Fleming & Associates
Graded against the actual result across 25 races (from 36 polls, through 2022).
Head-to-head vs VotePredictor Elections
The fair, apples-to-apples test: on the 15 races Fleming & Associates actually polled, how its final poll's margin compared to what VotePredictor Elections predicted for those same races.
| Model | Avg miss (pts) | Called right |
|---|---|---|
| Fleming & Associates | 5.39 | 87% |
| VotePredictor Elections | 5.61 | 87% |
VotePredictor Elections 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 (25)
Each race Fleming & 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 |
|---|---|---|---|---|
| 2022 RI Governor | D+13.0 | D+19.1 | 6.1 | ✓ |
| 2022 RI-2 House | R+6.0 | D+3.7 | 9.7 | ✗ |
| 2018 RI Governor | D+10.8 | D+15.5 | 4.7 | ✓ |
| 2018 RI Senate | D+19.0 | D+23.1 | 4.1 | ✓ |
| 2014 RI Governor | D+6.0 | D+4.5 | 1.5 | ✓ |
| 2012 RI-1 House | D+1.0 | D+12.1 | 11.1 | ✓ |
| 2012 RI-2 House | D+17.0 | D+20.6 | 3.6 | ✓ |
| 2012 RI President | D+21.0 | D+27.4 | 6.4 | ✓ |
| 2012 RI Senate | D+21.3 | D+29.8 | 8.5 | ✓ |
| 2010 RI Governor | D+0.7 | R+10.5 | 11.2 | ✗ |
| 2010 MA-4 House | D+12.0 | D+10.5 | 1.5 | ✓ |
| 2010 RI-1 House | D+6.0 | D+6.0 | 0.0 | ✓ |
| 2010 RI-2 House | D+30.4 | D+28.1 | 2.3 | ✓ |
| 2006 RI Governor | R+9.0 | R+2.0 | 7.0 | ✓ |
| 2006 RI Senate | D+4.0 | D+7.0 | 3.0 | ✓ |
| 2002 FL Governor | R+6.0 | R+12.8 | 6.8 | ✓ |
| 2002 RI Governor | R+3.0 | R+9.5 | 6.5 | ✓ |
| 2002 RI-1 House | D+21.0 | D+22.6 | 1.6 | ✓ |
| 2002 RI Senate | D+59.0 | D+56.9 | 2.1 | ✓ |
| 2000 FL President | R+5.0 | EVEN | 5.0 | ✓ |
| 2000 RI President | D+15.0 | D+29.1 | 14.1 | ✓ |
| 2000 FL Senate | D+3.0 | D+4.9 | 1.9 | ✓ |
| 2000 RI Senate | R+16.0 | R+15.7 | 0.3 | ✓ |
| 1998 RI Governor | R+4.0 | R+8.8 | 4.8 | ✓ |
| 1998 FL Senate | D+30.0 | D+24.9 | 5.1 | ✓ |
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 |
|---|---|---|---|---|
| 1–3 wk | 17 | 5.28 | +0.22 | 94% |
| 3–6 wk | 14 | 6.85 | +1.15 | 86% |
| 6–9 wk | 5 | 8.05 | +1.91 | 60% |
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 | 3 | 4.6 | D+4.6 |
| 2000 | 5 | 4.6 | R+3.8 |
| 2002 | 5 | 8.1 | D+7.5 |
| 2010 | 6 | 8.9 | D+8.9 |
| 2012 | 8 | 5.8 | R+5.0 |
| 2018 | 4 | 5.3 | R+5.3 |
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.