Landmark Communications
Graded against the actual result across 19 races (from 45 polls, through 2022).
Head-to-head vs VotePredictor Elections
The fair, apples-to-apples test: on the 19 races Landmark Communications actually polled, how its final poll's margin compared to what VotePredictor Elections predicted for those same races.
| Model | Avg miss (pts) | Called right |
|---|---|---|
| Landmark Communications | 3.23 | 79% |
| VotePredictor Elections | 2.82 | 90% |
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 Landmark Communications 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 GA Governor | R+6.6 | R+7.5 | 0.9 | ✓ |
| 2022 GA Senate | R+0.4 | D+1.0 | 1.4 | ✗ |
| 2020 GA President | R+4.1 | D+0.2 | 4.3 | ✗ |
| 2020 GA Senate | R+2.1 | R+1.8 | 0.3 | ✓ |
| 2018 GA Governor | R+1.7 | R+1.4 | 0.3 | ✓ |
| 2017 GA-6 House | D+0.1 | R+3.6 | 3.7 | ✗ |
| 2016 GA President | R+4.0 | R+5.1 | 1.1 | ✓ |
| 2016 GA Senate | R+13.6 | R+13.8 | 0.2 | ✓ |
| 2014 GA Governor | R+6.3 | R+7.9 | 1.6 | ✓ |
| 2014 GA-1 House | R+23.0 | R+21.8 | 1.2 | ✓ |
| 2014 GA-10 House | R+28.4 | R+33.0 | 4.6 | ✓ |
| 2014 GA-12 House | R+1.6 | R+9.5 | 7.9 | ✓ |
| 2014 GA-7 House | R+24.0 | R+30.8 | 6.8 | ✓ |
| 2014 GA Senate | R+4.2 | R+7.7 | 3.5 | ✓ |
| 2010 GA Governor | R+6.9 | R+10.0 | 3.1 | ✓ |
| 2010 GA-2 House | R+4.0 | D+2.9 | 6.9 | ✗ |
| 2010 GA-7 House | R+30.0 | R+34.1 | 4.1 | ✓ |
| 2010 GA-8 House | R+13.4 | R+5.4 | 8.0 | ✓ |
| 2010 GA Senate | R+20.8 | R+19.3 | 1.5 | ✓ |
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 | 19 | 3.88 | -0.36 | 63% |
| 1–3 wk | 12 | 4.96 | -0.10 | 67% |
| 3–6 wk | 11 | 2.43 | -3.27 | 73% |
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) |
|---|---|---|---|
| 2010 | 8 | 5.1 | R+2.8 |
| 2014 | 15 | 6.3 | D+6.1 |
| 2016 | 3 | 0.6 | D+0.3 |
| 2017 | 5 | 4.2 | D+4.2 |
| 2020 | 9 | 2.2 | R+1.4 |
| 2022 | 4 | 1.0 | R+0.1 |
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.