Opinion Dynamics
Graded against the actual result across 18 races (from 37 polls, through 2012).
Head-to-head vs the VotePredictor model
The fair, apples-to-apples test: on the 14 races Opinion Dynamics actually polled, how its final poll's margin compared to what the VotePredictor model predicted for those same races.
| Model | Avg miss (pts) | Called right |
|---|---|---|
| Opinion Dynamics | 4.01 | 71% |
| VotePredictor | 1.96 | 86% |
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 (18)
Each race Opinion Dynamics 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.
| Race | Their call | Result | Miss | Call |
|---|---|---|---|---|
| 2012 MA Senate | D+4.0 | D+7.5 | 3.5 | ✓ |
| 2008 US President | D+7.0 | D+7.4 | 0.4 | ✓ |
| 2004 FL President | D+5.0 | R+5.0 | 10.0 | ✗ |
| 2004 IA President | R+4.0 | R+0.7 | 3.3 | ✓ |
| 2004 MI President | D+2.0 | D+3.4 | 1.4 | ✓ |
| 2004 OH President | R+3.0 | R+2.1 | 0.9 | ✓ |
| 2004 PA President | D+3.0 | D+2.5 | 0.5 | ✓ |
| 2004 US President | D+2.0 | R+2.4 | 4.4 | ✗ |
| 2004 WI President | R+3.0 | D+0.4 | 3.4 | ✗ |
| 2004 FL Senate | D+6.0 | R+1.1 | 7.1 | ✗ |
| 2004 IA Senate | R+47.0 | R+42.3 | 4.7 | ✓ |
| 2004 OH Senate | R+30.0 | R+27.7 | 2.3 | ✓ |
| 2004 PA Senate | R+22.0 | R+10.6 | 11.4 | ✓ |
| 2004 WI Senate | D+14.0 | D+11.2 | 2.8 | ✓ |
| 2000 IL President | R+3.0 | D+12.0 | 15.0 | ✗ |
| 2000 MI President | D+3.0 | D+5.1 | 2.1 | ✓ |
| 2000 PA President | D+3.0 | D+4.2 | 1.2 | ✓ |
| 2000 US President | EVEN | D+0.5 | 0.5 | ✗ |
Accuracy by time to election
Lower is better. Time to election runs right (election week) to left (~2 months out).
By the numbers
| Time to election | Polls | Avg miss | vs field | Called right |
|---|---|---|---|---|
| ≤1 wk | 12 | 3.35 | -0.93 | 58% |
| 1–3 wk | 9 | 3.86 | -1.21 | 78% |
| 3–6 wk | 13 | 2.07 | -3.67 | 100% |
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?
| Year | Polls | Avg miss | Lean (house effect) |
|---|---|---|---|
| 2000 | 7 | 3.3 | R+3.2 |
| 2004 | 23 | 3.0 | R+0.5 |
| 2008 | 6 | 3.1 | R+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.