VotePredictor
← All pollsters

Selzer & Co.

Graded against the actual result across 45 races (from 74 polls, through 2022).

Races polled
45
Polls
74
Avg miss
4.4 pts
Most recent
2022

Head-to-head vs the VotePredictor model

The fair, apples-to-apples test: on the 40 races Selzer & 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
Selzer & Co.3.3283%
VotePredictor3.5880%

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 (45)

Each race Selzer & 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
2022 IA GovernorR+17.0R+18.51.5
2022 IA SenateR+12.0R+12.20.2
2020 IA PresidentR+7.0R+8.21.2
2020 IA SenateR+4.0R+6.62.6
2018 IA GovernorD+2.0R+2.74.7
2016 NC GovernorD+6.0D+0.25.8
2016 FL PresidentR+2.0R+1.20.8
2016 IA PresidentR+7.0R+9.42.4
2016 NC PresidentD+1.0R+3.74.7
2016 OH PresidentR+5.0R+8.13.1
2016 PA PresidentD+9.0R+0.79.7
2016 US PresidentD+3.0D+2.20.8
2016 FL SenateR+10.0R+7.72.3
2016 IA SenateR+23.0R+24.41.4
2016 NC SenateD+2.0R+5.77.7
2016 OH SenateR+17.0R+20.93.9
2016 PA SenateD+2.0R+1.43.4
2014 IA GovernorR+24.0R+21.72.3
2014 IA SenateR+7.0R+8.31.3
2012 CO PresidentD+4.0D+5.41.4
2012 IA PresidentD+5.0D+5.80.8
2012 US PresidentD+6.0D+3.92.1
2010 IA GovernorR+12.0R+9.62.4
2010 IA SenateR+31.0R+31.00.0
2008 IN GovernorR+18.0R+17.80.2
2008 IA PresidentD+17.0D+9.57.5
2008 IN PresidentD+1.0D+1.00.0
2008 MI PresidentD+16.0D+16.50.5
2008 IA SenateD+26.0D+25.40.6
2008 MI SenateD+21.0D+28.87.8
2006 IA GovernorD+9.0D+9.60.6
2006 MI GovernorD+13.0D+14.11.1
2006 IA-1 HouseD+21.0D+11.99.1
2006 IN-7 HouseR+3.0D+7.510.5
2006 MI SenateD+19.0D+15.73.3
2004 IN GovernorR+6.0R+7.71.7
2004 TX-32 HouseR+6.0R+10.34.3
2004 IA PresidentD+3.0R+0.73.7
2004 IN PresidentR+20.0R+20.70.7
2004 IN SenateD+39.0D+24.414.6
2002 IA GovernorD+12.0D+8.23.8
2002 IA SenateD+9.0D+10.41.4
2000 IA PresidentD+2.0D+0.31.7
1998 IA GovernorR+4.0D+5.89.8
1998 IA SenateR+44.0R+37.96.1

Accuracy by time to election

24686–9 wk3–6 wk1–3 wk≤1 wkavg miss (pts)
Selzer & 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 wk282.68-1.6089%
1–3 wk40.84-4.23100%
3–6 wk295.81+0.0783%
6–9 wk136.07-0.0862%

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)
200246.2D+5.5
200475.3D+3.2
2006105.0R+2.5
2008114.1D+0.7
201043.0R+2.9
201241.5R+0.5
201454.2D+3.3
2016164.4D+3.5
202045.4D+5.4
202243.1D+3.1

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.