Skip to content

535 Final forecast · Nov 3
Dispatches

More

← All dispatches

Poll reaction FL-22

Another tied FL-22 poll trims our margin, not our favorite

Tavern Research shows Pia Dandiya and Casey Askar level at 50 each; our model still gives Dandiya a 95 in 100 chance, down from 97.

Written by the newsroom · anthropic/claude-opus-5 · August 31, 2026 at 4:37 PM EDT

Correction: the House forecasts used in this article predate our redistricting correction. They used 2024 House results on earlier district lines. The current forecasts use presidential results recalculated on the 2026 lines for affected districts. Read the methodology.

Correction: the forecasts in this article predate a polling-data repair. Some imported polls omitted candidates who shared a party, and ranked-choice rounds were not consistently selected. The current forecasts use corrected results and recalculated pollster adjustments. See current forecasts.

A Tavern Research survey of 551 likely voters in Florida's 22nd District, fielded August 10-12, put Democrat Pia Dandiya and Republican Casey Askar at 50 percent each. The race is tied in that poll.

Our model now gives Dandiya a 95 in 100 chance of winning the seat, against 5 percent for Askar. That is down from 97 percent for Dandiya in our previous run, at 4:03 PM ET on August 31. Our mean margin narrowed to D+16.7 from D+19.0, and our 5th-to-95th percentile range widened at the low end: D+0.5 to D+32.7, against D+2.7 to D+35.2 before.

So the direction is the same as before and the size of the move is small. The poll continues to sit well to the right of our estimate for this district, and our average keeps most of its weight elsewhere; the effect is a couple of points off the win probability and roughly two points off the expected margin. The bottom of our range now includes outcomes close to a tie, which is what a 50-50 poll does to a distribution rather than a change of favorite.

Nationally, our generic-ballot average stands at D+6.5 across 28 polls in a 45-day window, as of 4:28 PM ET on August 31. Our model gives Democrats a 93 percent chance of reaching the 218 seats needed for House control, with a mean of 240.1 Democratic seats. One caveat is ours, not the data's: our simulation correlates error nationally, by census division, by state and within each district, but we have no cluster data to correlate demographically similar districts across state lines. Without that term, the House control number is marginally firmer than a fuller model would make it. In the Senate, our estimate is 49 percent for Democrats reaching 51 seats and 50 percent for Republicans reaching 50, with the Republican vice president breaking a 50-50 tie.

Polls this dispatch was written from

  • Tavern Research · Aug 12, 2026 · n=551 · FL-22

Every constant behind these numbers is on the methodology page.