Our Kansas Senate estimate moves toward Hamilton without new polling

Our model now gives Adam Hamilton, the Democrat, a 14 in 100 chance of winning the Kansas Senate race, up from 6% in our run of August 25. That is a shift of 8.3 points of Democratic win probability over seven days. Roger Marshall, the Republican incumbent, is at 86%.

No new Kansas polling arrived between the two runs. The most recent survey in our file is a Public Policy Polling poll of registered voters sponsored by Senate Majority PAC, fielded August 7-8, which had Marshall at 46% and Hamilton at 45%. Before that, a GBAO poll of 600 likely voters fielded July 8-13 had Marshall at 47% and Hamilton at 43%, and a GQR poll of 500 likely voters fielded April 23-27 had Marshall at 49% and the Democrat at 45%. All three predate the August 25 run. The model moved; the payload does not identify what drove it, and we are not going to guess.

The internals point to wider uncertainty rather than a better position for Hamilton in the center of the distribution. Our mean margin is R+12.0, slightly further toward Marshall than the R+10.1 we had a week ago. But the spread around it grew: our 5th-95th percentile range now runs from R+30.5 to D+6.6, against R+20.8 to D+0.7 in the previous run. A distribution that reaches further in both directions can raise the probability of an upset while the average outcome stays Republican, and that is what happened here.

Nationally, our model gives Democrats a 49% chance of Senate control and Republicans 51%, with a mean of 50.2 Democratic seats. Democrats need 51 seats for control; Republicans need 50, because a 50-50 tie goes to them under a Republican vice president. Kansas remains a long-shot component of that arithmetic.