Why Parlays Lose
A parlay is the most fun bet on the board and the worst one for your bankroll. The reason isn't bad luck — it's built into the math. We know because we built parlay finders, ran them on real games, and then deleted the whole feature. Here's what the data showed.
A parlay has to win every leg
A single bet wins or loses on its own. A parlay only pays if every leg hits — miss one and the whole ticket is dead. That sounds obvious, but it's the entire problem. Stringing legs together doesn't add their chances; it multiplies them, and multiplying numbers below 1 makes them shrink fast.
Say you find three bets you genuinely like, each a coin-flip-ish 52% to win. Individually that's fine. Stacked into a parlay:
| Parlay | Each leg | Chance all hit | Reality |
|---|---|---|---|
| 2 legs | 52% | 27.0% | roughly 1 in 4 |
| 3 legs | 52% | 14.1% | roughly 1 in 7 |
| 4 legs | 52% | 7.3% | roughly 1 in 14 |
Three solid 52% bets become a 14% shot. That's not a knock on the legs — it's just what multiplication does. And the books know it, which brings us to the part that actually empties your wallet.
You pay the vig on every leg
Every betting price has the sportsbook's cut baked in — the vig (or juice). A standard -110 line carries about a 4.5% hold. The catch with a parlay is that the vig compounds across legs, just like the probabilities do. So a parlay isn't paying you fairly for the longer odds — it's quietly charging you the house edge again and again:
That's why parlays are the most profitable product on a sportsbook's menu and why they're pushed so hard. The payout number is the bait; the compounded hold is the hook. A four-leg parlay can carry four times the house edge of betting those same four games as singles.
We tested it on real games — then cut the feature
We don't run on theory, so we built it. Lakeshore Edge had several parlay finders that combined our model's calibrated single-bet probabilities into "safe" parlays, cashable tickets, and custom builds. We ran them on live slates and graded every one against the result.
The finding was brutal and consistent. The legs the finders selected won only about 21% of the time, versus about 52% for our straight single bets. The reason is subtle but important: when you optimize for a bigger combined payout, the math pulls you toward longshots and away from the boring near-coin-flips that actually carry our edge. Multiply a handful of imperfect-but-calibrated probabilities together and one soft leg sinks the ticket.
| What we bet | Leg / pick win rate | Verdict |
|---|---|---|
| Single bets | ~52% | kept — roughly break-even with strong closing-line value |
| "Safe" parlay finder | ~21% (legs) | cut — deeply negative ROI |
| Custom parlay builder | ~21% (legs) | cut — worst performer of all |
We tried tightening the leg quality and cash-probability floors four separate times. It never moved the needle — multiplying calibrated-but-imperfect probabilities is unforgiving by nature. So in 2026 we removed all six parlay finders and kept only single bets. We'd rather delete a fun feature than leave a structurally losing one on the page.
What to do instead
- Bet the legs you like as singles. Same games, lower combined house edge, and you can win on some even when others lose.
- Shop the line. Getting -105 instead of -110 on the same bet is free expected value — and it isn't compounded away.
- Track your closing line value, not your parlay dreams. Beating the closing price is the only signal that you actually have an edge.
- If you parlay for fun, size it like entertainment. Small, occasional, and with eyes open about the built-in house edge — not as a bankroll strategy.
FAQ
Why do parlays lose money?
They need every leg to win and they multiply the sportsbook's cut across all of them. A single -110 bet holds ~4.5%; a four-leg parlay holds ~19%. One soft leg out of four kills the whole ticket, so parlays punish any imperfect pick far harder than singles do.
Is a parlay better than betting singles?
No. The same selections as separate singles have a lower combined house edge, let you cash some legs when others miss, and swing far less. The large payout is the marketing; the compounded vig is the cost.
Are parlays ever +EV?
Only if every leg is independently positive expected value, which is hard to find even one at a time. Recreationally a small parlay is fine if you accept the higher house edge. As a money-making strategy, singles win.
What did your own parlay test show?
The legs our finders picked won ~21% of the time vs ~52% for our singles, because chasing payout selects longshots. The tickets posted deeply negative ROI, so we removed all six finders and kept only single bets.