VotePredictor
← All pollsters

Tarrance Group

Graded against the actual result across 63 races (from 76 polls, through 2024).

Races polled
63
Polls
76
Avg miss
6.49 pts
Most recent
2024

Head-to-head vs the VotePredictor model

The fair, apples-to-apples test: on the 49 races Tarrance Group actually polled, how its final poll's margin compared to what the VotePredictor model predicted for those same races.

ModelAvg miss (pts)Called right
Tarrance Group6.1371%
VotePredictor5.5884%

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 (63)

Each race Tarrance Group 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
2024 MI SenateR+2.0D+0.32.3
2022 IA GovernorR+17.0R+18.51.5
2022 NY-25 HouseD+4.0D+11.37.3
2022 IA SenateR+11.0R+12.21.2
2020 IL-17 HouseD+5.0D+4.10.9
2020 NM-2 HouseR+1.0R+7.46.4
2020 PA-10 HouseR+4.0R+6.62.6
2020 MI SenateD+1.3D+1.70.4
2018 CO-6 HouseD+2.0D+11.29.2
2018 NM-2 HouseR+4.0D+1.95.9
2018 TX-31 HouseR+21.0R+2.918.1
2018 MI SenateD+7.0D+6.50.5
2018 WV SenateD+4.0D+3.30.7
2016 CA-24 HouseR+3.0D+6.89.8
2016 IA-3 HouseR+15.0R+13.71.3
2016 NV-3 HouseR+5.0D+1.36.3
2016 VA PresidentD+9.0D+5.33.7
2014 AZ GovernorR+7.0R+11.84.8
2014 HI GovernorD+3.0D+12.49.4
2014 IA-3 HouseR+2.0R+10.58.5
2014 KS-2 HouseR+12.0R+18.46.4
2014 MA-6 HouseR+1.0D+13.814.8
2014 MN-7 HouseR+1.0D+8.59.5
2014 VA-10 HouseR+12.0R+16.14.1
2012 CA-52 HouseR+4.0D+2.46.4
2012 CA-9 HouseR+1.0D+11.112.1
2012 NV-4 HouseR+10.0D+8.018.0
2012 PA-4 HouseR+22.0R+25.33.3
2012 AZ SenateR+6.0R+3.03.0
2010 CA GovernorEVEND+12.912.9
2010 IA-2 HouseR+1.0D+5.16.1
2010 IL-14 HouseR+6.0R+6.30.3
2010 IL-17 HouseR+1.0R+9.68.6
2010 KY-6 HouseR+1.0D+0.31.3
2010 MS-1 HouseR+11.0R+14.53.5
2010 MS-4 HouseR+2.0R+5.13.1
2010 NM-2 HouseR+9.0R+10.81.8
2010 PA-10 HouseR+5.0R+10.45.4
2010 TN-8 HouseR+15.0R+20.25.2
2010 CA SenateEVEND+10.010.0
2008 CA-50 HouseR+13.0R+5.17.9
2008 FL-16 HouseR+26.0R+20.25.8
2008 IA-4 HouseR+22.0R+21.10.9
2008 MN-1 HouseD+18.0D+29.611.6
2008 OH-7 HouseR+24.0R+16.47.6
2006 NJ-7 HouseR+15.0R+1.513.5
2004 LA-3 HouseR+7.0D+0.57.5
2004 CO SenateD+2.0D+4.82.8
2004 WI SenateD+5.0D+11.26.2
2002 CA GovernorD+3.0D+4.91.9
2002 CO-7 HouseR+2.0R+0.11.9
2002 FL-5 HouseD+2.0R+1.73.7
2002 GA-11 HouseD+4.0R+3.37.3
2002 GA-12 HouseR+15.0R+10.44.6
2002 NH-2 HouseR+15.0R+16.01.0
2002 TX-11 HouseD+8.0D+4.43.6
2002 UT-2 HouseD+1.0D+0.70.3
2002 CO SenateR+6.0R+4.91.1
2000 FL PresidentR+4.0EVEN4.0
2000 MN SenateD+1.0D+5.54.5
2000 NJ SenateR+4.0D+3.07.0
1998 FL GovernorR+12.0R+10.51.5
1998 AR SenateD+8.0D+12.84.8

Accuracy by time to election

24686–9 wk3–6 wk1–3 wk≤1 wkavg miss (pts)
Tarrance GroupAll 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 wk244.64-0.4375%
3–6 wk357.59+1.8569%
6–9 wk147.77+1.6264%

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)
200035.2R+5.2
2002103.6D+1.7
200435.5R+5.5
2008715.2D+5.5
2010165.9D+1.6
201258.6R+7.2
2014107.7R+1.8
201655.5R+4.0
201856.9R+6.4
202052.1D+2.0
202233.3R+1.5

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.