Saint Anselm College Survey Center
Graded against the actual result across 16 races (from 25 polls, through 2022).
Head-to-head vs VotePredictor Elections
The fair, apples-to-apples test: on the 16 races Saint Anselm College Survey Center actually polled, how its final poll's margin compared to what VotePredictor Elections predicted for those same races.
| Model | Avg miss (pts) | Called right |
|---|---|---|
| Saint Anselm College Survey Center | 5.94 | 88% |
| VotePredictor Elections | 3.51 | 94% |
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 (16)
Each race Saint Anselm College Survey Center 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 NH Governor | R+18.0 | R+15.5 | 2.5 | ✓ |
| 2022 NH-1 House | R+6.0 | D+8.1 | 14.1 | ✗ |
| 2022 NH-2 House | D+8.0 | D+11.7 | 3.7 | ✓ |
| 2022 NH Senate | R+1.0 | D+9.1 | 10.1 | ✗ |
| 2020 NH Governor | R+25.0 | R+31.8 | 6.8 | ✓ |
| 2020 NH-1 House | D+5.0 | D+5.1 | 0.1 | ✓ |
| 2020 NH-2 House | D+15.0 | D+10.2 | 4.8 | ✓ |
| 2020 NH President | D+8.0 | D+7.4 | 0.6 | ✓ |
| 2020 NH Senate | D+15.0 | D+15.6 | 0.6 | ✓ |
| 2018 NH Governor | R+10.0 | R+7.0 | 3.0 | ✓ |
| 2018 NH-1 House | D+8.9 | D+8.6 | 0.3 | ✓ |
| 2018 NH-2 House | D+27.1 | D+13.4 | 13.7 | ✓ |
| 2008 NH Governor | D+66.0 | D+42.6 | 23.4 | ✓ |
| 2008 NH-1 House | D+7.0 | D+5.9 | 1.1 | ✓ |
| 2008 NH President | D+12.0 | D+9.6 | 2.4 | ✓ |
| 2008 NH Senate | D+14.0 | D+6.3 | 7.7 | ✓ |
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 | 9 | 4.82 | -0.25 | 78% |
| 3–6 wk | 16 | 4.96 | -0.78 | 100% |
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) |
|---|---|---|---|
| 2008 | 4 | 8.7 | D+8.7 |
| 2018 | 3 | 5.7 | D+3.7 |
| 2020 | 10 | 3.5 | D+3.0 |
| 2022 | 8 | 4.5 | R+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.