VotePredictor
← All pollsters

Noble Predictive Insights

Graded against the actual result across 13 races (from 25 polls, through 2022).

Races polled
13
Polls
25
Avg miss
4.39 pts
Most recent
2022

Head-to-head vs VotePredictor Elections

The fair, apples-to-apples test: on the 12 races Noble Predictive Insights actually polled, how its final poll's margin compared to what VotePredictor Elections predicted for those same races.

ModelAvg miss (pts)Called right
Noble Predictive Insights3.3967%
VotePredictor Elections2.9775%

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 (12)

Each race Noble Predictive Insights 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.

RaceTheir callResultMissCall
2022 AZ GovernorR+2.0D+0.72.7
2022 NV GovernorD+4.0R+1.55.5
2022 AZ SenateD+2.0D+4.92.9
2022 NV SenateD+2.0D+0.81.2
2020 AZ-6 HouseR+3.0R+4.31.3
2020 AZ PresidentD+3.0D+0.32.7
2020 AZ SenateD+5.0D+2.32.7
2018 AZ GovernorR+18.0R+14.23.8
2018 AZ-4 HouseR+32.0R+37.75.7
2018 AZ-8 HouseR+10.0R+4.75.3
2018 AZ SenateR+1.0D+2.33.3
2016 AZ PresidentEVENR+3.53.5

Accuracy by time to election

24686–9 wk3–6 wk1–3 wk≤1 wkavg miss (pts)
Noble Predictive InsightsAll pollsters (field average)

Lower is better. Time to election runs right (election week) to left (~2 months out).

By the numbers

Time to electionPollsAvg missvs fieldCalled right
1–3 wk104.07-1.0070%
3–6 wk104.41-1.3370%
6–9 wk45.40-0.7575%

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?

YearPollsAvg missLean (house effect)
201875.5R+3.9
202074.3D+4.3
2022103.7D+0.2

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.