DraftKings made $2.8 billion in revenue in 2025. A significant chunk of that came from parlay bettors who took four reasonable opinions and turned them into one bet that needed all four to hit. Three did. The fourth didn’t. The book kept everything.
That’s not a horror story. That’s the design.
A standard -110 straight bet carries roughly 4.5% house edge. That’s the vig, baked into the price, and it’s what the book earns on average every time you bet a side or a total. It’s a real edge, but it’s a thin one. Sharp bettors who hit 54% of their straight bets can overcome it and show a profit over time.
Parlays are a different machine. When you combine two -110 bets into a parlay, true odds would pay you out at around +260. Most books pay +264, which sounds like they’re giving you something extra. They’re not. The fair payout on a two-teamer at -110 is closer to +284. That gap, roughly 20 cents on the dollar, is where the book’s edge on parlays lives. Add a third leg and the gap widens. Add a fourth and you’re looking at a house edge somewhere between 15% and 20% depending on the book and the juice on each leg.
The vig doesn’t add across legs. It multiplies.
If your goal is long-term profitability, straight bets are the only logical vehicle. The math is cleaner. Your edge on an individual game, if you have one, stays intact. A bettor who genuinely hits 55% of their NFL sides at -110 over a full season makes money. That same bettor parlaying three of those plays together needs to hit 16.6% of their three-teamers to break even, which requires hitting roughly 55% on each leg independently, and that’s before the compounded vig eats into the payout.
Every leg you add is another opportunity for variance to kill a ticket that had real value on each individual game. You can be right three times and lose because the fourth game went to overtime and the wrong team scored last. The individual edges didn’t fail. The format did.
Bankroll management also works with straight bets in a way it simply can’t with parlays. Flat betting 1% to 2% of your bankroll per game is a legitimate long-term strategy. There’s no equivalent discipline for parlay betting because the bet sizing and payout structure don’t allow for the kind of gradual, compounding growth that serious bettors build over years.
Here’s the honest version of this conversation that most betting content skips. Not everyone betting on sports is trying to build a spreadsheet and beat the closing line. Some people have $50 for a Sunday of NFL games and want a shot at making it $400. Straight bets at $10 a game don’t produce that outcome. A five-team parlay at $20 might.
That’s not irrational. It’s a different goal. Recreational bettors who treat parlays as entertainment, who know they’re paying a premium for the thrill of sweat across multiple games, are making a reasonable choice with their own money. The mistake isn’t betting parlays. The mistake is betting parlays while believing you’re playing a skill game that rewards your research.
FanDuel reported in 2022 that same-game parlays accounted for a larger share of handle than any other single bet type among casual users [VERIFY: exact FanDuel same-game parlay handle percentage]. That number tells you everything about who parlays are designed for and who they’re designed by.
There is one corner of parlay betting where the expected value math flips, at least in theory. Correlated parlays.
A correlated parlay is one where two outcomes in the same game are statistically linked, meaning if one happens, the other becomes more likely. The most straightforward example is parlaying a team to win with the game going over the total. If a team wins by scoring a lot of points, the total is more likely to go over than if they won a 13-10 defensive slog. Those outcomes aren’t independent. They’re connected.
Standard parlay pricing assumes independence between legs. When outcomes are positively correlated, the true probability of both hitting is higher than the book’s pricing reflects. That gap is where the value lives.
Books know this. Most major sportsbooks have adjusted same-game parlay pricing to account for obvious correlations, particularly on high-volume markets like first-half result combined with game total, or quarterback passing yards combined with game script. The easy spots are largely gone at books like DraftKings and FanDuel, which run algorithm-generated same-game parlay pricing at scale.
But smaller books occasionally post same-game parlays without hand-checking every correlation. A backup quarterback starting unexpectedly, a weather shift that affects game script in a specific way, a divisional game with known tendencies toward a particular style of play. These spots exist. They require real work to find.
The dangerous middle ground is the bettor who has learned just enough about correlated parlays to feel like they have an edge, but not enough to actually price one correctly. Same-game parlay builders on major apps are designed to feel like research tools. They’re not. They’re engagement features. The book sets the price on every combination, and that price already accounts for the correlation the bettor thinks they discovered.
A same-game parlay that “makes sense narratively” is not the same as a same-game parlay with a mathematical edge. Patrick Mahomes throwing for 300 yards and the Chiefs covering a spread is a story. Whether the book has underpriced that combination relative to true probability is a calculation, and it requires knowing the true probability first.
Most bettors skip the calculation and trust the narrative. That’s precisely what the parlay product is built to encourage.
Straight bets are better for anyone who takes their results seriously. The edge is preserved, the math is transparent, and long-term profitability is at least theoretically achievable with enough skill and discipline. Parlays are fine for anyone who understands they’re trading expected value for variance and excitement, and who has budgeted accordingly for the difference.
The line between those two bettors is not about how much money they have or how many games they watch. It’s about what they’re trying to accomplish. Decide which one you are before you build the ticket.