co/efficient puts Hinson up four in Iowa, and our estimate ticks back

A co/efficient survey of 831 likely voters in Iowa, fielded September 14-16, has Republican Ashley Hinson at 46 percent and Democrat Josh Turek at 42 percent, with 2 percent for another candidate. The payload gives no margin of error for the poll, and a four-point gap in a sample that size is the kind of separation that leaves the race close rather than settled.

Our forecast moved slightly toward Hinson. The model now gives Turek a 42 in 100 chance of winning the seat, against 58 percent for Hinson. In our previous run, at 2:32 PM ET today, those figures were 44 percent and 56 percent. The mean margin shifted from R+0.9 to R+1.2. The spread around it is essentially unchanged: our 5th-to-95th percentile range runs from R+11.0 to D+8.7, which is to say a comfortable Republican win and a clear Turek win are both inside the interval 48 days out.

That is a small step back from the direction this race has been moving. We published on September 14 that our Iowa estimate had moved 13 points toward Turek over a week, and on September 15 that a tied Cygnal poll left the estimate where it was. The co/efficient numbers are the first in that sequence to nudge our margin the other way, and they nudge it by three tenths of a point. This is a race our model still treats as a genuine tossup leaning Republican, in a seat with no incumbent on the ballot.

Iowa matters to the chamber arithmetic. Our model gives Democrats a 52 in 100 chance of winning Senate control and puts their mean seat count at 50.4, just below the 51 they need; Republicans need 50, because the payload notes a 50-50 tie goes to them on the vice president's vote. A one-seat shortfall on average is another way of saying races like this one decide it.