One Nation's primary vote has surged from ~9% in mid-2025 to 23.5% in my latest poll aggregation. That's nearly 4x their 2025 federal election result of 6.4%.
The trajectory is remarkable. Steady climb through late 2025, accelerating from October onward, then plateauing around 23% from February 2026. The leadership changes (Taylor for the Libs, Canavan for the Nats) coincide with the plateau rather than the surge itself.
Pollster house effects tell an interesting story. YouGov (+2.3), DemosAU MRP (+2.1) and Redbridge (+1.9) show ONP highest. Resolve (-1.6), Freshwater (-1.5) and Newspoll (-1.1) show them lowest. A spread of nearly 4 points between houses.
The model estimates observation error (sigma_obs) at about 1.9 points, meaning individual polls typically sit within ±2 points of the underlying trend (after adjusting for house effects). Nu sits around 20, so the StudentT likelihood has moderately heavy tails. Outlier polls don't overly distort the estimate.
Roy Morgan has a systemic issue: residuals show a mean shift of +1.99 points (p=0.009) and a clear upward trend. Compared with the other pollsters, Roy Morgan was underestimating ONP early, now overestimating. Put another way, Roy Morgan has found a larger swing to One Nation than the other polling houses. Worth watching.
YouGov shows variance decreasing over time (p=0.020). Early readings were volatile, recent ones tighter. Small sample (n=7) so treat cautiously, but the pattern is consistent with a pollster recalibrating as ONP became a bigger story.
Bottom line: ONP polling in the low-20s is now a robust finding, even accounting for house effects and typical poll noise. What happens between now and the next election is another question entirely.
No comments:
Post a Comment