Home / Strategy / Parlay Construction
Smart parlay construction
Parlays aren't inherently a mistake. Building them without thinking about correlation and sizing usually is.
On this page
Why parlays are appealing, and what that appeal costs
A parlay combines multiple individual bets ("legs") into a single wager that only pays out if every leg wins, in exchange for odds that multiply together into a much bigger potential payout than any single leg offered alone. A three-leg parlay of bets each priced around -110 can return roughly six times your stake instead of the roughly 1.9x a single -110 bet pays - which is exactly why parlays are marketed so heavily. The appeal is real. So is the cost: each additional leg multiplies not just the potential payout but the sportsbook's effective edge on the combined bet, and it multiplies the chance that the whole thing fails, since every single leg has to hit.
The math sportsbooks don't emphasize
If each leg of a parlay independently has, say, a 50% chance of winning, a two-leg parlay wins 25% of the time, a three-leg parlay wins 12.5% of the time, and a five-leg parlay wins just over 3% of the time - probabilities fall fast as legs are added, even though it doesn't always feel that way looking at the growing payout number. Because each leg individually carries the sportsbook's standard vig, and that vig compounds across every leg, the effective house edge on a multi-leg parlay is meaningfully higher than the edge on any one of its legs bet straight. This is simple, well-understood math, and it's the entire reason parlays are among the most profitable bet types for sportsbooks on a percentage basis.
None of this means parlays are "bad" to bet - it means the big number on the screen already has the extra house edge baked into it, and it's worth knowing that going in.
Correlated vs. uncorrelated legs
The math above assumes each leg is independent of the others, which is often a poor assumption in practice - and understanding when it's a poor assumption is where actual parlay skill lives. If two legs are positively correlated (a team's win and their star player going over a passing-yardage prop, for instance, since a team likely to win by a comfortable margin often has that player performing well too), the real combined probability of both hitting is higher than treating them as independent would suggest, which can make an otherwise-unappealing parlay quietly reasonable. If legs are negatively correlated or simply unrelated (a coin-flip NFL game and an unrelated tennis match), the independent-probability math applies more cleanly, and there's no hidden correlation boost to rely on.
Sportsbooks are aware of correlation too, and same-game parlay pricing tools generally adjust odds down when they detect meaningfully correlated legs specifically to prevent the payout from overstating the real probability - so "the app let me combine them at these odds" isn't proof the correlation was ignored in the pricing.
Same-game parlays specifically
Same-game parlays (SGPs), which combine multiple bets within a single game, are the fastest-growing bet type at nearly every major U.S. sportsbook, largely because they let a bettor express a detailed opinion about how a specific game will unfold. They're also where correlation matters most and is hardest to eyeball correctly - a bettor combining "Team A wins," "Team A's quarterback over 250 passing yards," and "game goes over the total" is stacking three legs that are all more likely to hit together than independently, which the book's own pricing has usually already partially accounted for. Treat an SGP's advertised payout as a starting point for evaluation, not confirmation that the bet is a good idea.
Sizing a parlay responsibly
Because parlays are higher-variance than straight bets by design - a single missed leg out of five wastes the entire stake - most disciplined bettors size parlay wagers smaller than they would a straight bet with similar conviction, often treating them explicitly as a smaller, separate part of a betting budget rather than a primary strategy. Our bankroll management guide covers unit sizing in more depth; the short version for parlays specifically is: size them like the long-shot bets they are, not like your core wagers.
When a parlay is a reasonable choice
Parlays make the most sense as a small, deliberate part of a betting budget rather than a primary strategy: when you have a specific, considered view on genuinely correlated outcomes, when the stake size reflects the real, lower probability of winning, and when you're treating the bet as entertainment with asymmetric upside rather than as an expected source of steady profit. They make the least sense when they're built by throwing together a handful of favorites you feel confident about individually without considering how the combined odds compare to the combined true probability, which is the most common way recreational bettors overpay for parlays without realizing it.
FAQ
Are parlays a bad bet type in general?
Not inherently - they carry a mathematically higher house edge than straight bets on average, but that doesn't make every parlay a mistake, especially when correlation is genuinely favorable and stakes are sized appropriately.
Do same-game parlay odds already account for correlation?
Sportsbooks generally adjust same-game parlay pricing to account for known correlations between legs, but that doesn't guarantee the adjustment is generous to the bettor - the book still builds in its standard margin on top.
How many legs is "too many" in a parlay?
There's no fixed number, but each additional leg meaningfully lowers the win probability and raises the effective house edge, so longer parlays should generally be sized smaller and viewed more as a lottery-style bet than a core wager.