VotePredictor
← All pollsters

Normington, Petts & Associates

Graded against the actual result across 49 races (from 52 polls, through 2020).

Races polled
49
Polls
52
Avg miss
8.05 pts
Most recent
2020

Head-to-head vs the VotePredictor model

The fair, apples-to-apples test: on the 47 races Normington, Petts & Associates actually polled, how its final poll's margin compared to what the VotePredictor model predicted for those same races.

ModelAvg miss (pts)Called right
Normington, Petts & Associates7.9855%
VotePredictor5.5379%

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 (49)

Each race Normington, Petts & Associates 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
2020 CA-25 HouseD+6.0R+0.16.1
2020 MN-2 HouseD+18.0D+2.315.7
2020 MO-2 HouseEVENR+6.46.4
2018 CO-6 HouseD+11.0D+11.20.2
2018 VA-7 HouseEVEND+1.91.9
2016 IA-1 HouseR+1.0R+7.76.7
2016 IL-10 HouseD+5.0D+5.20.2
2016 IN-9 HouseR+2.0R+13.711.7
2016 ME-2 HouseD+1.0R+9.610.6
2016 NH-1 HouseD+10.0D+1.38.7
2016 IL PresidentD+17.0D+17.10.1
2016 M2 PresidentD+4.0R+10.314.3
2016 IL SenateD+9.0D+15.16.1
2014 ME GovernorD+2.0R+4.86.8
2014 MA-6 HouseD+12.0D+13.81.8
2014 ME-2 HouseD+8.0R+5.213.2
2014 NH-1 HouseD+2.0R+3.65.6
2014 NH-2 HouseD+6.0D+10.04.0
2012 ME-2 HouseD+25.0D+16.48.6
2012 WI-8 HouseR+6.0R+12.06.0
2010 AZ-5 HouseD+3.0R+8.811.8
2010 CA-20 HouseD+6.0D+3.42.6
2010 CO-4 HouseEVENR+11.111.1
2010 IA-3 HouseD+8.0D+4.23.8
2010 SD-1 HouseD+13.0R+2.215.2
2010 TX-17 HouseR+4.0R+25.221.2
2010 KY SenateD+2.0R+11.513.5
2010 PA SenateR+1.0R+2.01.0
2008 IL-10 HouseR+6.0R+5.10.9
2008 IL-11 HouseD+21.0D+23.92.9
2008 LA-6 HouseD+17.0R+7.824.8
2008 MN-3 HouseD+5.0R+7.612.6
2008 VA-2 HouseR+5.0D+4.99.9
2008 WA-8 HouseD+5.0R+5.610.6
2006 AZ-5 HouseR+1.0D+4.05.0
2006 AZ-8 HouseD+25.0D+12.212.8
2006 CT-4 HouseD+3.0R+3.46.4
2006 IA-1 HouseD+11.0D+11.90.9
2006 IL-6 HouseD+1.0R+2.73.7
2006 IL-8 HouseD+16.0D+6.99.1
2006 PA-10 HouseD+14.0D+5.98.1
2006 TX-17 HouseD+21.0D+17.83.2
2006 MO SenateD+5.0D+2.32.7
2006 NJ SenateD+9.0D+16.07.0
2004 IL-8 HouseR+2.0D+3.45.4
2004 TX-1 HouseD+4.0R+23.827.8
2004 TX-17 HouseD+10.0D+3.86.2
2002 ME-2 HouseD+12.0D+4.08.0
2002 TX SenateR+2.0R+12.010.0

Accuracy by time to election

24686–9 wk3–6 wk1–3 wk≤1 wkavg miss (pts)
Normington, Petts & AssociatesAll 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 wk96.53+1.4656%
3–6 wk318.37+2.6358%
6–9 wk128.37+2.2250%

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)
2004313.1D+9.5
2006105.9D+3.3
2008710.2D+3.5
2010810.0D+10.0
201456.3D+4.0
2016107.4D+6.1
202039.4D+9.4

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.