Head-to-head vs the VotePredictor model
The fair, apples-to-apples test: on the 82 races GQR actually polled, how its final poll's margin compared to what the VotePredictor model predicted for those same races.
| Model | Avg miss (pts) | Called right |
|---|---|---|
| GQR | 5.90 | 70% |
| VotePredictor | 4.12 | 79% |
VotePredictor 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 (86)
Each race GQR 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.
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 wk | 5 | 3.59 | -0.69 | 40% |
| 1–3 wk | 31 | 4.24 | -0.83 | 65% |
| 3–6 wk | 68 | 5.77 | +0.03 | 68% |
| 6–9 wk | 29 | 5.20 | -0.95 | 69% |
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) |
|---|---|---|---|
| 2000 | 6 | 1.5 | D+1.3 |
| 2002 | 5 | 5.4 | D+5.4 |
| 2004 | 15 | 3.6 | D+2.7 |
| 2006 | 10 | 6.3 | D+1.2 |
| 2008 | 27 | 4.2 | D+0.7 |
| 2009 | 6 | 5.7 | D+5.7 |
| 2010 | 14 | 6.5 | D+5.7 |
| 2012 | 11 | 3.2 | D+0.4 |
| 2014 | 12 | 8.1 | D+8.1 |
| 2016 | 12 | 6.5 | D+2.9 |
| 2018 | 6 | 4.5 | D+1.8 |
| 2020 | 5 | 8.7 | D+8.7 |
| 2022 | 3 | 6.2 | D+6.2 |
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.