NFL Key Numbers: Why 3 and 7 Decide Your Bets

Football doesn't score in ones. It scores in 3s and 7s, and that single fact bends the entire distribution of final margins. Roughly 15% of NFL games end with a 3-point margin. About 9% end on 7. No other number comes close.

Every serious bettor knows this. What almost nobody does is put a price on it. "Buy the half point off 3" is repeated constantly as if it's free. It isn't. It costs juice, and past a certain price it turns a winning position into a losing one. We ran the numbers below and found the exact break-even price for both key numbers — and they're nowhere near each other.

For context on how we price our own plays, our sports betting record averages −137, where break-even sits at 57.8%. Knowing what a number costs is the whole job.

📊 Where NFL games actually land

Here's the distribution of final margins. The shape is not a smooth curve — it's a set of spikes, and the spikes are where your money lives or dies.

Margin	Share of games	Why it happens
3	~15%	One field goal. The default way close games end.
7	~9%	One touchdown plus the extra point.
10	~6%	Touchdown plus field goal.
6	~5.5%	Two field goals, or a missed extra point.
4	~5%	The dead zone just past the 3.
14	~4.5%	Two touchdowns.
1	~4%	Two-point conversions and missed kicks.

Notice the 4. It's the number immediately past the biggest spike, and it's worth almost nothing. That's why a line moving from 3 to 4 is a much bigger deal than one moving from 4 to 5 — the market just stepped over a cliff. If line movement is new to you, start with reading betting lines.

🧮 What a half point off 3 is actually worth

Here's the calculation nobody publishes. Take 100 games where the spread is exactly 3 and you're backing the favourite. A realistic split looks like this:

• 45 wins (favourite wins by 4+)
• 15 pushes (favourite wins by exactly 3)
• 40 losses

At −3 (−110), the pushes return your stake. You risk $8,500 net of pushes and profit $90.95 — a threadbare +1.1% ROI.

Now buy down to −2.5. Every one of those 15 pushes becomes a win. Your record is 60–40. But you pay for it in juice, and the price decides everything:

Read that chart carefully, because it contains the whole lesson. −130 turns +1.1% into +6.2%. The same buy at −160 turns it negative. The advice "always buy off the 3" is only true with a price attached — and the price is about −150.

⚖️ Why the 7 is worth much less

Here's where most bettors get it wrong. They treat 3 and 7 as twin key numbers and buy off both at the same price. But 7 lands in about 9% of games versus 15% for the 3 — you're buying materially fewer pushes, so you can afford materially less juice.

Buy	Pushes converted	Worth paying up to	Verdict at −130
−3 → −2.5	~15%	~−150	✅ Strong buy
−7 → −6.5	~9%	~−120	❌ Overpriced
−10 → −9.5	~6%	~−115	❌ Overpriced
−4 → −3.5	~5%	~−112	❌ Skip entirely

That last row deserves a note. Buying from −4 to −3.5 is the trap — it feels like you're approaching the 3, but you haven't crossed it. You're paying for the 4, which is worth almost nothing. The value is in crossing 3, not in getting near it.

Books know all of this. That's why the price to buy off the 3 is almost always worse than the price to buy off the 8. If you want to see where the juice really goes, read our breakdown of vig, then alternate lines betting for how to shop these prices properly.

📱 What this means on a Sunday

Three practical rules that follow directly from the maths:

1. A game sitting on 3 is worth shopping harder than any other number. Half a point of difference between two books is worth ~15% of outcomes. Nothing else on the board offers that. Our list of reduced juice sportsbooks exists for exactly this.
2. Take the free half point, pay for it only up to the number. −2.5 at −140 beats −3 at −110. −2.5 at −165 does not.
3. Never buy off 4, 5, 8 or 9. You're paying key-number prices for dead numbers.

Key numbers are also why teasers work when they work: a 6-point teaser that moves a favourite from −8.5 through both the 7 and the 3 is buying two spikes at once. One that moves −12 to −6 crosses only the 7 and is a far weaker proposition, even though it's the same 6 points.

Where pushes fit in

A push isn't a loss, which makes it easy to shrug off. But it's dead money — capital tied up in a game that returned you nothing while you carried full variance. Over a season, 15% of your action on 3-point spreads landing in the void is a real drag on compounding. Our guide to push rules covers how books grade them, which is not always what you'd assume.

And the deeper point: if you're consistently getting the better side of key numbers, that shows up in closing line value before it shows up in your bankroll. CLV is the leading indicator; the bankroll is the lagging one. It's why we grade every play at the price it was released — the whole record sits on one public page, losses included.

✅ The short version

Roughly 1 in 4 NFL games is decided by exactly 3 or 7. Getting on the right side of those two numbers is the highest-leverage decision on a football card. But leverage isn't the same as free: the 3 is worth paying up to about −150 to cross, the 7 only about −120, and everything else is close to worthless.

If you take one thing from this: the number matters, and so does the price of the number. Most bettors only ever check the first one. For more on turning that into a repeatable process, see expected value betting and Kelly criterion staking.

❓ Frequently asked questions

What are the NFL key numbers?

In order of importance: 3, 7, 10, 6, 4, 14 and 1. The 3 dominates at roughly 15% of games and the 7 follows at about 9%. Everything after 10 is a minor number.

Why are 3 and 7 the most common NFL margins?

Because football scores in those increments. A field goal is 3, a touchdown with the extra point is 7. Close games resolve on one score, and one score is almost always a 3 or a 7.

Is it worth buying the half point off 3?

Yes, up to about −150. In our model, moving from −3 (−110) to −2.5 (−130) lifts ROI from +1.1% to +6.2%. At −150 it's break-even. Past that you're paying more than the half point returns.

Should I buy the half point off 7 too?

Only up to about −120. The 7 converts roughly 9% of outcomes versus 15% for the 3, so it supports far less juice. Paying −130 to cross the 7 is an overpay.

What's the worst half point to buy?

Anything that doesn't cross a key number — −4 to −3.5, −5 to −4.5, −9 to −8.5. You're paying a premium to land on a number that decides about 5% of games or fewer.

Do key numbers matter for totals as well?

Much less. Totals cluster loosely around 41, 43, 44 and 47, but the distribution is far flatter than for margins, so no single total carries anything like the weight of the 3.

How do key numbers affect teasers?

They're the whole point. A 6-point teaser that drags a favourite through both the 7 and the 3 is buying two spikes. One that crosses only the 7 is much weaker for the same 6 points.

Do key numbers still hold in 2026?

Yes. Scoring rules have shifted the extra point and nudged two-point attempts up, which slightly fattens the 1 and 2, but 3 and 7 remain far and away the two biggest spikes. The structure of scoring hasn't changed.

⚠️ Responsible gambling: betting involves risk and no service can guarantee profit. Only wager money you can afford to lose, never chase losses, and set limits before you play. If gambling stops being entertainment, call 1-800-GAMBLER for confidential help. 21+ where legal.