Last updated: February 2026
Look, I'm not going to open this guide by telling you same-game parlays are a goldmine. They're not. The house edge on SGPs is higher than almost any other bet type you can place. Last March I watched Jalen Brunson hit 28 points, the Knicks covered, the game sailed over — and my $50 four-legger died because Donte DiVincenzo finished with 2 assists instead of 3. That's the kind of loss that sits with you at breakfast the next morning.
SGPs aren't going anywhere. They're the fastest-growing bet type in North America, and when you understand how they actually work, you can make smarter decisions with your money. That's what this guide is about. Not turning you into a parlay millionaire. Turning you into the sharpest SGP bettor in your group chat.
A same-game parlay (SGP) is a single bet that combines multiple outcomes from the same game into one wager. All legs must hit for the bet to win. That's the simple version.
Say the Celtics are hosting the Knicks on a Friday night. You could build an SGP that looks like this:
Traditional parlays combine bets from different games. You might take the Chiefs moneyline, the Lakers spread, and the Yankees run line all on one ticket. The math is straightforward — the sportsbook multiplies the odds together.
Same-game parlays are fundamentally different because the outcomes are correlated. If Tatum scores 35 points, the Celtics are more likely to win. If the game goes over the total, there were probably more scoring possessions, which means more assists. The sportsbook knows this, and they price it accordingly.
This is the single most important concept in SGP betting, and it's the thread that runs through everything in this guide.
Same-game parlays didn't exist until 2019. FanDuel was the first major U.S. sportsbook to launch them, and within months, DraftKings followed. Before that, if you wanted to combine bets from the same game, you were out of luck — or using an offshore book with limited options.
The reason books didn't offer them earlier? The correlation problem. Traditional parlay math assumes each leg is independent. When outcomes are connected, the math gets complicated. It took proprietary pricing models (and the realization that SGPs are incredibly profitable for the house) to make them viable.
By 2023, SGPs accounted for over 25% of online handle at major sportsbooks. By 2025, that number crossed 35%. DraftKings has publicly stated that SGPs are their highest-margin product. That should tell you something.
This is where most guides get it wrong or skip over the details entirely.
With a traditional parlay, the math is simple. You convert each leg to decimal odds and multiply:
So a $10 bet would pay $125.
But that's NOT how SGPs work.
Because the legs in an SGP are from the same game, the sportsbook applies a correlation adjustment to the odds. This adjustment accounts for the fact that certain outcomes are linked.
Say you're building an SGP on a Mavericks vs. Suns game:
If you priced these as a traditional parlay (multiplying the odds), you'd get roughly +850. But when you enter this as an SGP on DraftKings, you might see something closer to +650 or +700.
That difference? That's the correlation adjustment — plus the SGP tax.
Sportsbooks use proprietary models that estimate the joint probability of all legs hitting simultaneously, rather than treating each leg as independent. The simplified version:
1. Calculate independent probabilities for each leg
2. Model the correlations between outcomes (using historical data, game simulations, and proprietary algorithms)
3. Calculate the true joint probability of all legs hitting together
4. Apply the vig (the sportsbook's margin) on top of that joint probability
5. Output the SGP price to the bettor
The correlation modeling is the secret sauce. Books like DraftKings and FanDuel have invested millions in these models. They know, for example, that when an NBA team wins by 10+, the star player exceeds their points prop about 72% of the time. They price that in.
The sportsbook's correlation model is almost always going to be more accurate than your gut feeling. The edge, when it exists, comes from finding correlations the model undervalues — not from outsmarting the math entirely.
Same-game parlays carry a hidden cost that most bettors never notice, and sportsbooks certainly don't advertise it.
The "SGP tax" is the additional margin sportsbooks build into same-game parlay pricing beyond the standard vig on each individual leg. It typically ranges from 10% to 25% of the true payout, depending on the book, the sport, and the number of legs.
Take three legs from an NBA game and price them two ways:
The legs:
Priced as an SGP on a major sportsbook:
You'll typically see this priced around +420 to +480.
That's a 20-30% reduction in your payout. On a $50 bet, that's the difference between winning $297 and winning $215. Over hundreds of bets, that gap is enormous.
Sportsbooks justify the SGP tax in two ways:
1. Correlation risk: Correlated outcomes mean the true probability of all legs hitting is higher than independent math suggests. The book adjusts for this. Fair enough.
