How Same Game Parlays
Actually Work:
Correlation, Penalties & Hidden Juice

The Uncomfortable Truth About SGPs

Every sportsbook in the country is pushing same game parlays. DraftKings has SGP promos on the homepage. FanDuel runs SGP boosts every night. BetMGM has an entire tab dedicated to them.

Ask yourself: why would sportsbooks aggressively promote a product that's good for bettors?

They wouldn't. SGPs are the single most profitable product sportsbooks offer. The hold percentage on SGPs is estimated at 15-25% — compared to 4-5% on straight bets. That means for every $100 wagered on SGPs across the market, the house keeps $15-25. On straight bets, they keep $4-5.

That doesn't mean SGPs can never have value. It means you need to understand exactly where the juice is hidden and when the math actually works in your favor. Most bettors never look under the hood.

What Correlation Actually Means

Correlation is the reason SGPs exist as a separate product — and the reason they're so profitable for books.

The Basic Concept

Two events are correlated when the outcome of one makes the other more or less likely.

Positive correlation: Both outcomes move together. If one hits, the other is more likely to hit too.

Negative correlation: The outcomes move against each other. If one hits, the other is less likely.

No correlation: The outcomes are independent. One has no effect on the other.

Why This Matters for Parlays

A standard parlay assumes each leg is independent. The math multiplies the probabilities together.

But in a same game parlay, the legs are NOT independent. They're correlated — because they're happening in the same game, involving the same players, affected by the same game flow.

The Correlation Penalty

Here's what sportsbooks actually do when you build an SGP:

Step 1: Price Each Leg Independently

Each prop or spread gets its own odds — just like if you were betting them as straight bets. The starting point is the same market price you'd see on a single bet.

Step 2: Identify Correlated Legs

The book's algorithm scans your parlay for legs that move together. Player over points + team over total? Correlated. Player over rebounds + same player over minutes? Correlated. Two players on the same team both over points? Correlated.

Step 3: Apply the Correlation Penalty

The book reduces your payout to account for correlation. If the true probability is higher than what independent math suggests, they pay you less.

The problem: You can't see this adjustment. The book shows you a final odds number — say +450 — but you have no way to know whether that's fair or whether they over-penalized the correlation.

Step 4: Add Extra Juice

On top of the correlation penalty, the book adds additional juice. This is where the 15-25% hold comes from. You're getting hit twice: once for correlation, once for juice.

Common SGP Correlations

Positive Correlations (Legs That Move Together)

Player over points + Team over total: If the team scores a lot, the star probably contributed. These are positively correlated — both hit together more often than independent math suggests.

Player over assists + Teammate over points: Assists literally create made baskets. If the point guard dishes 10 assists, his teammates scored at least 20 points off those assists. Strong positive correlation.

Player over rebounds + Game over total: Higher game totals mean faster tempo — more possessions, more shots, more rebound opportunities for everyone. Positively correlated.

Player over minutes + Player over points: More time on court = more scoring opportunities. Very strong correlation — and books penalize this heavily.

Two teammates both over points: If the team is scoring, multiple players benefit. Positively correlated, especially for the top 2-3 scorers.

Negative Correlations (Legs That Oppose Each Other)

Player over points + Team under total: If the team scores fewer points overall, the star probably didn't go off. These fight each other. Negative correlation.

Player over points + Opponent spread (opponent winning): If the opponent is winning, your player's team is losing — which means blowout risk, reduced minutes, garbage time. Negatively correlated.

Two opposing players both over points: In a low-scoring game, it's hard for stars on both teams to go over. Weakly negatively correlated through game total.

Low/No Correlation

Player over points + Different player on opposing team over rebounds: These are largely independent. One player scoring doesn't meaningfully affect the other team's rebounding.

Player over steals + Teammate over assists: These are functionally independent stats driven by different game actions.

Where the Hidden Juice Lives

The "Obvious" SGP

The SGPs that sportsbooks promote — the ones with easy-to-understand legs that "make sense together" — are the ones with the most hidden juice.

The "Safe" Low-Alt SGP

Bettors love building SGPs with low alt lines — "easy" legs that feel safe.

This is the most common way recreational bettors lose money on SGPs. See our guide on NBA Prop Betting Mistakes — chasing "safe" parlays is mistake #7.

When SGPs Actually Have Value

SGPs aren't always -EV. There are specific situations where the math can work in your favor.

Situation 1: Book Under-Penalizes Correlation

Sometimes the book's correlation algorithm doesn't catch the full relationship between legs.

This happens most often when one underlying factor (injury, minutes change, rotation shift) drives multiple legs simultaneously. The book adjusts each leg's individual price but doesn't fully penalize the correlation between them.

This is exactly what double-double betting exploits — correlated stats driven by one underlying factor (minutes).

Situation 2: Boosted SGPs with Real Edge

Sportsbook boosts on SGPs are marketing. But sometimes the boost is large enough to overcome the built-in juice.

