Vox Populi Polling
Graded against the actual result across 27 races (from 35 polls, through 2018).
Head-to-head vs the VotePredictor model
The fair, apples-to-apples test: on the 26 races Vox Populi Polling 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 |
|---|---|---|
| Vox Populi Polling | 4.21 | 73% |
| VotePredictor | 3.96 | 77% |
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 (26)
Each race Vox Populi Polling 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 |
|---|---|---|---|---|
| 2018 AZ Governor | R+8.0 | R+14.2 | 6.2 | ✓ |
| 2018 CA Governor | D+20.0 | D+23.9 | 3.9 | ✓ |
| 2018 FL Governor | D+6.0 | R+0.4 | 6.4 | ✗ |
| 2018 NV Governor | D+4.0 | D+4.1 | 0.1 | ✓ |
| 2018 TN Governor | R+12.0 | R+21.0 | 9.0 | ✓ |
| 2018 TX Governor | R+10.0 | R+13.3 | 3.3 | ✓ |
| 2018 AZ Senate | D+4.0 | D+2.3 | 1.7 | ✓ |
| 2018 FL Senate | EVEN | R+0.1 | 0.1 | ✗ |
| 2018 IN Senate | D+10.0 | R+5.9 | 15.9 | ✗ |
| 2018 MO Senate | R+2.0 | R+5.8 | 3.8 | ✓ |
| 2018 NJ Senate | D+8.0 | D+11.2 | 3.2 | ✓ |
| 2018 NV Senate | D+2.0 | D+5.0 | 3.0 | ✓ |
| 2018 TN Senate | R+6.0 | R+10.8 | 4.8 | ✓ |
| 2018 TX Senate | EVEN | R+2.6 | 2.6 | ✗ |
| 2018 WV Senate | D+6.0 | D+3.3 | 2.7 | ✓ |
| 2015 KY Governor | EVEN | R+8.7 | 8.7 | ✗ |
| 2014 CO Governor | D+5.0 | D+3.3 | 1.7 | ✓ |
| 2014 GA Governor | R+7.0 | R+7.9 | 0.9 | ✓ |
| 2014 NH Governor | R+3.0 | D+4.9 | 7.9 | ✗ |
| 2014 CO Senate | R+3.0 | R+1.9 | 1.1 | ✓ |
| 2014 GA Senate | R+5.0 | R+7.7 | 2.7 | ✓ |
| 2014 LA Senate | R+11.0 | R+11.9 | 0.9 | ✓ |
| 2014 NC Senate | R+5.0 | R+1.6 | 3.4 | ✓ |
| 2014 NH Senate | R+4.0 | D+3.2 | 7.2 | ✗ |
| 2014 NM Senate | D+6.0 | D+11.1 | 5.1 | ✓ |
| 2014 VA Senate | D+4.0 | D+0.8 | 3.2 | ✓ |
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 | 18 | 4.11 | -0.96 | 72% |
| 3–6 wk | 8 | 4.77 | -0.97 | 75% |
| 6–9 wk | 9 | 5.81 | -0.34 | 67% |
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) |
|---|---|---|---|
| 2014 | 13 | 3.9 | R+1.2 |
| 2018 | 21 | 5.0 | D+2.9 |
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.