North Star Opinion Research
Graded against the actual result across 19 races (from 21 polls, through 2024).
Head-to-head vs VotePredictor Elections
The fair, apples-to-apples test: on the 13 races North Star Opinion Research actually polled, how its final poll's margin compared to what VotePredictor Elections predicted for those same races.
| Model | Avg miss (pts) | Called right |
|---|---|---|
| North Star Opinion Research | 6.65 | 46% |
| VotePredictor Elections | 5.15 | 77% |
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 (19)
Each race North Star Opinion Research 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 |
|---|---|---|---|---|
| 2024 PA President | EVEN | R+1.7 | 1.7 | ✗ |
| 2024 PA Senate | D+3.0 | R+0.2 | 3.2 | ✗ |
| 2016 IL-10 House | R+7.0 | D+5.2 | 12.2 | ✗ |
| 2016 NE-2 House | R+4.0 | R+1.2 | 2.8 | ✓ |
| 2016 NH-1 House | R+3.0 | D+1.3 | 4.3 | ✗ |
| 2016 N2 President | R+4.0 | R+2.2 | 1.8 | ✓ |
| 2014 AZ-1 House | R+6.0 | D+5.2 | 11.2 | ✗ |
| 2012 MA-6 House | R+8.0 | D+1.1 | 9.1 | ✗ |
| 2012 MI-9 House | D+2.0 | D+27.9 | 25.9 | ✓ |
| 2012 US President | R+1.0 | D+3.9 | 4.9 | ✗ |
| 2010 VA-2 House | R+5.0 | R+10.7 | 5.7 | ✓ |
| 2008 TN President | R+15.0 | R+15.1 | 0.1 | ✓ |
| 2008 TN Senate | R+37.0 | R+33.5 | 3.5 | ✓ |
| 2002 AL Governor | R+1.0 | R+0.2 | 0.8 | ✓ |
| 2002 TN Senate | R+16.0 | R+9.9 | 6.1 | ✓ |
| 2000 GA President | R+6.0 | R+11.7 | 5.7 | ✓ |
| 2000 MO President | EVEN | R+3.3 | 3.3 | ✗ |
| 2000 MO Senate | R+5.0 | D+2.1 | 7.1 | ✗ |
| 1998 SC Senate | R+1.0 | D+7.0 | 8.0 | ✗ |
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 | 7 | 3.03 | -2.04 | 57% |
| 3–6 wk | 8 | 8.70 | +2.96 | 75% |
| 6–9 wk | 5 | 6.33 | +0.18 | 20% |
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 | 3 | 5.4 | D+0.7 |
| 2002 | 4 | 5.0 | R+5.0 |
| 2012 | 3 | 13.3 | R+13.3 |
| 2016 | 4 | 5.3 | R+5.3 |
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.