Selzer
Graded against the actual result across 50 races (from 79 polls, through 2024).
Head-to-head vs VotePredictor Elections
The fair, apples-to-apples test: on the 45 races Selzer actually polled, how its final poll's margin compared to what VotePredictor Elections predicted for those same races.
| Model | Avg miss (pts) | Called right |
|---|---|---|
| Selzer | 4.60 | 78% |
| VotePredictor Elections | 4.24 | 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 (50)
Each race Selzer 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 | 28 | 2.68 | -1.56 | 89% |
| 1–3 wk | 9 | 8.61 | +3.55 | 67% |
| 3–6 wk | 29 | 5.81 | +0.11 | 83% |
| 6–9 wk | 13 | 6.07 | -0.07 | 62% |
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) |
|---|---|---|---|
| 2002 | 4 | 6.2 | D+5.5 |
| 2004 | 7 | 5.3 | D+3.2 |
| 2006 | 10 | 5.0 | R+2.5 |
| 2008 | 11 | 4.1 | D+0.7 |
| 2010 | 4 | 3.0 | R+2.9 |
| 2012 | 4 | 1.5 | R+0.5 |
| 2014 | 5 | 4.2 | D+3.3 |
| 2016 | 16 | 4.4 | D+3.5 |
| 2020 | 4 | 5.4 | D+5.4 |
| 2022 | 4 | 3.1 | D+3.1 |
| 2024 | 5 | 14.8 | D+14.8 |
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.