VotePredictor
← All pollsters

Gallup

Graded against the actual result across 38 races (from 120 polls, through 2012).

Races polled
38
Polls
120
Avg miss
4.42 pts
Most recent
2012

Head-to-head vs the VotePredictor model

The fair, apples-to-apples test: on the 29 races Gallup actually polled, how its final poll's margin compared to what the VotePredictor model predicted for those same races.

ModelAvg miss (pts)Called right
Gallup3.9676%
VotePredictor2.5793%

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 (38)

Each race Gallup 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
2012 US PresidentR+1.0D+3.94.9
2008 US PresidentD+11.0D+7.43.6
2006 MD GovernorD+12.0D+6.55.5
2006 RI GovernorR+1.0R+2.01.0
2006 TN GovernorD+40.0D+38.91.1
2006 MD SenateD+15.0D+10.05.0
2006 MO SenateD+4.0D+2.31.7
2006 MT SenateD+9.0D+0.98.1
2006 NJ SenateD+10.0D+16.06.0
2006 RI SenateD+3.0D+7.04.0
2006 TN SenateR+3.0R+2.70.3
2006 VA SenateR+3.0D+0.43.4
2004 CO PresidentR+6.0R+4.71.3
2004 FL PresidentD+1.0R+5.06.0
2004 IA PresidentR+2.0R+0.71.3
2004 MI PresidentD+6.0D+3.42.6
2004 MN PresidentD+8.0D+3.54.5
2004 MO PresidentR+14.0R+7.26.8
2004 NM PresidentR+3.0R+0.82.2
2004 NV PresidentR+9.0R+2.66.4
2004 OH PresidentD+4.0R+2.16.1
2004 OR PresidentD+8.0D+4.23.8
2004 PA PresidentR+4.0D+2.56.5
2004 US PresidentR+2.0R+2.40.4
2004 WA PresidentD+8.0D+7.20.8
2004 WI PresidentR+8.0D+0.48.4
2004 WV PresidentR+6.0R+12.96.9
2004 CO SenateD+1.0D+4.83.8
2004 FL SenateD+1.0R+1.12.1
2003 CA GovernorR+15.0R+17.12.1
2002 AR GovernorR+15.0R+6.18.9
2002 CO GovernorR+30.0R+29.01.0
2002 SD GovernorR+14.0R+14.80.8
2002 AR SenateD+8.0D+7.80.2
2002 CO SenateR+2.0R+4.92.9
2002 MO SenateR+4.0R+1.12.9
2002 SD SenateR+3.0D+0.23.2
2000 US PresidentR+2.0D+0.52.5

Accuracy by time to election

24686–9 wk3–6 wk1–3 wk≤1 wkavg miss (pts)
GallupAll 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 wk293.73-0.5559%
1–3 wk224.56-0.5164%
3–6 wk443.77-1.9766%
6–9 wk256.24+0.0964%

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)
2000204.8R+1.7
200272.9R+1.7
2004444.4R+1.0
2006154.3D+0.3
2008224.8R+3.7
2012103.9R+3.6

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.