New: OddsHype is now on Google PlayGet the Android app →App Store: coming soon
OddsHypePricingStart free
☰

Same-Game Parlays and Correlation: How SGPs Are Priced

How sportsbooks price same-game parlays, what correlation means, why SGP margins are so high, and how to judge whether an SGP or SGP boost has any value.

By the OddsHype team · · 3 min read
In this guide
  1. What correlation means
  2. How books price SGPs
  3. Why SGPs carry big margins
  4. How to evaluate an SGP
  5. SGP boosts
  6. Playing SGPs responsibly
  7. Frequently asked questions

Same-game parlays (SGPs) combine several bets from one game into a single ticket. They are the fastest-growing bet type in the US and among the most profitable for sportsbooks. To understand why, you need to understand correlation.

What correlation means

Two outcomes are correlated when one happening makes the other more (or less) likely.

  • Positive: a team winning big and its running back going over his rushing yards
  • Positive: a QB's passing yards over and his top receiver's yards over
  • Negative: a game total under and a star's points over

In a normal (multi-game) parlay, legs are independent, so the parlay probability is the product of the leg probabilities. In an SGP, they are not.

How books price SGPs

Because correlated legs are more likely to hit together, books cannot simply multiply odds. Each book runs a correlation model (often from a data provider) to estimate the joint probability, then adds margin, and on SGPs the margin is usually large: commonly 20–35% of the stake across a ticket.

Example: QB over 249.5 passing yards (fair 50%) and WR1 over 74.5 receiving yards (fair 50%).

  • Independent probability: 25% (fair +300)
  • With positive correlation, the true joint probability might be 31% (fair +223)
  • A book might offer +240, below the independent +300 but above the true fair +223

In that example, the SGP is slightly +EV. More often, the book's correlation estimate plus margin leaves you well below fair.

Why SGPs carry big margins

  • Each leg already includes margin (props often 5–9%)
  • Correlation models are conservative on purpose
  • Bettors are less price-sensitive on SGPs, so books charge more
  • Longer SGPs compound everything (Parlay math)

How to evaluate an SGP

  1. Start with fair probabilities for each leg (devig sharp prop lines). (How to devig odds.)
  2. Estimate correlation honestly. Direction is easy; size is hard.
  3. Compare the SGP price with the fair joint probability.
  4. Compare across books: different books use different correlation models, so the same SGP can be priced very differently. This is one of the few places SGP value shows up.

SGP boosts

A boost on an SGP adds value, but if the SGP margin is 30%, a 25% profit boost may still leave it −EV. Boosts on straight bets are usually easier wins. (Odds boosts · Profit boosts.)

Playing SGPs responsibly

  • Treat them as entertainment-sized bets
  • Keep legs few (2–3)
  • Shop the same SGP across books
  • Track SGPs separately from straight bets (How to track sports bets)

Frequently asked questions

Are same-game parlays a good bet?

Usually not, because books charge large margins on them. Occasionally a book underprices a correlation, especially when compared with other books.

How do sportsbooks calculate same-game parlay odds?

They estimate the joint probability of all legs with a correlation model and then add margin, rather than simply multiplying each leg's odds.

Why won't my sportsbook let me combine certain bets?

Some combinations are so highly correlated (or contradictory) that books block them rather than price them.

Put this to work automatically
OddsHype compares every legal sportsbook in your state, finds +EV bets, arbitrage and props, and tracks every bet you place. Free account, no card. 21+.
Start free

Educational content, not betting advice. No strategy guarantees a profit; past results do not predict future results. Bet only where it is legal for you and only what you can afford to lose. Gambling problem? Call or text 1-800-GAMBLER.