Debrief · 2014 cycle
The 2014 Debrief · post-mortem · permanent archive

How did the polling industry do this cycle?

The 2014 industry MAE landed at 4.8pt — graded across 476 historical races with a known outcome — and 225 polls were analyzed across 86 pollsters. Signed bias −3.06pt: pollsters again systematically over-predicted Democrats.

INDUSTRY MAE
4.8pt
vs midterm avg
SIGNED BIAS
−3.06pt
D-leaning
POLLS · POLLSTERS
225
86 firms
RACES SCORED
476
w/ known outcome
CYCLE 2012 2014 2016 2018 2020 2022 2024
Industry mean absolute error · 2012–2014
2012 4.7pt 2014 4.8pt 2016 6.5pt 2018 4.5pt 2020 5.9pt 2022 4.6pt 2024 4.1pt MAE · 2-WAY MARGIN
Per-pollster scorecard · 2014 · sorted by MAE
Signed bias · ±5.3pt
MAE · 0–5.5pt
0ptimus
+0.93pt
1.0pt
cnnorc
+0.07pt
1.5pt
mitchell-research
−2.06pt
2.5pt
new-england-college
−0.06pt
3.4pt
yougov
−2.94pt
3.4pt
vox-populi-polling
+1.81pt
3.5pt
magellan-strategies
+0.30pt
3.5pt
suffolk-university
+0.72pt
3.9pt
public-policy-polling
−2.93pt
4.0pt
clarity-campaign-labs
−4.06pt
4.1pt
quinnipiac-university
−0.61pt
4.1pt
landmark-communications
−4.19pt
4.2pt
nbc-newsmarist
−4.11pt
4.5pt
boston-globe
+3.47pt
4.7pt
monmouth-university
−4.77pt
4.8pt
rasmussen-reports
−2.71pt
5.0pt
Biggest-miss races · 2014
Race Expected Actual Miss Direction
us-governor-2014-oh D+19.00 D+30.61 -11.6pt R-margin underestimate
us-governor-2014-sd D+36.00 D+45.04 -9.0pt R-margin underestimate
us-governor-2014-ny R+23.00 R+15.13 -7.9pt D-margin underestimate
us-governor-2014-ks R+3.50 D+3.69 -7.2pt D-overestimate
us-governor-2014-ar D+7.00 D+13.95 -7.0pt R-margin underestimate
us-governor-2014-il R+2.00 D+3.92 -5.9pt D-overestimate
us-governor-2014-sc D+9.00 D+14.48 -5.5pt R-margin underestimate
us-governor-2014-md R+1.50 D+3.78 -5.3pt D-overestimate