Blum & Weprin Associates
Graded against the actual result across 10 races (from 27 polls, through 2010).
Every race (10)
Each race Blum & Weprin Associates 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 |
|---|---|---|---|---|
| 2010 TX Governor | R+12.0 | R+12.7 | 0.7 | ✓ |
| 2006 TX Governor | R+23.0 | R+9.2 | 13.8 | ✓ |
| 2002 NY Governor | R+16.0 | R+13.7 | 2.3 | ✓ |
| 2002 TX Governor | R+12.0 | R+17.8 | 5.8 | ✓ |
| 2002 TX Senate | R+9.0 | R+12.0 | 3.0 | ✓ |
| 2000 NY President | D+20.0 | D+24.2 | 4.2 | ✓ |
| 2000 NY Senate | D+7.0 | D+12.0 | 5.0 | ✓ |
| 1998 NY Governor | R+26.0 | R+14.1 | 11.9 | ✓ |
| 1998 TX Governor | R+47.0 | R+37.1 | 9.9 | ✓ |
| 1998 NY Senate | D+4.0 | D+14.2 | 10.2 | ✓ |
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 | 8 | 4.44 | +0.16 | 100% |
| 1–3 wk | 5 | 5.59 | +0.52 | 100% |
| 3–6 wk | 10 | 9.33 | +3.59 | 80% |
| 6–9 wk | 4 | 5.02 | -1.13 | 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) |
|---|---|---|---|
| 1998 | 6 | 14.5 | R+14.5 |
| 2000 | 12 | 4.1 | R+3.6 |
| 2002 | 6 | 3.4 | D+2.7 |
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.