Skip to main content

Public validation record

2026 Football Model Performance

These results use official OSPR projections locked before kickoff and evaluated against final scores. Mirrored schedule records and live forecasts are excluded.

Evaluated through Aug 29, 2026

Accuracy calculated Aug 31, 2026 9:59 AM EDT

Reset

Performance summary

707 games evaluated with ospr-2026.1

Games evaluated

707

Winner accuracy

59.7%

422 correct

Average margin error

22.6

Lower is better

Average score error

13.1

Per-team points

Brier score

0.236

Probability accuracy

Close projection rate

12.4%

Correct and within 7

Regular season vs. playoffs

A direct comparison using the selected season, week, team, and OSPR model version.

Regular season

Games
707
Winner accuracy
59.7%
Margin error
22.6

Playoffs

Games
0
Winner accuracy
Margin error

Performance by week

Results for the selected model version and season phase.

Week Games Winner accuracy Margin error
Week 1 355 58.3% 22.0
Week 2 351 61.0% 23.3

Probability calibration

Observed win rate should approach the average predicted probability as sample sizes grow.

Probability Games Predicted Observed Gap Brier
0–9% 1 8.5% 0.0% -8.5 0.007
10–19% 32 16.7% 18.8% +2.0 0.160
20–29% 81 25.9% 34.6% +8.6 0.234
30–39% 117 35.2% 40.2% +5.0 0.237
40–49% 167 45.5% 48.5% +3.0 0.250
50–59% 128 54.6% 50.8% -3.8 0.256
60–69% 96 65.2% 67.7% +2.6 0.217
70–79% 73 74.6% 68.5% -6.1 0.218
80–89% 12 83.1% 58.3% -24.7 0.301

Performance by prediction confidence

Newer snapshots use historical calibration and comparable sample size. Legacy snapshots fall back to rating-evidence confidence.

Confidence Games Winner accuracy Average margin error
High 252 55.6% 22.8 points
Medium 21 76.2% 21.5 points
Low 434 61.3% 22.5 points

Performance by risk indicator

A game may appear in more than one group because indicators describe different kinds of uncertainty.

Indicator Games Winner accuracy Average margin error
Limited Data 706 59.6% 22.6 points
Close Game 535 56.8% 22.8 points
Upset Possible 505 56.0% 22.4 points

Largest misses

The largest margin errors in this selection. A miss is a model result to learn from, not automatically a data error.

Date Matchup Projected Final Margin error
Aug 21, 2026 Anna vs. Spencerville 19–22 60–0 63.2 points
Aug 29, 2026 Warrensville Heights vs. Cleveland JFK 18–20 57–0 59.2 points
Aug 28, 2026 West Clermont vs. Anderson 24–17 19–70 58.8 points
Aug 28, 2026 Ross vs. Thurgood Marshall 19–20 57–0 57.4 points
Aug 29, 2026 John Marshall vs. Rocky River 17–24 0–63 56.8 points

Evaluated games

The game-level record behind the summary above.

Aug 21, 2026

Cleveland Heights vs. Wadsworth

Correct
Projected
21–26
Final
6–28
Projected winner
Wadsworth
Margin error
16.3

Aug 21, 2026

Canton McKinley vs. Warren G. Harding

Correct
Projected
24–20
Final
33–0
Projected winner
Canton McKinley
Margin error
28.4

Aug 21, 2026

Berea-Midpark vs. Brunswick

Missed
Projected
26–16
Final
42–47
Projected winner
Berea-Midpark
Margin error
15.6

Aug 20, 2026

Cardinal vs. LaBrae

Correct
Projected
18–20
Final
0–49
Projected winner
LaBrae
Margin error
46.8

Aug 20, 2026

River vs. Portsmouth Notre Dame

Missed
Projected
18–25
Final
41–6
Projected winner
Portsmouth Notre Dame
Margin error
41.9

