An NFL point spread handicaps the favorite by a set number of points. Three and seven are called key numbers because NFL margins cluster on them: across 1,615 regular-season games from 2020 to 2025, 14.61% finished exactly three points apart and 7.68% exactly seven, together 22.29% of all games.

Current numbers are on the NFL odds board.

What an NFL point spread is

A point spread handicaps the favorite by a number of points. The favorite is listed with a minus sign and must win by more than that number; the underdog is listed with a plus sign and can lose by less than it, or win outright.

LineFinal scoreResult
Chiefs −5.5Chiefs win 27-20 (margin 7)Chiefs side wins
Chiefs −5.5Chiefs win 23-20 (margin 3)Broncos side wins
Chiefs −3Chiefs win 23-20 (margin 3)Push — stake returned
Broncos +5.5Broncos lose 20-24 (margin 4)Broncos side wins

Why 3 and 7 are called key numbers

NFL scoring is built from field goals and touchdowns, so final margins pile up on a few values instead of spreading evenly. Across 1,615 regular-season games from 2020 to 2025, the single most common margin was 3 points.

14.61%of NFL regular-season games ended with a margin of exactly 3 points2020–2025 regular season · n = 1,615 games · as of 2026-07-28

Seven is the next spike, at 7.68%. Together, 3 and 7 account for 22.29% of all regular-season results — roughly one game in five landing on just two of the dozens of possible margins.

The full margin distribution

Final marginGamesShare
050.31%
1784.83%
2825.08%
323614.61%
4825.08%
5734.52%
61106.81%
71247.68%
8734.52%
9301.86%
10815.02%
11291.80%
12311.92%
13291.80%
14744.58%
15 or more47829.60%

NFL regular season, 2020–2025, n = 1,615. Margin 0 is a tie. Source: our own results archive; the query is committed as repo_check/seo/f31_lake_stats.py.

The shape is what matters. Three is not merely common, it is discontinuously common: it occurs 2.9 times as often as 2 (5.08%) and 2.9 times as often as 4 (5.08%), its immediate neighbours.

Seven is a genuine but much smaller spike. At 7.68% it sits only a little above 6 (6.81%) and clearly above 8 (4.52%). Six and ten (5.02%) are the secondary numbers the same scoring arithmetic produces, and the table shows them plainly rather than rounding the story down to two figures.

What that means for a half point

Because so much probability sits exactly on 3, the difference between −3 and −3.5 is much larger than the difference between −5 and −5.5. Moving a favorite from −3 to −3.5 gives away the 14.61% of games that land exactly on 3; moving from −5 to −5.5 gives away the 4.52% that land on 5.

This is also why a line sitting on 3 tends to hold there while the price beside it moves instead. A book would rather charge −120 on −3 than move to −3.5.

Is it stable season to season?

Yes, within the noise you would expect from about 270 games a year.

SeasonGamesMargin 3Margin 7
202025614.45%7.42%
202127213.97%6.62%
202227116.61%7.01%
202327213.97%8.09%
202427213.60%7.35%
202527215.07%9.56%

Regular season only. Single seasons carry visible sampling noise; the six-season pooled figures above are the ones worth quoting.

Playoff games behave similarly on this measure: across 78 playoff games in the same window, 26.92% finished with a 3-point margin and 8.97% with a 7-point margin. The playoff sample is small enough that those two figures should be read as indicative rather than precise.

How this was measured

Frequently asked questions

What are the key numbers in NFL betting?

Three and seven. In 1,615 NFL regular-season games from 2020 to 2025, 14.61% ended with a 3-point margin and 7.68% with a 7-point margin.

Why does a line stop at −3 instead of moving to −3.5?

Because 14.61% of games land exactly on 3, crossing that number changes the bet far more than crossing 5 or 8. Books typically adjust the price instead.

What happens if the margin lands exactly on the spread?

The bet pushes and the stake is returned. Only whole-number spreads can push; a half point removes the possibility.

Do playoff games follow the same pattern?

Broadly yes. In 78 playoff games over the same seasons, 26.92% landed on 3 and 8.97% on 7, though that sample is small.

Related: NFL odds board · how to read NFL odds · moneyline vs spread · betting glossary

Every figure on this page is a historical measurement of what already happened, produced by a committed, re-runnable script over our own odds archive. Past frequencies are not forecasts and are not betting advice: Nebula Insights publishes no picks, no selections and no recommended wagers. 18+ only.