VotePredictor
← All pollsters

Field Research Corp. (Field Poll)

Graded against the actual result across 16 races (from 34 polls, through 2014).

Races polled
16
Polls
34
Avg miss
5.59 pts
Most recent
2014

Head-to-head vs VotePredictor Elections

The fair, apples-to-apples test: on the 10 races Field Research Corp. (Field Poll) actually polled, how its final poll's margin compared to what VotePredictor Elections predicted for those same races.

ModelAvg miss (pts)Called right
Field Research Corp. (Field Poll)2.74100%
VotePredictor Elections2.24100%

VotePredictor Elections 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 (16)

Each race Field Research Corp. (Field Poll) 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
2014 CA GovernorD+21.0D+19.91.1
2012 CA PresidentD+15.0D+23.18.1
2012 CA SenateD+21.0D+25.04.0
2010 CA GovernorD+10.0D+12.92.9
2010 CA SenateD+8.0D+10.02.0
2008 CA PresidentD+22.0D+24.12.1
2006 CA GovernorR+16.0R+17.01.0
2006 CA SenateD+22.0D+24.42.4
2004 CA PresidentD+7.0D+9.92.9
2004 CA SenateD+19.0D+19.90.9
2003 CA GovernorR+10.0R+17.17.1
2002 CA GovernorD+7.0D+4.92.1
2000 CA PresidentD+5.0D+11.86.8
2000 CA SenateD+20.0D+19.30.7
1998 CA GovernorD+14.0D+19.65.6
1998 CA SenateD+9.0D+10.01.0

Accuracy by time to election

24686–9 wk3–6 wk1–3 wk≤1 wkavg miss (pts)
Field Research Corp. (Field Poll)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
1–3 wk173.61-1.46100%
3–6 wk87.31+1.5775%
6–9 wk87.93+1.7863%

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)
199859.4R+9.4
200052.9R+1.8
2003316.4D+16.4
200452.1R+2.1
200643.5D+2.3
201047.0R+7.0
201243.5R+2.6

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.