VotePredictor
← All pollsters

Grove Insight

Graded against the actual result across 45 races (from 58 polls, through 2012).

Races polled
45
Polls
58
Avg miss
4.39 pts
Most recent
2012

Head-to-head vs the VotePredictor model

The fair, apples-to-apples test: on the 44 races Grove Insight actually polled, how its final poll's margin compared to what the VotePredictor model predicted for those same races.

ModelAvg miss (pts)Called right
Grove Insight4.8489%
VotePredictor3.5080%

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 Grove Insight 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 AZ-2 HouseD+14.0D+0.813.2
2012 CA-52 HouseD+5.0D+2.42.6
2012 CO-3 HouseR+3.0R+12.39.3
2012 NY-19 HouseR+2.0R+0.81.2
2012 NY-24 HouseD+8.0D+9.31.3
2012 CO PresidentD+3.0D+5.42.4
2012 FL PresidentD+1.0D+0.90.1
2012 IA PresidentD+3.0D+5.82.8
2012 MI PresidentD+7.0D+9.52.5
2012 NC PresidentD+3.0R+2.05.0
2012 NH PresidentD+3.0D+5.62.6
2012 NV PresidentD+6.0D+6.70.7
2012 OH PresidentD+4.0D+3.01.0
2012 WI PresidentD+6.0D+6.80.8
2012 NV SenateR+1.0R+1.20.2
2010 OR GovernorD+4.0D+1.52.5
2010 DE-1 HouseD+18.0D+15.72.3
2010 GA-8 HouseD+12.0R+5.417.4
2010 KY-6 HouseD+14.0D+0.313.7
2010 MN-1 HouseD+16.0D+5.310.7
2010 NC-7 HouseD+11.0D+7.43.6
2010 NY-20 HouseD+13.0R+1.514.5
2010 OR-4 HouseD+14.0D+10.93.1
2010 OR-5 HouseD+13.0D+5.37.7
2010 PA-12 HouseD+7.0D+1.65.4
2010 PA-15 HouseR+2.0R+14.612.6
2008 CO-4 HouseD+9.0D+12.43.4
2008 MD-1 HouseEVEND+0.80.8
2008 MI-9 HouseD+10.0D+9.50.5
2008 MN-6 HouseR+4.0R+3.01.0
2008 NJ-3 HouseD+8.0D+4.23.8
2008 PA-11 HouseD+8.0D+3.34.7
2008 AZ PresidentR+4.0R+8.54.5
2008 NV PresidentD+2.0D+12.510.5
2008 OR PresidentD+13.0D+16.33.3
2008 NM SenateD+16.0D+22.76.7
2008 OR SenateD+5.0D+3.41.6
2006 AZ-5 HouseD+3.0D+4.01.0
2006 CT-2 HouseD+4.0EVEN4.0
2006 CT-5 HouseD+5.0D+12.97.9
2006 NY-20 HouseD+3.0D+8.95.9
2006 OH-1 HouseD+9.0R+4.513.5
2006 PA-8 HouseD+4.0D+0.63.4
2006 WA-8 HouseR+1.0R+2.91.9
2002 AZ-1 HouseR+5.0R+3.61.4

Accuracy by time to election

24686–9 wk3–6 wk1–3 wk≤1 wkavg miss (pts)
Grove InsightAll 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 wk41.79-2.49100%
1–3 wk223.43-1.6486%
3–6 wk204.78-0.9690%
6–9 wk126.38+0.2383%

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)
200685.9R+0.3
2008143.4R+0.4
2010118.5D+8.5
2012242.7D+0.6

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.