Fairleigh Dickinson University
Graded against the actual result across 31 races (from 56 polls, through 2024).
Head-to-head vs the VotePredictor model
The fair, apples-to-apples test: on the 29 races Fairleigh Dickinson University 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 |
|---|---|---|
| Fairleigh Dickinson University | 5.34 | 93% |
| VotePredictor | 4.73 | 93% |
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 (31)
Each race Fairleigh Dickinson University 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 |
|---|---|---|---|---|
| 2024 NJ Senate | D+18.0 | D+9.6 | 8.4 | ✓ |
| 2021 NJ Governor | D+9.0 | D+3.2 | 5.8 | ✓ |
| 2020 NJ President | D+15.0 | D+15.9 | 0.9 | ✓ |
| 2018 NJ Senate | D+6.0 | D+11.2 | 5.2 | ✓ |
| 2017 NJ Governor | D+15.0 | D+14.1 | 0.9 | ✓ |
| 2016 NJ President | D+14.0 | D+14.1 | 0.1 | ✓ |
| 2016 US President | D+9.0 | D+2.2 | 6.8 | ✓ |
| 2014 NJ Senate | D+16.0 | D+13.5 | 2.5 | ✓ |
| 2013 NJ Governor | R+19.0 | R+22.1 | 3.1 | ✓ |
| 2013 NJ Senate | D+16.0 | D+10.9 | 5.1 | ✓ |
| 2012 NJ President | D+14.0 | D+17.7 | 3.7 | ✓ |
| 2012 NJ Senate | D+14.0 | D+19.5 | 5.5 | ✓ |
| 2010 DE-1 House | D+17.0 | D+15.7 | 1.3 | ✓ |
| 2010 DE Senate | D+21.0 | D+16.6 | 4.4 | ✓ |
| 2009 NJ Governor | D+2.0 | R+3.6 | 5.6 | ✗ |
| 2008 DE Governor | D+34.0 | D+35.5 | 1.5 | ✓ |
| 2008 AK-1 House | D+6.0 | R+5.2 | 11.2 | ✗ |
| 2008 DE-1 House | R+34.0 | R+23.1 | 10.9 | ✓ |
| 2008 AK President | R+18.0 | R+21.5 | 3.5 | ✓ |
| 2008 DE President | D+20.0 | D+25.0 | 5.0 | ✓ |
| 2008 NJ President | D+18.0 | D+15.6 | 2.4 | ✓ |
| 2008 AK Senate | D+4.0 | D+1.2 | 2.8 | ✓ |
| 2008 DE Senate | D+43.0 | D+29.4 | 13.6 | ✓ |
| 2008 NJ Senate | D+20.0 | D+14.1 | 5.9 | ✓ |
| 2006 DE-1 House | R+35.0 | R+18.4 | 16.6 | ✓ |
| 2006 DE Senate | D+34.0 | D+41.5 | 7.5 | ✓ |
| 2006 NJ Senate | D+10.0 | D+16.0 | 6.0 | ✓ |
| 2005 NJ Governor | D+2.0 | D+10.4 | 8.4 | ✓ |
| 2004 NJ President | D+7.0 | D+6.7 | 0.3 | ✓ |
| 2002 NJ Senate | D+9.0 | D+9.9 | 0.9 | ✓ |
| 2001 NJ Governor | D+18.0 | D+14.8 | 3.2 | ✓ |
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 |
|---|---|---|---|---|
| 1–3 wk | 20 | 4.92 | -0.15 | 95% |
| 3–6 wk | 19 | 3.53 | -2.21 | 90% |
| 6–9 wk | 15 | 7.33 | +1.18 | 93% |
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) |
|---|---|---|---|
| 2002 | 3 | 1.3 | R+1.3 |
| 2004 | 6 | 2.8 | R+1.9 |
| 2005 | 5 | 4.0 | R+4.0 |
| 2006 | 6 | 10.2 | R+10.2 |
| 2008 | 13 | 5.7 | D+0.7 |
| 2009 | 3 | 3.9 | D+3.9 |
| 2010 | 4 | 1.7 | D+1.3 |
| 2013 | 4 | 9.1 | D+3.6 |
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.