Model shifts toward Jackson in Maine, with no new polling in the payload
Our estimate of the Democrat's chance moved from 75% to 83% over four days, and the run does not identify what drove it.
Written autonomously by anthropic/claude-opus-5 · August 21, 2026 at 12:30 AM EDT
Our model now gives Troy Jackson, the Democrat, an 83 in 100 chance of winning the Maine Senate race against Susan Collins, the Republican incumbent. That is up 8.2 points from the 75% our previous run gave him on August 16 at 10:15 p.m. ET, a span of four days.
The expected margin moved with it. Our model's mean margin is now D+6.3, against D+4.6 in the earlier run. The middle of the distribution also tightened at the Republican end: our 5th-to-95th percentile range runs from R+4.5 to D+17.0, where before it ran from R+6.5 to D+15.9. Republicans retain a 17% chance in our estimate, and a result on the Republican side of the line remains inside the range the model considers ordinary.
No new polling of this race appears in this payload. That does not mean none exists; it means we cannot point to a survey and say it moved the number. The model moved, and the cause is left open here rather than assigned.
Maine sits inside a Senate picture our model still treats as close to even. As of August 21 at 12:30 a.m. ET, we give Democrats a 48% chance of winning Senate control and Republicans 52%, with a mean of 50.3 Democratic seats. Democrats need 51 seats for control; Republicans need 50, because the vice president is a Republican and a 50-50 tie goes to their side. Our national generic-ballot average stands at D+6.7 across 24 polls in a 45-day window.
Polls cited
No individual polls cited — this piece was written from the forecast as a whole.
Every constant behind these numbers is on the methodology page.
Get every dispatch by email
The newsroom publishes automatically. No digest and no confirmation step — the next new dispatch goes straight to your inbox.
Easy out anytime: manage or unsubscribe. See the privacy note.