OH Predictive Insights
Graded against the actual result across 21 races (from 33 polls, through 2024).
Head-to-head vs VotePredictor Elections
The fair, apples-to-apples test: on the 20 races OH Predictive Insights actually polled, how its final poll's margin compared to what VotePredictor Elections predicted for those same races.
| Model | Avg miss (pts) | Called right |
|---|---|---|
| OH Predictive Insights | 3.54 | 65% |
| VotePredictor Elections | 4.03 | 80% |
VotePredictor Elections aggregates all the pollsters, so it's expected to beat any single one on margin — that's the value of averaging. The honest comparison among forecasters is on the combined board.
Every race (20)
Each race OH Predictive Insights polled, scored on its final poll — the call right before the vote — against the actual Dem−Rep result. Click a race for its full detail.
| Race | Their call | Result | Miss | Call |
|---|---|---|---|---|
| 2024 UT Governor | R+17.0 | R+24.4 | 7.4 | ✓ |
| 2024 PA-8 House | D+7.0 | R+1.6 | 8.6 | ✗ |
| 2024 AZ President | R+1.0 | R+5.5 | 4.5 | ✓ |
| 2024 NV President | D+1.0 | R+3.1 | 4.1 | ✗ |
| 2024 US President | D+3.0 | R+1.5 | 4.5 | ✗ |
| 2024 UT President | R+20.0 | R+21.6 | 1.6 | ✓ |
| 2024 AZ Senate | D+4.0 | D+2.4 | 1.6 | ✓ |
| 2024 NV Senate | D+2.0 | D+1.6 | 0.4 | ✓ |
| 2024 UT Senate | R+28.0 | R+30.8 | 2.8 | ✓ |
| 2022 AZ Governor | R+2.0 | D+0.7 | 2.7 | ✗ |
| 2022 NV Governor | D+4.0 | R+1.5 | 5.5 | ✗ |
| 2022 AZ Senate | D+2.0 | D+4.9 | 2.9 | ✓ |
| 2022 NV Senate | D+2.0 | D+0.8 | 1.2 | ✓ |
| 2020 AZ-6 House | R+3.0 | R+4.3 | 1.3 | ✓ |
| 2020 AZ President | D+3.0 | D+0.3 | 2.7 | ✓ |
| 2020 AZ Senate | D+5.0 | D+2.3 | 2.7 | ✓ |
| 2018 AZ Governor | R+18.0 | R+14.2 | 3.8 | ✓ |
| 2018 AZ-4 House | R+32.0 | R+37.7 | 5.7 | ✓ |
| 2018 AZ Senate | R+1.0 | D+2.3 | 3.3 | ✗ |
| 2016 AZ President | EVEN | R+3.5 | 3.5 | ✗ |
Accuracy by time to election
Lower is better. Time to election runs right (election week) to left (~2 months out).
By the numbers
| Time to election | Polls | Avg miss | vs field | Called right |
|---|---|---|---|---|
| 1–3 wk | 18 | 3.94 | -1.12 | 67% |
| 3–6 wk | 10 | 4.41 | -1.29 | 70% |
| 6–9 wk | 4 | 5.40 | -0.74 | 75% |
vs field is this pollster's average miss minus all pollsters' at the same lead time — green beats the field, redtrails it. Our historical polls reach ~2 months out; earlier polling isn't in the record.
Track record by cycle — getting better?
| Year | Polls | Avg miss | Lean (house effect) |
|---|---|---|---|
| 2018 | 6 | 5.6 | R+3.7 |
| 2020 | 7 | 4.3 | D+4.3 |
| 2022 | 10 | 3.7 | D+0.2 |
| 2024 | 9 | 3.9 | D+3.9 |
Do we credit a pollster for fixing its bias? Each cycle, the model re-estimates every pollster's lean from all its earlier polls (walk-forward) and subtracts it before using the poll. We tested weighting recent cycles more — it doesn't help: a pollster's lean in one cycle barely predicts the next (correlation 0.28), so the swings above are mostly noise, and averaging over more history beats chasing the latest cycle. The all-time estimate we use came out within ~0.5% of the best option.