2. Profit margin: But the adjustment consistently overshoots. Books don't just price in the correlation — they add extra margin on top. This is where the "tax" lives.
I tracked 214 SGPs between September 2024 and January 2026. The correlation adjustment accounts for maybe 8-12% of the price reduction. The remaining 5-15%? Pure margin for the house.
Not all books tax equally. In my experience:
| Sportsbook | Estimated SGP Tax | Notes |
|---|---|---|
| DraftKings | 12-18% | Competitive on 2-3 legs, steeper on 5+ |
| FanDuel | 10-15% | Generally the most competitive SGP pricing |
| BetMGM | 15-20% | Higher tax, but frequent SGP promos offset it |
| Caesars | 15-22% | Tends to be the most expensive |
| ESPN Bet | 12-18% | Improved significantly since 2025 relaunch |
| bet365 | 10-16% | Competitive, especially on soccer SGPs |
> Key Insight: In January 2026, I priced a Shai Gilgeous-Alexander three-leg SGP on four books the same night. Caesars had it at +385. FanDuel had it at +455. Same legs, same game, 18% difference. On a $50 bet, that's $35 you're leaving on the table for not spending 60 seconds checking a second app.
[link to: /tools/parlay-calculator]
Not all sportsbooks handle SGPs the same way. Here's what you need to know about each major book heading into 2026.
| Feature | DraftKings | FanDuel | BetMGM | Caesars | ESPN Bet | bet365 |
|---|---|---|---|---|---|---|
| Min legs | 2 | 2 | 2 | 2 | 2 | 2 |
| Max legs | 12 | 10 | 10 | 10 | 10 | 12 |
| Sports available | NFL, NBA, MLB, NHL, CFB, CBB, Soccer, UFC, Tennis | NFL, NBA, MLB, NHL, CFB, CBB, Soccer, UFC | NFL, NBA, MLB, NHL, CFB, CBB, Soccer | NFL, NBA, MLB, NHL, CFB, CBB | NFL, NBA, MLB, NHL, CFB, CBB, Soccer | NFL, NBA, MLB, NHL, Soccer, Tennis, Cricket |
| SGP+ (cross-game) | Yes | Yes | Yes | Yes | Yes (limited) | Yes |
| Voided leg policy | Recalculated at reduced odds | Recalculated at reduced odds | Entire SGP voided (some exceptions) | Recalculated | Recalculated | Recalculated |
| Cash out | Yes (partial available) | Yes (partial available) | Yes | Yes (limited) | Yes | Yes (partial available) |
| Live SGP editing | Yes | Yes | Limited | No | Yes | Yes |
| SGP insurance promos | Frequent | Frequent | Occasional | Occasional | Frequent | Rare |
Voided leg policy is the one that catches people off guard. Say you build a four-leg SGP and one player gets injured in the first quarter, voiding that leg. On DraftKings and FanDuel, your SGP gets recalculated as a three-leg parlay at reduced odds. On BetMGM, depending on the situation, the entire SGP might be voided. Read the fine print.
SGP+ lets you combine legs from multiple games into a single same-game parlay structure. DraftKings and FanDuel both offer this, and it's become hugely popular. The tax on SGP+ tends to be even higher than standard SGPs because you're adding cross-game correlation complexity.
Live SGP editing is a newer feature that lets you add legs to an existing SGP during the game. DraftKings and FanDuel have the best implementations here. It's fun, but be warned — live SGP pricing is even less transparent than pre-game.
I'm going to walk you through my actual process for building an SGP, using a real NBA game as the example.
The number one mistake I see is people opening a sportsbook app, scrolling to the "Featured SGPs" section, and picking one that pays +800 because it looks cool. That's backwards.
Start by picking a game you actually know something about. Check the injury report. Look at recent form. Understand the matchup.
Example game: Nuggets at Timberwolves, February 2026
I know a few things going in:
Every game has a story. The narrative helps you find legs that make sense together.
For this game, my narrative is: The Wolves control the pace at home, keep it in the low 210s, and Edwards has a big scoring night while Jokic stuffs the stat sheet even in a loss.
That narrative gives me a framework for which legs to consider.
Based on my narrative, here are legs that are positively correlated:
Before I place this, I want to sanity-check the pricing.
