How Much NFL Cover Rates Swing From Season to Season

A cover rate is the share of games a side beat the spread, and even a group as big as every home team in the league moves a lot from one NFL season to the next. From 2015 to 2025, home teams covered between 43.8% and 51.8% in a single season, according to nflverse's closing lines. That is the yardstick for reading any short record, including the one on the NFL Weekly Record.

One group, eleven seasons

Home teams are a fixed category of side, a simple group to follow across seasons. Their against-the-spread record by season, regular season and playoffs, games with a closing line:

Season	Home ATS record (W-L-P)	Games	Cover rate
2015	117-136-10	263	46.2%
2017	131-122-8	261	51.8%
2019	110-141-10	261	43.8%
2024	139-135-4	278	50.7%
2025	140-136-1	277	50.7%
2015–2025	1415-1477-75	2,967	48.9%

Source: nflverse games.csv, closing lines. Cover rates leave pushes out of the division.

Each season is roughly 260 to 280 games, and the cover rate still ranged from 43.8% in 2019 to 51.8% in 2017. Pooled over all eleven seasons, home teams covered 48.9% of 2,967 games.

The same swing in totals

Over/under results move just as much. Across the same seasons, the over hit 48.8% of 3,028 games. By season it ran from 44.5% in 2022 (284 games) to 53.9% in 2024 (285 games).

How wide the results spread around the line

In the 2015–2025 nflverse file the final margin landed within 3 points of the closing line in 23.2% of games and within 7 points in 46.9%. It missed by more than 14 points in 24% of games, and the standard deviation of the result minus the line was 12.72 points. A season is a few hundred draws from a spread of outcomes that wide, so a gap of a few percentage points between two seasons is ordinary.

Where the break-even number fits

At standard -110 pricing a side needs 52.38% to break even. The season swings above sit on both sides of that figure, which is why a single season above or below it says little about the method behind it. The Weekly Record's own season hub prints a coin-flip range for the number of games graded so far, so you can see how much room luck has before reading anything into the percentage.

A worked example from the Weekly Record

The NFL Weekly Record grades the Spread Dial's lean on every game against nflverse's closing spread, one week at a time. A week is a small slice of a season, and a season is already a small slice of the eleven-year picture above. If a full season of home-team results can run from 43.8% to 51.8%, a few weeks of leans can land almost anywhere. The longer test is the out-of-sample record on the NFL Spread Dial, which the site reports as near a coin flip.

What this cannot tell you

These are closed seasons. They describe how often a fixed group covered, not which side any coming game will favor. Nothing here is a reason to lean one way, and a swing in the past does not predict a swing in the next season. The Weekly Record uses the closing line, which may differ from the line the dial showed earlier in the week, and it is not a betting record.

Data: nflverse schedules and results, CC-BY 4.0. Not affiliated with or endorsed by the NFL or any team. Written by Eric, OddsDial founder.

Last reviewed: October 2026. Informational only, not betting advice.