How to check:

1. Calculate the fair odds of the SGP accounting for correlation
2. Compare to the boosted price
3. If the boosted price is higher than your fair odds, there's value
4. If it's lower, the boost didn't overcome the juice

Reality: Most boosts are breakeven or slightly +EV. The book uses them to get you in the door, hoping you'll build un-boosted SGPs the rest of the week.

Situation 3: Two Legs Max with Strong Edge on Both

The correlation penalty and juice compound with each leg you add. A 2-leg SGP has less total juice than a 6-leg SGP.

If you have genuine edge on two props in the same game — your projection diverges meaningfully from the market on both — a 2-leg SGP is the least bad way to combine them. The correlation penalty is smaller with fewer legs, and the total juice is manageable.

The rule: Never build an SGP with more than 3 legs unless you have a specific mathematical reason. Every additional leg adds more juice and more correlation penalty.

SGPs vs. Straight Bets: The Math

Let's compare the expected cost of SGPs vs straight bets over 100 bets.

The takeaway: If you have edge on a prop, bet it straight. You're paying 4.5% juice instead of 15-25%. The only reason to SGP is when the correlation between legs creates value that exceeds the additional cost — and that's rare.

The Biggest Tell

Here's the tell that confirms everything above:

Sportsbooks never limit recreational SGP bettors.

Think about that. Books aggressively limit and ban winning straight bettors. They cut your props limits to $50 the moment they identify you as sharp. But recreational SGP bettors? Unlimited. Promoted. Encouraged.

If SGPs were beatable at scale, books would limit SGP bettors the same way they limit prop bettors. They don't. Because SGPs are structurally profitable for the house at a level that straight bets aren't.

That doesn't mean every SGP is -EV. It means the product is designed to extract maximum juice, and you need to be very selective about when you use it.

How to Evaluate an SGP

1. Price each leg independently. What would each prop pay as a straight bet? Use SwishLand's fair value projections as your baseline.
2. Identify the correlations. Which legs move together? Which oppose each other? How strong is the correlation?
3. Estimate fair correlated odds. For positively correlated legs, the true probability is HIGHER than independent math suggests — so fair odds are LOWER.
4. Compare to the SGP price. Is the book paying more or less than your fair correlated estimate? If more, there might be value. If less, you're getting crushed.
5. Ask: would straight bets be better? If you have edge on individual legs, straight bets pay you at 4.5% juice instead of 15-25%. The SGP needs to offer something straight bets can't to justify the cost.

Common SGP Mistakes

Mistake #1: Building SGPs for "Fun"

The error: "It's just a small bet for entertainment."

The problem: Small bets add up. If you're dropping $20 on SGPs three nights a week at 20% hold, that's $600+ per year in expected losses. That's not entertainment — that's a subscription to losing.

The fix: If you want entertainment, bet straight bets. Same excitement, 75% less juice.

Mistake #2: Adding Legs to "Boost" the Payout

The error: "I'll add one more leg to get from +300 to +500."

The problem: That extra leg adds correlation penalty and juice. The +500 looks better but your expected return is worse. You're paying for the illusion of a bigger payout.

The fix: Fewer legs = less juice = better expected value. Always.

Mistake #3: Parlaying Negatively Correlated Legs

The error: Player over points + Team under total.

The problem: These fight each other. If the team scores fewer points, the star probably didn't go off either. The probability of BOTH hitting is lower than independent math suggests — but the book might not give you better odds for the negative correlation.

The fix: Only parlay legs that move in the same direction. Positive correlation or no correlation — never negative.

Mistake #4: Using SGPs Instead of Straight Bets

The error: You have edge on a single prop but parlay it with other legs "to make it more interesting."

The problem: You diluted your edge. The prop you have edge on is now dependent on legs you might NOT have edge on. And you're paying 3-5x more juice.

The fix: If you have edge on one prop, bet it straight. Period. Don't contaminate good bets with bad ones.

Mistake #5: Not Checking Correlation Direction

The error: Assuming all legs in the same game are positively correlated.

The problem: Some same-game legs are negatively correlated, some are uncorrelated. Treating them all the same means you're mispricing your own SGP.

The fix: Think through the causal chain. Does one leg hitting make the other more likely (positive), less likely (negative), or have no effect (independent)?

Conclusion

Same game parlays are the most profitable product sportsbooks offer — for the house. The correlation penalty plus additional juice creates a 15-25% hold that's 3-5x worse than straight bets.

SGPs can have value when:

1. The book under-penalizes correlation between legs (especially when one factor like an injury drives multiple stats)
2. A boost is large enough to overcome the built-in juice
3. You limit to 2-3 legs where you have genuine edge on each

SGPs are -EV when:

1. You're stacking low-alt "safe" legs (juice on juice on juice)
2. You're adding legs for a bigger payout (each leg adds cost)
3. You're parlaying negatively correlated outcomes
4. You have edge on individual legs but parlay them instead of betting straight

The tell: Books never limit SGP bettors. That tells you everything about who's winning.

Use SGPs surgically — when the correlation math works in your favor. Otherwise, bet straight. Your bankroll will thank you.