VotePredictor
← All pollsters

Research Co.

Graded against the actual result across 82 races (from 86 polls, through 2025).

Races polled
82
Polls
86
Avg miss
4.23 pts
Most recent
2025

Head-to-head vs the VotePredictor model

The fair, apples-to-apples test: on the 81 races Research Co. actually polled, how its final poll's margin compared to what the VotePredictor model predicted for those same races.

ModelAvg miss (pts)Called right
Research Co.4.1884%
VotePredictor3.4389%

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 (81)

Each race Research Co. 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
2025 NJ GovernorD+2.0D+14.412.4
2025 VA GovernorD+17.0D+15.41.6
2024 MO GovernorR+14.0R+20.46.4
2024 WA GovernorD+14.0D+11.22.8
2024 CA PresidentD+32.0D+20.211.8
2024 FL PresidentR+7.0R+13.16.1
2024 MI PresidentD+2.0R+1.43.4
2024 MN PresidentD+7.0D+4.22.8
2024 MO PresidentR+15.0R+18.43.4
2024 NJ PresidentD+17.0D+5.911.1
2024 NY PresidentD+16.0D+12.63.4
2024 PA PresidentD+1.0R+1.72.7
2024 US PresidentD+2.0R+1.53.5
2024 VA PresidentD+6.0D+5.80.2
2024 WA PresidentD+15.0D+18.33.3
2024 WI PresidentD+3.0R+0.83.8
2024 CA SenateD+24.5D+17.86.7
2024 FL SenateR+7.0R+12.85.8
2024 MI SenateD+6.0D+0.35.7
2024 MN SenateD+17.0D+15.81.2
2024 MO SenateR+11.0R+13.82.8
2024 NJ SenateD+18.0D+9.68.4
2024 NY SenateD+28.0D+18.39.7
2024 PA SenateD+2.0R+0.22.2
2024 VA SenateD+12.0D+9.03.0
2024 WA SenateD+20.0D+18.51.5
2024 WI SenateD+1.0D+0.90.1
2022 AZ GovernorR+2.0D+0.72.7
2022 CA GovernorD+20.0D+18.41.6
2022 FL GovernorR+14.0R+19.45.4
2022 GA GovernorR+7.0R+7.50.5
2022 IL GovernorD+20.0D+12.57.5
2022 NV GovernorR+2.0R+1.50.5
2022 NY GovernorD+10.0D+7.32.7
2022 OH GovernorR+22.0R+25.03.0
2022 PA GovernorD+12.0D+14.82.8
2022 WI GovernorEVEND+3.43.4
2022 AZ SenateD+4.0D+4.90.9
2022 CA SenateD+16.0D+22.16.1
2022 FL SenateR+10.0R+16.46.4
2022 GA SenateEVEND+1.01.0
2022 IL SenateD+18.0D+15.32.7
2022 NV SenateR+1.0D+0.81.8
2022 NY SenateD+20.0D+14.06.0
2022 OH SenateR+8.0R+6.11.9
2022 PA SenateD+1.0D+4.93.9
2022 WI SenateR+6.0R+1.05.0
2021 NJ GovernorD+6.0D+3.22.8
2021 VA GovernorR+1.0R+1.90.9
2020 IL PresidentD+17.0D+17.00.0
2020 MI PresidentD+7.0D+2.84.2
2020 MN PresidentD+9.0D+7.11.9
2020 NJ PresidentD+18.0D+15.92.1
2020 NY PresidentD+28.0D+22.15.9
2020 OH PresidentEVENR+8.08.0
2020 PA PresidentD+6.0D+1.24.8
2020 US PresidentD+8.0D+4.43.6
2020 WI PresidentD+8.0D+0.67.4
2020 IL SenateD+22.0D+16.15.9
2020 MI SenateD+15.0D+1.713.3
2020 MN SenateD+11.0D+5.25.8
2020 NJ SenateD+23.0D+16.36.7
2018 AZ GovernorR+15.0R+14.20.8
2018 CA GovernorD+20.0D+23.93.9
2018 FL GovernorD+1.0R+0.41.4
2018 MI GovernorD+4.0D+9.65.6
2018 MN GovernorD+6.0D+11.45.4
2018 NM GovernorD+12.0D+14.42.4
2018 NY GovernorD+17.0D+24.07.0
2018 OH GovernorEVENR+3.73.7
2018 PA GovernorD+15.0D+17.12.1
2018 WI GovernorD+1.0D+1.10.1
2018 AZ SenateD+1.0D+2.31.3
2018 FL SenateD+1.0R+0.11.1
2018 MI SenateD+16.0D+6.59.5
2018 MN SenateD+20.0D+24.14.1
2018 NM SenateD+14.0D+23.69.6
2018 NY SenateD+28.0D+33.45.4
2018 OH SenateD+14.0D+6.87.2
2018 PA SenateD+17.0D+13.13.9
2018 WI SenateD+11.0D+10.80.2

Accuracy by time to election

24686–9 wk3–6 wk1–3 wk≤1 wkavg miss (pts)
Research Co.All 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 wk844.25-0.0385%

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)
2018214.1R+1.1
2020155.1D+5.1
2022213.4D+0.0
2024254.5D+4.2

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.