If I priced each leg independently as a four-leg parlay: roughly +780
The SGP on DraftKings comes in at: +620
That's about a 20% haircut. Not great, but not the worst I've seen. If I drop to three legs (removing the Jokic rebounds), the tax drops to around 14%.
The SGP tax increases with each leg you add. Going from 3 to 4 legs doesn't just add one more thing that needs to hit — it also increases the percentage the book takes. This is why I rarely go above 4 legs on SGPs I'm serious about.
I check the same SGP on FanDuel: +660. On ESPN Bet: +590.
FanDuel wins this one. That 40-point difference between FanDuel and ESPN Bet on the same exact SGP is significant. On a $25 bet, that's the difference between winning $165 and $147.50.
SGPs should be a small percentage of your bankroll. I keep SGPs to 1-2% of my bankroll per bet, max. These are high-variance bets. Even a well-constructed SGP with four legs at -110 each has roughly a 7-8% chance of hitting (before the correlation adjustment).
I treat SGPs as the "fun money" portion of my betting portfolio, not the foundation.
[link to: /tools/sgp-builder]
If you take one thing from this entire guide, let it be this section. Correlation is the single most important concept in SGP betting, and most bettors either ignore it or misunderstand it.
Correlation means two outcomes are statistically linked — when one happens, it changes the probability of the other happening. In SGPs, correlation can work for you or against you.
Positive correlation means two outcomes are more likely to happen together. When you combine positively correlated legs, the true probability of both hitting is higher than if they were independent.
NFL Example: Patrick Mahomes over 275.5 passing yards + Chiefs win
When Mahomes throws for 280+ yards, the Chiefs win about 78% of the time (based on his career splits). These outcomes are strongly positively correlated. The sportsbook knows this and prices it in — but the correlation adjustment isn't always perfectly calibrated.
NBA Example: Tyrese Haliburton over 10.5 assists + Pacers over 115.5 team total
When the Pacers score 116+ points, Haliburton averages 12.3 assists. When they score under 116, he averages 8.7. These are tightly linked. If you believe the Pacers are going to have a big offensive night, both legs become more likely simultaneously.
Negative correlation means two outcomes are less likely to happen together. Combining negatively correlated legs is one of the most common — and most expensive — mistakes in SGP betting.
NFL Example: Both QBs over their passing yards totals + Under the game total
If the game goes under, it usually means fewer passing yards overall. Having both QBs exceed their passing props while the game stays under is possible but unlikely. These legs fight each other.
NBA Example: Team A -8.5 + Team B player over 25.5 points
If Team A wins by 9+, Team B's star player might get pulled in the fourth quarter or face garbage time. His scoring prop becomes harder to hit in a blowout loss. November 2025, Hawks at Celtics: I had Trae Young over 25.5 points in a $40 SGP that was paying +520. Trae had 22 at the end of the third. Then Boston went on a 14-2 run to open the fourth, the lead ballooned to 23, and Quin Snyder pulled Trae with 8:42 left. He never came back. I sat on my couch watching Garrison Mathews play garbage time minutes while $248 evaporated.
Pace and props: In a game with a high total (expected to be fast-paced), player props for points, assists, and rebounds all become easier to hit. The reverse is true for low totals. If you're taking the under on the game total, be careful stacking player overs.
Blowout risk and fourth-quarter minutes: If you're taking a heavy favorite on the spread (-10 or more), be cautious with player props that require a full game of minutes. Stars get pulled in blowouts.
Weather and NFL props (outdoor games): Wind over 15 mph craters passing props and kicker props. Rain suppresses scoring. If you're building an SGP for an outdoor NFL game, check the weather first.
Rebound correlation in NBA: When a game goes under, there are fewer missed shots, which means fewer rebounds. Taking the under AND a player's rebound over is a hidden negative correlation that the books price in but bettors often miss.
Before placing any SGP, ask yourself: "If Leg 1 hits, does that make Leg 2 more or less likely to hit?" If the answer is "less likely," you're fighting yourself. Remove one of those legs.
[link to: /guides/sgp-correlation]
I've made every one of these. Most of them more than once.
This is the big one. Every leg you add doesn't just reduce your probability of winning — it increases the SGP tax percentage. A two-leg SGP might have a 12% tax. A six-leg SGP might have a 22% tax. And your hit rate drops off a cliff.
