Targoz Market Research
Graded against the actual result across 28 races (from 31 polls, through 2024).
Head-to-head vs the VotePredictor model
The fair, apples-to-apples test: on the 28 races Targoz Market Research 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 |
|---|---|---|
| Targoz Market Research | 5.86 | 68% |
| VotePredictor | 2.84 | 82% |
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 (28)
Each race Targoz Market 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 |
|---|---|---|---|---|
| 2024 US House | D+3.0 | R+2.8 | 5.8 | ✗ |
| 2024 US President | D+3.0 | R+1.5 | 4.5 | ✗ |
| 2022 AZ Governor | R+2.0 | D+0.7 | 2.7 | ✗ |
| 2022 GA Governor | R+14.0 | R+7.5 | 6.5 | ✓ |
| 2022 OH Governor | R+30.0 | R+25.0 | 5.0 | ✓ |
| 2022 PA Governor | D+6.0 | D+14.8 | 8.8 | ✓ |
| 2022 AZ Senate | D+3.0 | D+4.9 | 1.9 | ✓ |
| 2022 GA Senate | R+2.0 | D+1.0 | 3.0 | ✗ |
| 2022 OH Senate | R+7.0 | R+6.1 | 0.9 | ✓ |
| 2022 PA Senate | R+5.0 | D+4.9 | 9.9 | ✗ |
| 2021 VA Governor | D+3.0 | R+1.9 | 4.9 | ✗ |
| 2020 AZ President | R+1.0 | D+0.3 | 1.3 | ✗ |
| 2020 FL President | D+3.7 | R+3.4 | 7.1 | ✗ |
| 2020 MI President | D+14.0 | D+2.8 | 11.2 | ✓ |
| 2020 MN President | D+9.0 | D+7.1 | 1.9 | ✓ |
| 2020 PA President | D+14.0 | D+1.2 | 12.8 | ✓ |
| 2020 US President | D+1.2 | D+4.4 | 3.2 | ✓ |
| 2020 AZ Senate | D+10.0 | D+2.3 | 7.7 | ✓ |
| 2020 GA Senate | D+2.0 | D+2.1 | 0.1 | ✓ |
| 2020 MI Senate | D+11.0 | D+1.7 | 9.3 | ✓ |
| 2020 MN Senate | D+9.0 | D+5.2 | 3.8 | ✓ |
| 2019 KY Governor | D+19.0 | D+0.4 | 18.6 | ✓ |
| 2019 LA Governor | D+4.0 | D+2.7 | 1.3 | ✓ |
| 2019 MS Governor | R+1.0 | R+5.1 | 4.1 | ✓ |
| 2018 FL Governor | R+1.0 | R+0.4 | 0.6 | ✓ |
| 2018 TN Governor | R+9.0 | R+21.0 | 12.0 | ✓ |
| 2018 FL Senate | R+4.0 | R+0.1 | 3.9 | ✓ |
| 2018 TN Senate | EVEN | R+10.8 | 10.8 | ✗ |
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 | 14 | 4.03 | -0.25 | 50% |
| 1–3 wk | 12 | 8.01 | +2.94 | 83% |
| 3–6 wk | 4 | 6.19 | +0.45 | 75% |
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) |
|---|---|---|---|
| 2018 | 6 | 7.2 | D+5.7 |
| 2019 | 3 | 8.0 | D+8.0 |
| 2020 | 11 | 5.4 | D+4.4 |
| 2022 | 8 | 4.8 | R+4.8 |
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.