DFM Research
Graded against the actual result across 20 races (from 21 polls, through 2022).
Head-to-head vs VotePredictor Elections
The fair, apples-to-apples test: on the 19 races DFM Research actually polled, how its final poll's margin compared to what VotePredictor Elections predicted for those same races.
| Model | Avg miss (pts) | Called right |
|---|---|---|
| DFM Research | 9.40 | 79% |
| VotePredictor Elections | 5.03 | 84% |
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 (19)
Each race DFM Research 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 |
|---|---|---|---|---|
| 2020 ND Governor | R+32.0 | R+40.5 | 8.5 | ✓ |
| 2020 ND President | R+14.0 | R+33.4 | 19.4 | ✓ |
| 2018 NE-2 House | R+7.0 | R+2.0 | 5.0 | ✓ |
| 2018 PA-18 House | R+3.0 | D+0.3 | 3.3 | ✗ |
| 2018 NE Senate | R+15.0 | R+19.1 | 4.1 | ✓ |
| 2016 MO Governor | D+6.0 | R+5.6 | 11.6 | ✗ |
| 2016 WY-1 House | R+16.0 | R+32.1 | 16.1 | ✓ |
| 2016 MO President | R+9.0 | R+18.6 | 9.6 | ✓ |
| 2016 ND President | R+11.0 | R+35.7 | 24.7 | ✓ |
| 2016 WY President | R+35.0 | R+46.3 | 11.3 | ✓ |
| 2016 MO Senate | EVEN | R+2.8 | 2.8 | ✗ |
| 2014 IA-4 House | R+3.0 | R+23.3 | 20.3 | ✓ |
| 2014 ND-1 House | R+7.0 | R+17.1 | 10.1 | ✓ |
| 2014 NE-2 House | D+5.0 | D+3.3 | 1.7 | ✓ |
| 2012 ND President | R+12.0 | R+19.6 | 7.6 | ✓ |
| 2012 ND Senate | D+4.0 | D+0.9 | 3.1 | ✓ |
| 2008 ND Governor | R+44.0 | R+50.9 | 6.9 | ✓ |
| 2008 ND-1 House | D+25.0 | D+23.9 | 1.1 | ✓ |
| 2008 ND President | D+3.0 | R+8.6 | 11.6 | ✗ |
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 | 7 | 6.40 | +1.33 | 71% |
| 3–6 wk | 7 | 10.00 | +4.26 | 86% |
| 6–9 wk | 7 | 14.12 | +7.97 | 86% |
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) |
|---|---|---|---|
| 2008 | 3 | 6.5 | D+6.5 |
| 2014 | 3 | 10.7 | D+10.7 |
| 2016 | 6 | 12.7 | D+12.7 |
| 2018 | 3 | 4.1 | R+1.4 |
| 2020 | 3 | 14.1 | D+14.1 |
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.