Head-to-head vs VotePredictor Elections
The fair, apples-to-apples test: on the 20 races Wick actually polled, how its final poll's margin compared to what VotePredictor Elections predicted for those same races.
| Model | Avg miss (pts) | Called right |
|---|---|---|
| Wick | 3.84 | 60% |
| VotePredictor Elections | 3.95 | 75% |
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 Wick 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 |
|---|---|---|---|---|
| 2022 AZ Governor | R+2.4 | D+0.7 | 3.1 | ✗ |
| 2022 GA Governor | R+9.1 | R+7.5 | 1.6 | ✓ |
| 2022 MI Governor | D+1.3 | D+10.5 | 9.2 | ✓ |
| 2022 NH Governor | R+10.6 | R+15.5 | 4.9 | ✓ |
| 2022 PA Governor | D+5.9 | D+14.8 | 8.9 | ✓ |
| 2022 WI Governor | R+0.3 | D+3.4 | 3.7 | ✗ |
| 2022 AZ Senate | D+2.3 | D+4.9 | 2.6 | ✓ |
| 2022 GA Senate | EVEN | D+1.0 | 1.0 | ✗ |
| 2022 NC Senate | R+4.5 | R+3.2 | 1.3 | ✓ |
| 2022 NH Senate | R+0.1 | D+9.1 | 9.2 | ✗ |
| 2022 PA Senate | R+1.7 | D+4.9 | 6.6 | ✗ |
| 2022 WI Senate | R+4.1 | R+1.0 | 3.1 | ✓ |
| 2020 FL President | R+2.9 | R+3.4 | 0.5 | ✓ |
| 2020 GA President | R+2.5 | D+0.2 | 2.7 | ✗ |
| 2020 MI President | D+0.3 | D+2.8 | 2.5 | ✓ |
| 2020 NC President | R+2.2 | R+1.3 | 0.9 | ✓ |
| 2020 OH President | R+2.9 | R+8.0 | 5.1 | ✓ |
| 2020 PA President | R+2.1 | D+1.2 | 3.3 | ✗ |
| 2020 GA Senate | R+2.5 | D+2.1 | 4.6 | ✗ |
| 2018 ME Senate | R+28.0 | R+24.8 | 3.2 | ✓ |
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 | 13 | 4.01 | -1.06 | 62% |
| 3–6 wk | 11 | 4.84 | -0.90 | 55% |
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) |
|---|---|---|---|
| 2020 | 8 | 3.1 | R+1.7 |
| 2022 | 17 | 5.4 | R+4.8 |
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.