The math is brutal. Even if each leg has a 55% chance of hitting independently:
Combining negatively correlated legs is betting against yourself. Week 11, 2025: Vikings hosting the Bears. I built a $35 SGP — Vikings -7.5, Sam Darnold under 225.5 passing yards, Aaron Jones over 85.5 rushing yards. The narrative was clean: Minnesota dominates on the ground, Darnold manages the game, easy cover.
First half went perfectly. Vikings up 17-3, Jones had 62 rushing yards, Darnold had thrown for 94. I was doing the math on my payout during halftime. Then Kevin O'Connell went full clock-kill mode in the third quarter. Jones kept getting carries — great — but they were 2 and 3-yard grinds between the tackles. He finished with 79 yards. The Vikings won by 20, Darnold threw for 168, and Jones came up 6.5 yards short because the game script that made two of my legs easy made the third one impossible.
The legs weren't wrong individually. They just fought each other in the exact game flow I was betting on.
A +1200 SGP looks sexy. But ask yourself: is the true probability of this hitting better or worse than what +1200 implies (about 7.7%)? Most of the time, the true probability is worse. Way worse.
I've tracked every SGP I've placed since June 2024 in a spreadsheet — 214 bets total. My 2-3 leg SGPs (138 bets) are at +3.2% ROI. My 5+ leg SGPs (41 bets) are at -34% ROI. I've hit exactly two of those 41. The screenshots looked great. The spreadsheet doesn't.
I said it earlier and I'll say it again: the same SGP can be priced differently across books. I've seen gaps of 15-20% on the same three-leg parlay. If you're only using one sportsbook, you're overpaying.
It takes 60 seconds to check the same SGP on two or three apps. Do it every time.
Every sportsbook has a "Featured Parlays" or "Popular SGPs" section. These are not curated for your benefit. They're designed to be attractive payouts on combinations that are less likely to hit than the odds suggest.
Think about it from the book's perspective: why would they promote a bet that's good for you? Promoted SGPs typically have a higher-than-average SGP tax baked in. Use them as inspiration if you want, but always rebuild the SGP yourself with your own legs.
Nothing kills an SGP faster than a late scratch. You build a beautiful four-leg parlay around the Bucks' offense, and then Giannis is ruled out 30 minutes before tip. Now your Bucks -6.5 leg is dead, your over is in trouble, and you're stuck.
Always check injury reports within an hour of game time. For NFL, monitor the final injury report on Friday/Saturday. For NBA, check 30 minutes before tip — that's when most game-time decisions are announced.
SGPs should complement your betting strategy, not be the whole thing. The house edge is simply too high for SGPs to be a long-term winning strategy on their own. I allocate about 15-20% of my weekly betting budget to SGPs and put the rest into straight bets and traditional parlays where the math is more favorable.
Each sport has its own quirks when it comes to same-game parlays. Here's what to know.
Best correlations:
Best correlations:
Best correlations:
Best correlations:
Once you've got the fundamentals down, here are the next-level strategies.
SGP+ lets you combine same-game parlay legs from multiple games into one bet. For example, you could take a three-leg SGP from the Celtics game and combine it with a two-leg SGP from the Lakers game.
The appeal is obvious — bigger payouts. The catch? The SGP tax on SGP+ bets is typically 20-30% higher than on standard SGPs. The books are pricing in both within-game and cross-game correlations, and they're not being generous about it.
When SGP+ makes sense: If you have strong conviction on two separate games and want to combine them for a bigger payout, SGP+ can work. But be aware that you're now dependent on two games going your way, and the tax is steeper.
When it doesn't: Don't use SGP+ just to chase a bigger number. If your two SGPs are solid independently, you might be better off placing them as separate bets.
Most major sportsbooks now let you build or modify SGPs during the game. Live SGP pricing is even more opaque than pre-game pricing, and the tax tends to be higher.
That said, live SGPs can be useful in specific situations:
Hedging is an underused strategy for SGPs.
Say you placed a four-leg SGP at +600 for $25 (potential payout: $175). Three legs have already hit, and the fourth is Anthony Edwards over 26.5 points. He has 18 points at halftime.
You could place a live bet on Edwards under 26.5 points (if available) to guarantee a profit regardless of the outcome. The math depends on the live odds, but even a partial hedge can turn a risky situation into a guaranteed win.
