Research
NFL dome scoring and the closing total
Everyone knows domes add points. The useful question is whether the market adds enough. We measured every NFL game since 2006 against both teams' own season scoring, split indoor from outdoor, and then checked 1,164 games against the closing total.
The headline
Indoor NFL games finished +2.2 points above the closing total on average across 385 games with lines, going 208–174 to the Over. Outdoor games were flat against the total. Every full season we could test leaned the same way.
Indoor vs closing total
+2.2 pts
208 Overs · 174 Unders
385 games · t = 3.27
Outdoor vs closing total
-0.6 pts
347 Overs · 422 Unders
779 games · t = -1.22
Low-scoring matchups indoors
+6.2 pts
vs both teams' season scoring
474 games · t = 10.06
Key finding in plain English
NFL games played indoors have scored more than the betting market expected. Between 2019 and 2022, dome totals closed about two points too low on average. The pattern was strongest when two weaker offenses met indoors. We cannot yet confirm it has survived into the most recent seasons.
Every matchup group, indoor and outdoor
Matchups are grouped by expected score: both teams' season points per game, leaving out the game itself. "vs expected" is season-adjusted. "vs close" uses only games with a closing total.
| Expected score | Venue | Games | Avg points | vs expected | Lined | vs close | Over–Under |
|---|---|---|---|---|---|---|---|
| Under 42 | Indoor | 474 | 44.3 | +6.2t = 10.06 | 88 | +1.2t = 0.87 | 42–45 |
| Under 42 | Outdoor | 1,206 | 41.1 | +3.1t = 8.30 | 227 | -1.0t = -1.27 | 91–133 |
| 42-47 | Indoor | 449 | 45.5 | +1.0t = 1.72 | 114 | +2.3t = 1.78 | 62–50 |
| 42-47 | Outdoor | 1,068 | 43.3 | -1.2t = -2.97 | 227 | -1.1t = -1.24 | 102–124 |
| 47-52 | Indoor | 416 | 48.7 | -0.6t = -0.83 | 106 | +2.2t = 1.70 | 60–46 |
| 47-52 | Outdoor | 848 | 46.1 | -3.1t = -6.69 | 187 | -0.4t = -0.39 | 83–102 |
| 52+directional only | Indoor | 216 | 53.1 | -2.3t = -2.37 | 77 | +3.4t = 2.17 | 44–33 |
| 52+ | Outdoor | 464 | 50.1 | -5.3t = -7.91 | 138 | +0.7t = 0.59 | 71–63 |
Bold figures clear |t| ≥ 1.96 with at least 40 games. The high-scoring drop-off against expectation appears outdoors too, so it is regression to the mean rather than a dome effect.
Indoor games against the closing total, by season
| Season | Indoor games | vs close | Over–Under |
|---|---|---|---|
| 2019 | 72 | +1.1 | 37–35 |
| 2020 | 91 | +4.0 | 52–38 |
| 2021 | 92 | +1.4 | 48–44 |
| 2022 | 96 | +2.0 | 51–43 |
| 2025partial season | 29 | +3.2 | 18–11 |
2023 has only 5 indoor games with a closing total and 2024 has none, so both are left out of this table.
Method and limits
- 5,141 regular and postseason games, 2006–2025; indoor means a fixed dome or a retractable-roof stadium. NFL game records do not say whether a retractable roof was open, so those venues count as indoor for every game; their roofs are closed for the large majority of games. Leaving retractable venues out entirely, fixed-dome games still finished 1.9 points above the closing total (113 Overs, 88 Unders, n=204, t=1.97), right at the edge of statistical significance; retractable venues alone finished 2.6 points above (n=181, t=2.70). The pattern is not driven by retractable roofs.
- Closing totals are ESPN consensus lines. Coverage is near-complete for 2019–2022, thin for 2023 and 2025, and missing for 2024 and earlier seasons.
- The 40-game minimum and the |t| ≥ 1.96 threshold were set before any group was inspected. The indoor 52+ group is labelled directional only.
- Not controlled: injuries, rest, pace, and line movement before the close.
- This is a historical pattern, not a prediction for any single game.
Data generated 2026-09-24.
Share this finding
Free to republish with a visible credit and a link to the study page.
Related research
- NFL Wind & Totals Study — how wind cuts scoring outdoors.
- NFL Climate Shock Study — dome teams playing outdoors in the cold.
- NFL stadium weather pages
References & data sources
- ESPN NFL scores and odds — espn.com/nfl (schedules, final scores and consensus closing totals)
- Internal NFL historical game archive: venue type, final scores and ESPN consensus closing totals.
- Every dated change we have made to a published study is listed on our corrections log.
Where this is applied
These findings feed the game-time read on every NFL venue page, using the same sample-size and significance rules published here.
OverUnderWeather Research Desk
Independent data analysis on public sports and weather records. Studies match historical game outcomes to archived game-window weather, apply pre-set sample and significance rules, and publish null results when that is what the data show. Editorial policy and corrections log.
Frequently asked
Do NFL games score more in domes?
Yes. Across 1,555 indoor games since 2006, domes averaged 47.1 total points against 44.1 outdoors. Measured against both teams' own season scoring, indoor games beat expectations in every matchup group from low-scoring to average.
Does the betting market already price the dome effect?
Not fully, in the window we can test. Across 385 indoor games with closing totals (2019–2022, plus partial 2023 and 2025), dome games finished +2.2 points above the closing total on average (t = 3.27), with 208 Overs and 174 Unders. Outdoor games were roughly flat against the total (-0.6, t = -1.22).
Do two high-scoring offenses in a dome go Under?
Against the teams' own scoring averages, yes: elite matchups (a combined expected score of 52 or more) fell 2.3 points short indoors and 5.3 points short outdoors, so this is regression to the mean, not a dome effect. Against the closing total, the same indoor group finished above the number. With 77 lined games it is directional evidence only.
Is this edge still there today?
We cannot say for certain. ESPN closing totals are almost missing for 2023 and 2024, so most of the market test comes from 2019–2022. The partial 2025 sample points the same way, but it is too small to confirm anything. Markets adjust once a pattern is widely known.
How is each game's expected score built?
Each team's points per game for that season, excluding the game being measured, are added together. The gap between that expected score and the actual score is then adjusted for the season's league-wide scoring level, so a high-scoring era cannot flatter any group.