NBA Same-Game Parlay Picks Guide: Scripts, Correlation & When SGPs Make Sense

A same-game parlay looks like free money on the app. Three legs from one game, a price north of +500 and a story that sounds airtight. Then the backup center grabs 14 boards, the game turns into a blowout by the third quarter and the whole ticket dies on a leg you threw in for juice.

This guide is about building NBA SGPs on purpose. Not "four overs and a prayer." We'll cover how books price these tickets, what a real game script looks like, which legs actually move together and when the smart play is a straight bet instead.

Affiliate disclosure: NBA Prop Hub may earn a commission if you sign up through links on this site. That never changes our analysis or the odds a book shows you.

Gambling problem? Call 1-800-GAMBLER. Must be 21+ (18+ in some jurisdictions). Only bet where it's legal.

How books price an SGP

A regular parlay multiplies independent prices. Two -110 legs pay about +264. Books can do that because the legs come from different games and don't affect each other.

Same-game legs do affect each other. If a point guard goes over his assists, someone on his team probably went over on points. Books know this, so the SGP engine takes your raw multiplied price and shaves it down based on how related the legs are. That shave is the correlation adjustment. On top of it sits a bigger hold than you'd pay on any single prop.

Rough numbers help. A single player prop at -110/-110 carries around 4.5% hold. A three-leg SGP at most US books often carries something in the 20% to 30% range once you account for the adjustment and the juice on each leg. Exact figures vary by book and market, and books don't publish them. But the direction never changes: the more legs you add, the more you pay.

So the first rule is simple. An SGP needs a real reason to exist. If you'd bet each leg alone, ask why you're bundling them.

Start with a game script, not a leg

The best SGPs come from one specific read on how the game plays out. Everything else follows from that read.

A few scripts that show up all season:

• Track meet. Two top-10 pace teams, weak perimeter defense, total sitting at 238 or higher. Script: lots of possessions, guards rack up points and assists, the game over hits.
• Grind. Two slow teams, one or both missing a primary scorer. Script: game under, a big man gets extra touches, star points line is too high.
• Size mismatch. A team starting two bigs against a small-ball lineup. Script: rebounds and second-chance points for the big side, maybe a center's points-plus-rebounds over.
• Blowout risk. Spread at 12 or more. Script: starters on the favorite sit most of the fourth, bench guys get run, unders on the stars become live.

Pick one script. Then ask which three or so legs would all cash if you're right about it.

Correlation that works for you

Positive correlation means legs that tend to hit together. That's the whole point of an SGP. Some pairings with real logic behind them:

1. Point guard assists + teammate's points. Assists only happen when someone else scores. If you like a passer's over, the main finisher on his team is a natural partner.
2. Player points + team total over. A star's 30 points is a big chunk of a 118-point team total. They move together.
3. Game over + both primary scorers' overs. Pace lifts everyone. This is the classic track-meet build.
4. Center rebounds + opponent's poor shooting. More missed shots means more boards to grab. If the opponent is a bad three-point team on a back-to-back, a rebound over has a tailwind.

The catch: books spot the obvious pairs and adjust harder on them. Star points plus team total over is priced tighter than you'd think. The edges, if they exist, tend to sit in less obvious links like a role player's threes paired with a pace-up script.

Correlation that kills tickets

Negative correlation is legs fighting each other. People build these all the time without noticing.

• Two high-usage teammates both going big. If one guy takes 28 shots, the other usually doesn't. Stacking 30-point overs on two stars from the same team works against itself.
• Star over + blowout spread. You want the favorite covering -13 and their best player clearing 31.5? A blowout is exactly when that player sits the fourth.
• Game under + multiple points overs. Possible, but you're betting on a narrow outcome. Usually one of those legs is just there to pump the price.

Read every ticket back once before you submit it. If one leg needs the game to go a different way than another leg, cut it.

How many legs?

Two to three, most nights. Every extra leg adds hold and adds another way to lose.

A four- or five-leg SGP at +1500 can be fine as a small fun bet. Treat it like a lottery ticket and size it like one. If you're building a ticket you expect to win over time, keep it short and keep every leg tied to the same script.

Quick test for each leg: would you bet it on its own at this price? If the answer is no, it's filler.

When to skip the SGP

Plenty of nights a straight prop is the better bet.

• One strong lean, nothing else. Bet the single. Don't add two weaker legs to make the payout look exciting.
• Injury news still pending. If a starter is listed questionable and the report lands at 6:30 ET, your script could flip. Wait or pass. Our guide to betting NBA player props covers how minutes and injury notes move lines.
• The app won't price your combo. Some books block legs they view as too correlated. That's a signal they've already priced the link tightly elsewhere.
• You're chasing. Down on the night and reaching for a +900 ticket to get it back is how bankrolls disappear.

Shop the SGP price

SGP prices differ between books more than single props do, because each book runs its own correlation model. The same three legs can pay +450 at one app and +520 at another.

Build the ticket at two or three books before you place it. It takes two minutes. Over a season that gap adds up. Our rundown of the best sportsbooks for NBA player props covers which apps tend to have the friendliest SGP builders and deepest menus.

A sample build, start to finish

Here's how the process looks on a made-up slate. Numbers are for illustration only.

Two fast teams, total 236.5, spread 3. Neither side has a key injury. The road point guard averages around 9 assists and faces a defense that gives up a lot of transition looks. His top finisher is a wing who scores off cuts and lobs.

Script: track meet, close game, starters play full minutes.

Legs:

1. Road PG over 8.5 assists
2. Road wing over 22.5 points
3. Game total over 236.5

All three cash in the same version of the game. No leg needs a blowout or a slow pace. Price it at two books, take the better number and size it smaller than a straight bet.

Quick FAQ

Are NBA same-game parlays worth it? They can be when every leg follows one game script and you've shopped the price. Long tickets built for a big payout usually carry far more hold than single props.

How many legs should an NBA SGP have? Two or three for most bettors. Each added leg raises the house edge and adds another way to lose.

What is correlation in a same-game parlay? It's how much the legs tend to win or lose together. Books cut your payout on correlated legs, so the obvious pairings pay less than the raw math suggests.

Can you build an SGP with players from both teams? Yes. Most books allow legs from both sides of the same game, as long as the combination isn't blocked by their rules.

Bottom line

Build from the game, not from the payout. One script, two or three legs that all need the same thing to happen, priced at more than one book. Anything else is a lottery ticket, and that's fine as long as you size it that way.

Bet responsibly. 21+ (18+ where applicable). Gambling problem? Call 1-800-GAMBLER.