How Often Does the NFL Spread Come Into Play?

Betting favorites dominated the 2024-25 NFL season.

Blind betting every favorite by NFL odds would have returned a 2.2% ROI. It marked the first time since the 2017-18 NFL season the return eclipsed 1%.

The results correlated with the worst underdog winning percentage and return on investment in quite some time.

At 77-195, underdogs produced a -21.3% return on investment. Not since the 2005-06 NFL season have underdogs finished with a ROI above -20%.

Last season, underdogs saw increased ATS success. Those teams covered 141 games while winning 94 outright.

Underdogs returned bettors a -1.3% ROI on those 94 victories.

ATS outcomes returned only slightly superior results. Betting $100 on every underdog in 2025 would have returned $15 for the entire season.

All of this data led me to ponder something: How often does the spread actually come into play?

Specifically, I’m looking to see how many times a favorite wins outright while the underdog covers the spread.

The goal? Trying to identify a consistent threshold where the points begin to matter. We automatically assume it’s above key numbers, but does that hold up lately?

NFL Betting Data: Underdogs SU vs. ATS Results

One simple formula can tell us how often the spread comes into play. First, we take every underdog ATS success in a season and subtract the number of outright wins.

Here is the data since 2021, the first year of a 17-game schedule:

• 2025-26: 141 ATS wins, 94 outright wins
• 2024-25: 124 ATS wins, 77 outright wins
• 2023-24: 124 ATS wins, 88 outright wins
• 2022-23: 148 ATS wins, 92 outright wins
• 2021-22: 142 ATS wins, 101 outright wins

Now, we can use that data to calculate the number of times dogs covered while failing to win:

• 2025-26: 47 games
• 2024-25: 47 games
• 2023-24: 36 games
• 2022-23: 56 games
• 2021-22: 41 games

As a percentage of total regular-season games (272 games):

• 2025-26: 17%
• 2024-25: 17%
• 2023-24: 13.2%
• 2022-23: 20.6%
• 2021-22: 15.1%

What’s most notable about these percentages is the last two years. It could easily be a coincidence, but 17% in two straight years featuring a new kickoff is noticeable.

The next logical question to ask is whether this data correlates with underdogs above key numbers.

I’ve always used +3 as the logical starting point for considering dogs ATS. Moving extra points back slightly impacted this strategy, but I remain skeptical that two is a key number.

Data backs up that belief.

Since 2015, 134 NFL games have been decided by exactly two points. Over the same span, 424 games landed on three, 251 landed on seven and 201 landed on six.

Now here’s a look at the results of underdogs that closed between -1.5 and +3:

• 2025-26: 51 ATS wins (+0.2% ROI), 45 SU wins (+0%)
• 2024-25: 47 ATS wins (-10.5% ROI), 42 SU wins (-12.1%)
• 2023-24: 47 ATS wins (-11.3% ROI), 43 SU wins (-20.5%)
• 2022-23: 47 ATS wins (-7.4% ROI), 43 SU wins (-8.4%)
• 2021-22: 48 ATS wins (+7.8% ROI), 44 SU wins (+12.5%)

Last year, seven teams failed to win outright while covering the spread below +3. Four of those seven teams closed at +3.

Over the last two years, dogs at +3 or lower covered 98 times while pushing four times (all at +3). They won 87 of those games, meaning the spread came into play 15 times.

In other words, 16% of the 94 games where the spread mattered came at +3 or shorter. Since 2021, dogs at +3 or lower went 244-251-15 ATS vs. 217-292-1 SU.

What this data tells me is that +3 is the first number where the spread begins to impact games.

Over the same timeframe, dogs at +2.5 or shorter covered while failing to win only 10 times, or 4% of the sample.

Range that Produces Largest ATS vs. SU Differential

What is most fascinating about this research, to me, are the numbers where a big swing occurs.

Since 2021, the largest ATS vs. SU discrepancy came between +4 and +7.5. Dogs are 248-205-3 ATS over that span vs. 146-307-3 SU.

Those records led to +4.8% and +4.4% returns on investment, respectively. Here’s the breakdown by each number:

• +4: 20-31 SU, 29-22 ATS
• +4.5: 22-47 SU, 39-30 ATS
• +5: 8-5 SU, 9-4 ATS
• +5.5: 25-39 SU, 37-27 ATS
• +6: 11-35 SU, 18-29 ATS
• +6.5: 20-61 SU, 45-37 ATS
• +7: 18-51 SU, 34-33 ATS
• +7.5: 22-38 SU, 37-23 ATS

The 102 times a dog covered while failing to win accounts for 45% of the 227 occurrences since 2021.

Potential Takeaways from ATS vs SU Data

Most notable about that data is the recent increase in games landing on five. From 2003 to 2014, only 88 games produced that margin. Since 2015, it rose to 127 occurrences.

Of those 127 games, 29 happened in the last two years. That’s a larger number than games decided by exactly eight points and 10 points since 2024.

It also tells me betting underdogs north of +8 is historically problematic.

The last regular season that produced a positive return on investment for every underdog north of +8: 2019-20. Not since the 2017-18 season have those teams returned north of 10%.

Over the last two years, underdogs at +8 or higher are 11-68 SU for a -31.8% return on investment. That includes 5-41 in 2025 for a -42.3% ROI.

Conversely, betting every underdog north of +8 ATS since 2024 saw bettors go 39-39-1, losing only 3.8%.

If you’re going off historical trends, dogs at +5 or +5.5 return the highest percentage. Over the last five years, those dogs netted a +30% ROI or higher three times.

Since 2021, dogs at +5 or +5.5 are returning better results on the moneyline. A $100 bettor would have netted 29.6% ROI on the moneyline vs. 14.8% ATS.

Accordingly, if a dog is priced between key numbers, it may prove a better strategy to evaluate their chances of winning outright.