Aug 20, 2026

Conotton Valley vs. Tuscarawas Central Catholic

Correct
Projected
23–16
Final
46–12
Projected winner
Conotton Valley
Margin error
26.6

Aug 20, 2026

Bridgeport vs. Frontier

Correct
Projected
17–22
Final
14–50
Projected winner
Frontier
Margin error
30.9

Aug 20, 2026

Delphos St. John's vs. Sidney Lehman

Missed
Projected
15–25
Final
40–6
Projected winner
Sidney Lehman
Margin error
44.0

Aug 20, 2026

Tiffin Calvert vs. Leipsic

Correct
Projected
21–23
Final
3–40
Projected winner
Leipsic
Margin error
35.2

Aug 20, 2026

Jackson-Milton vs. Mathews

Correct
Projected
15–22
Final
27–34
Projected winner
Mathews
Margin error
0.2

Aug 20, 2026

David Anderson vs. Leetonia

Correct
Projected
24–15
Final
14–0
Projected winner
David Anderson
Margin error
4.6

Aug 20, 2026

Gamble Montessori vs. Stivers School For The Arts

Missed
Projected
16–19
Final
50–6
Projected winner
Stivers School For The Arts
Margin error
47.0

Aug 20, 2026

Montpelier vs. Antwerp

Correct
Projected
24–15
Final
21–14
Projected winner
Montpelier
Margin error
2.6

Aug 20, 2026

Colonel Crawford vs. Fairbanks

Correct
Projected
23–22
Final
45–21
Projected winner
Colonel Crawford
Margin error
22.6

Aug 20, 2026

Columbus Grove vs. Pandora-Gilboa

Correct
Projected
27–22
Final
42–12
Projected winner
Columbus Grove
Margin error
25.3

Aug 20, 2026

McDermott Northwest vs. Piketon

Correct
Projected
20–24
Final
0–41
Projected winner
Piketon
Margin error
36.9

Aug 20, 2026

Ironton vs. Williamsburg

Missed
Projected
25–26
Final
42–0
Projected winner
Williamsburg
Margin error
43.0

Aug 20, 2026

Harvest Prep vs. Versailles

Missed
Projected
25–20
Final
18–21
Projected winner
Harvest Prep
Margin error
7.6

Aug 20, 2026

Rootstown vs. Ravenna Southeast

Missed
Projected
19–21
Final
14–13
Projected winner
Ravenna Southeast
Margin error
3.1

Aug 20, 2026

Youngstown Liberty vs. Campbell Memorial

Correct
Projected
25–18
Final
45–2
Projected winner
Youngstown Liberty
Margin error
35.2

Aug 20, 2026

Garrettsville Garfield vs. Warren JFK

Correct
Projected
25–21
Final
37–28
Projected winner
Garrettsville Garfield
Margin error
5.4

Aug 20, 2026

Cambridge vs. Indian Creek

Correct
Projected
13–27
Final
36–49
Projected winner
Indian Creek
Margin error
1.4

Aug 20, 2026

Fairless vs. Sandy Valley

Missed
Projected
18–21
Final
44–6
Projected winner
Sandy Valley
Margin error
40.8

Aug 20, 2026

Lima Shawnee vs. Lima Central Catholic

Missed
Projected
23–22
Final
20–51
Projected winner
Lima Shawnee
Margin error
31.9

Aug 20, 2026

Jackson vs. Unioto

Missed
Projected
27–25
Final
21–35
Projected winner
Jackson
Margin error
16.3

How to interpret these numbers

Winner accuracy shows whether OSPR favored the team that won. Average margin error shows how far the projected point margin was from the final margin.

Brier score measures probability quality from 0 to 1; lower is better. Calibration compares predicted win probabilities with observed outcomes.

Small groups can swing sharply, so every percentage is shown with its evaluated game count.

Injuries, roster changes, weather, and information not reflected in recorded results may affect individual games.