VotePredictor
← All pollsters

Digital Research

Graded against the actual result across 37 races (from 58 polls, through 2020).

Races polled
37
Polls
58
Avg miss
7.76 pts
Most recent
2020

Head-to-head vs the VotePredictor model

The fair, apples-to-apples test: on the 32 races Digital 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
Digital Research7.3691%
VotePredictor5.0097%

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 (37)

Each race Digital 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
2020 ME-1 HouseD+34.0D+24.39.7
2020 ME-2 HouseD+18.0D+6.111.9
2020 M1 PresidentD+32.0D+23.18.9
2020 M2 PresidentR+8.0R+7.40.6
2020 ME PresidentD+11.0D+9.11.9
2020 ME SenateD+1.0R+8.69.6
2018 ME SenateR+20.0R+24.84.8
2014 ME GovernorR+3.0R+4.81.8
2014 ME-1 HouseD+34.0D+29.84.2
2014 ME-2 HouseR+5.0R+5.20.2
2012 ME-1 HouseD+19.0D+29.610.6
2012 ME-2 HouseD+19.0D+16.42.6
2012 M1 PresidentD+6.0D+21.415.4
2012 M2 PresidentD+7.0D+8.61.6
2012 ME PresidentD+7.0D+15.38.3
2012 ME SenateR+22.0R+17.54.5
2010 ME GovernorR+12.0R+26.714.7
2010 PA GovernorR+1.0R+9.08.0
2010 ME-1 HouseR+4.0D+13.617.6
2010 ME-2 HouseD+4.0D+10.36.3
2010 PA-10 HouseR+9.0R+10.41.4
2010 PA-11 HouseD+8.0R+9.417.4
2010 PA SenateR+4.0R+2.02.0
2008 ME-1 HouseD+21.0D+9.811.2
2008 ME-2 HouseD+41.0D+34.96.1
2008 ME PresidentD+21.0D+17.33.7
2008 ME SenateR+12.0R+22.710.7
2006 ME GovernorD+17.0D+7.99.1
2006 ME-1 HouseD+44.0D+29.514.5
2006 ME-2 HouseD+37.0D+41.04.0
2006 ME SenateR+60.0R+53.76.3
2004 ME PresidentD+3.0D+9.06.0
2002 ME GovernorD+19.0D+5.713.3
2002 ME-2 HouseD+9.0D+4.05.0
2002 ME SenateR+29.0R+16.912.1
2000 ME PresidentR+5.0D+5.110.1
2000 ME SenateR+56.0R+37.918.1

Accuracy by time to election

24686–9 wk3–6 wk1–3 wk≤1 wkavg miss (pts)
Digital 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
1–3 wk198.39+3.3295%
3–6 wk227.78+2.0486%
6–9 wk156.36+0.21100%

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)
2002310.2D+2.1
200648.5D+3.3
200847.9D+7.9
2010229.4D+6.2
2012124.8R+2.7
201432.1D+2.1
202067.1D+6.9

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.