VotePredictor
← All pollsters

Market Shares Corp.

Graded against the actual result across 33 races (from 52 polls, through 2012).

Races polled
33
Polls
52
Avg miss
5.77 pts
Most recent
2012

Head-to-head vs the VotePredictor model

The fair, apples-to-apples test: on the 15 races Market Shares Corp. actually polled, how its final poll's margin compared to what the VotePredictor model predicted for those same races.

ModelAvg miss (pts)Called right
Market Shares Corp.3.9887%
VotePredictor2.2880%

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 (33)

Each race Market Shares Corp. 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 IL-8 HouseD+10.0D+9.50.5
2012 IL PresidentD+19.0D+16.92.1
2010 IL GovernorR+4.0D+0.94.9
2010 IL SenateR+3.0R+1.61.4
2008 IL PresidentD+24.0D+25.11.1
2006 IL GovernorD+15.0D+10.54.5
2006 IL-6 HouseR+4.0R+2.71.3
2006 IL-8 HouseD+19.0D+6.912.1
2004 IA PresidentR+2.0R+0.71.3
2004 IL PresidentD+8.0D+10.32.3
2004 MN PresidentD+2.0D+3.51.5
2004 OH PresidentD+4.0R+2.16.1
2004 WI PresidentD+4.0D+0.43.6
2004 IL SenateD+47.0D+42.94.1
2004 WI SenateD+24.0D+11.212.8
2002 IL GovernorD+10.0D+7.12.9
2002 WI GovernorD+2.0D+3.71.7
2002 IN-7 HouseD+1.0D+9.08.0
2002 IL SenateD+20.0D+22.32.3
2000 IN GovernorD+16.0D+14.91.1
2000 IL PresidentD+10.0D+12.02.0
2000 IN PresidentR+23.0R+15.67.4
2000 MI PresidentD+2.0D+5.13.1
2000 NJ PresidentD+18.0D+15.82.2
2000 OH PresidentD+2.0R+3.55.5
2000 PA PresidentD+5.0D+4.20.8
2000 WI PresidentR+5.0D+0.25.2
2000 IN SenateR+59.0R+34.724.3
2000 WI SenateD+33.0D+24.58.5
1998 IL GovernorR+15.0R+3.611.4
1998 WI GovernorR+41.0R+21.020.0
1998 IL SenateR+10.0R+2.97.1
1998 WI SenateR+3.0D+2.15.1

Accuracy by time to election

24686–9 wk3–6 wk1–3 wk≤1 wkavg miss (pts)
Market Shares Corp.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 wk216.09+1.0286%
3–6 wk226.03+0.2991%
6–9 wk84.30-1.8588%

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)
1998812.1R+12.1
2000156.2R+0.9
200274.5R+0.4
2004104.3D+2.7
200654.6D+4.0
201042.5R+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.