexively. A builder makes sense when the legs are negatively correlated and the bookmaker has not modelled the correlation correctly, or when the legs are mildly positively correlated and the operator’s discount happens to be conservative. Builders should not be used as a substitute for straight accumulators where the legs would be better placed as independent positions across different games.
Cashing out a builder, and when the offered price is actually generous
Cashout on a same-game builder works structurally like cashout on any other multi-leg ticket. The bookmaker offers a current price to settle the bet immediately based on which legs have already settled and the current live odds on the remaining legs. The punter accepts or declines. If they accept, the bet settles at the offered cashout value and the remaining legs become irrelevant.
The mechanics during a single NFL game create cashout opportunities that do not exist on cross-game accumulators. Approximately 290 million online bets on real events are placed monthly in the UK, and a meaningful slice of that volume flows through cashout offers during live games. On a same-game builder, the cashout offer updates after every meaningful play – every score, every turnover, every quarter end – and the value of the offered cashout depends on which specific legs have already locked in versus which remain live.
The interesting case is when the early legs of the builder have hit and the late legs still need outcomes. A four-leg builder where the spread leg and an early-scoring touchdown have already settled correctly, but the total over and a passing-yards over still need to hit, offers cashout values that depend heavily on the live total line and the live passing-yards projection. The cashout price sometimes overweights the difficulty of the remaining legs and offers values structurally close to the full winning return, which makes early acceptance attractive even if the punter believes the remaining legs are likely to hit.
The opposite case is the early-leg loss. A builder where the first leg has already failed shows a cashout value close to zero, because the ticket cannot win regardless of what happens with the remaining legs. Some operators offer “partial cashout” structures that let the punter keep some legs alive while cashing others, but the structure is unusual on builders and more common on cross-game accumulators with acca-insurance overlays.