Ipsos
Graded against the actual result across 124 races (from 408 polls, through 2024).
Head-to-head vs the VotePredictor model
The fair, apples-to-apples test: on the 120 races Ipsos 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 |
|---|---|---|
| Ipsos | 5.29 | 80% |
| VotePredictor | 3.40 | 88% |
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 (122)
Each race Ipsos 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 | 50 | 3.45 | -0.83 | 76% |
| 1–3 wk | 106 | 5.06 | -0.01 | 76% |
| 3–6 wk | 147 | 6.08 | +0.34 | 78% |
| 6–9 wk | 105 | 5.46 | -0.69 | 82% |
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) |
|---|---|---|---|
| 2004 | 8 | 5.1 | D+3.3 |
| 2008 | 10 | 3.9 | R+2.8 |
| 2010 | 22 | 6.5 | R+1.3 |
| 2012 | 35 | 2.4 | R+0.5 |
| 2014 | 9 | 6.3 | D+5.3 |
| 2016 | 211 | 6.0 | D+3.8 |
| 2018 | 34 | 5.0 | D+0.9 |
| 2020 | 72 | 4.7 | D+4.7 |
| 2024 | 5 | 4.0 | D+4.0 |
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.