VotePredictor
← All pollsters

Change Research

Graded against the actual result across 99 races (from 156 polls, through 2024).

Races polled
99
Polls
156
Avg miss
5.63 pts
Most recent
2024

Head-to-head vs the VotePredictor model

The fair, apples-to-apples test: on the 97 races Change Research actually polled, how its final poll's margin compared to what the VotePredictor model predicted for those same races.

ModelAvg miss (pts)Called right
Change Research5.6774%
VotePredictor3.7190%

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 (97)

Each race Change 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.

RaceTheir callResultMissCall
2024 MN PresidentD+2.6D+4.21.6
2024 NE PresidentR+18.0R+20.52.5
2024 MN SenateD+11.7D+15.84.1
2022 MN GovernorD+5.0D+7.72.7
2022 IA SenateR+3.0R+12.29.2
2020 IN GovernorR+6.0R+24.518.5
2020 MT GovernorR+4.0R+12.98.9
2020 AR-2 HouseEVENR+10.710.7
2020 FL-15 HouseR+2.0R+10.88.8
2020 FL-16 HouseR+3.0R+11.08.0
2020 IA-4 HouseR+5.0R+24.219.2
2020 IN-5 HouseEVENR+4.14.1
2020 ME-2 HouseD+9.0D+6.12.9
2020 MO-2 HouseEVENR+6.46.4
2020 NE-2 HouseD+1.0R+4.65.6
2020 NY-24 HouseD+2.0R+2.44.4
2020 OH-1 HouseR+2.0R+7.15.1
2020 OK-5 HouseEVENR+4.14.1
2020 AZ PresidentD+3.0D+0.32.7
2020 FL PresidentD+3.0R+3.46.4
2020 IA PresidentEVENR+8.28.2
2020 IN PresidentR+14.0R+16.12.1
2020 M1 PresidentD+21.0D+23.12.1
2020 M2 PresidentD+1.0R+7.48.4
2020 MD PresidentD+29.0D+33.24.2
2020 ME PresidentD+12.0D+9.12.9
2020 MI PresidentD+7.0D+2.84.2
2020 MN PresidentD+5.0D+7.12.1
2020 MT PresidentR+5.0R+16.411.4
2020 N2 PresidentD+3.0D+6.53.5
2020 NC PresidentD+2.0R+1.33.3
2020 PA PresidentD+4.0D+1.22.8
2020 US PresidentD+10.0D+4.45.6
2020 WI PresidentD+8.0D+0.67.4
2020 AK SenateR+3.0R+12.79.7
2020 AZ SenateD+4.0D+2.31.7
2020 IA SenateD+1.0R+6.67.6
2020 ME SenateD+8.0R+8.616.6
2020 MI SenateD+5.0D+1.73.3
2020 MN SenateD+4.0D+5.21.2
2020 MT SenateR+4.0R+10.06.0
2020 NC SenateD+4.0R+1.75.7
2018 AZ GovernorR+7.6R+14.26.6
2018 CA GovernorD+12.0D+23.911.9
2018 IA GovernorD+3.0R+2.75.7
2018 ID GovernorR+16.0R+21.65.6
2018 ME GovernorD+8.0D+7.70.3
2018 MI GovernorD+8.0D+9.61.6
2018 MN GovernorD+12.0D+11.40.6
2018 NH GovernorR+1.0R+7.06.0
2018 OH GovernorD+5.0R+3.78.7
2018 PA GovernorD+11.0D+17.16.1
2018 SD GovernorD+6.0R+3.49.4
2018 WY GovernorR+34.0R+39.65.6
2018 CA-22 HouseR+5.0R+5.40.4
2018 CO-3 HouseR+15.0R+8.07.0
2018 FL-18 HouseR+9.0R+8.60.4
2018 FL-19 HouseR+9.0R+24.515.5
2018 GA-6 HouseD+6.0D+1.05.0
2018 IA-4 HouseR+1.0R+3.32.3
2018 IL-16 HouseR+14.0R+18.24.2
2018 IN-5 HouseR+6.0R+13.57.5
2018 IN-9 HouseR+7.0R+13.06.0
2018 KS-2 HouseR+1.0R+0.80.2
2018 MI-1 HouseR+12.0R+12.60.6
2018 MI-6 HouseD+2.0R+4.66.6
2018 MI-8 HouseD+1.0D+3.82.8
2018 MT-1 HouseR+8.0R+4.63.4
2018 NC-13 HouseEVENR+6.06.0
2018 NE-1 HouseR+16.0R+20.74.7
2018 NY-1 HouseR+15.0D+1.016.0
2018 NY-23 HouseR+2.0R+5.73.7
2018 OH-1 HouseR+3.0R+4.41.4
2018 OH-2 HouseR+13.0R+16.43.4
2018 OH-4 HouseR+24.0R+30.56.5
2018 OH-7 HouseR+6.0R+17.511.5
2018 SD-1 HouseR+10.0R+24.314.3
2018 VA-2 HouseEVEND+2.22.2
2018 WI-1 HouseD+1.0R+12.313.3
2018 WI-6 HouseR+2.0R+11.09.0
2018 WY-1 HouseR+27.0R+33.86.8
2018 AZ SenateD+0.2D+2.32.1
2018 FL SenateD+2.0R+0.12.1
2018 MI SenateD+2.0D+6.54.5
2018 MN SenateD+15.0D+24.19.1
2018 MS SenateR+5.0R+7.32.3
2018 MT SenateR+3.0D+3.66.6
2018 NJ SenateD+10.0D+11.21.2
2018 OH SenateD+10.0D+6.83.2
2018 PA SenateD+7.0D+13.16.1
2018 TX SenateEVENR+2.62.6
2018 WY SenateR+29.0R+36.97.9
2017 NJ GovernorD+16.0D+14.11.9
2017 VA GovernorD+6.0D+8.92.9
2017 MT-1 HouseR+5.0R+5.60.6
2017 SC-5 HouseR+9.0R+3.15.9
2017 AL SenateR+6.0D+1.67.6

Accuracy by time to election

24686–9 wk3–6 wk1–3 wk≤1 wkavg miss (pts)
Change ResearchAll 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 wk685.42+1.1463%
1–3 wk395.80+0.7382%
3–6 wk265.91+0.1781%
6–9 wk235.67-0.4874%

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)
201794.6R+3.7
2018595.7D+1.7
2020795.7D+5.4
202236.3D+4.6
202465.0D+0.8

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.