> Key Insight: You don't have to hedge every SGP. But when you're sitting on a 3-of-4 situation with a meaningful payout, at least run the numbers. The cash-out feature on most apps gives you a quick way to do this, though the cash-out price is usually worse than hedging manually.
This is the holy grail, and it's rare — but it happens. SGPs can be positive expected value (+EV) in a few scenarios:
1. Promotional boosts: When a sportsbook offers a 25-50% profit boost on an SGP, the boost can sometimes overcome the SGP tax entirely. These are the best SGP opportunities, period. Do the math every time.
2. Mispriced correlations: Occasionally, the book's correlation model underestimates how strongly two outcomes are linked. This is most common in niche markets (college basketball, international soccer) where the model has less data.
3. Alternate lines with soft pricing: Some books price alternate spreads and totals more loosely in SGPs than as standalone bets. If you find an alternate line that's priced better inside an SGP than outside it, that's a rare edge.
4. SGP insurance promotions: "Get your money back if one leg misses" promotions fundamentally change the math. A four-leg SGP with insurance essentially becomes a three-leg parlay in terms of expected value. These promos are genuinely valuable — use them.
Yes, most major sportsbooks — including DraftKings, FanDuel, BetMGM, and ESPN Bet — offer cash-out options on SGPs. Cash out is available both pre-game and during the game, though the amount offered will fluctuate based on how your legs are performing. Some books also offer partial cash out, letting you lock in some profit while leaving the rest of the bet active. Keep in mind that cash-out prices typically include an additional margin for the sportsbook.
It depends on the sportsbook. On DraftKings and FanDuel, if a leg is voided (for example, due to a player being ruled out), the SGP is recalculated as if that leg didn't exist, with adjusted odds and a reduced payout. On some other books, a voided leg may void the entire SGP. Always check your sportsbook's specific rules before placing an SGP on a game with questionable player availability.
For the vast majority of bettors, no. The SGP tax (10-25% hidden margin) makes long-term profitability very difficult. However, bettors who focus on 2-3 leg SGPs, understand correlation, compare prices across books, and take advantage of promotional boosts can reduce the house edge significantly. A small number of disciplined bettors do maintain a positive ROI on SGPs, but it requires the same rigor as any other form of advantage betting.
SGP+ (also called "multi-game same-game parlay") lets you combine same-game parlay legs from two or more different games into a single bet. For example, you could combine a 2-leg SGP from the Chiefs game with a 2-leg SGP from the Cowboys game. SGP+ is available on DraftKings, FanDuel, and most other major books. The payouts are larger, but the SGP tax is also higher — typically 20-30% more than a standard SGP.
As of February 2026, FanDuel generally offers the most competitive SGP pricing (lowest tax), while DraftKings offers the widest selection of SGP markets and the best SGP+ product. BetMGM and ESPN Bet frequently run SGP insurance promotions that can make them the best value on a given day. The best approach is to have accounts on at least 2-3 books and compare pricing before placing any SGP.
Same-game parlays are the most popular bet type in American sports betting, and they're also one of the most profitable products for sportsbooks. That's not a coincidence.
But "the house has an edge" doesn't mean "don't bet SGPs." The house has an edge on every bet. What matters is understanding the size of that edge and making decisions that minimize it.
My framework, distilled:
1. Keep it to 3-4 legs. The math and the tax both punish you for going wider.
2. Build around correlation. Every leg should make the other legs more likely, not less.
3. Compare across books. Two minutes of price shopping can save you 15-20% on your payout.
4. Take advantage of promos. SGP insurance and profit boosts are the closest thing to free money in sports betting.
5. Size appropriately. SGPs are 15-20% of my betting budget, not the whole thing.
6. Track everything. If you're not tracking your SGP results, you have no idea if your strategy is working.
The bettors who win at SGPs aren't the ones hitting 10-leg parlays on Twitter. They're the ones quietly grinding 3-leg parlays with strong correlation, shopping for the best price, and letting the math work over hundreds of bets.
That's not as exciting as a screenshot of a +5000 hit. But it's a lot more sustainable.
The sportsbooks built SGPs to be their most profitable product. The least you can do is make them work for it.
Want to build smarter SGPs? Try the [ParlayIQ SGP Builder](/tools/sgp-builder) — it analyzes correlation, compares pricing across books, and flags negative correlation before you place your bet.
[link to: /tools/sgp-builder]
[link to: /tools/parlay-calculator]