co/efficient finds Michigan Senate race tied; our forecast eases to 69 percent

A co/efficient survey of 843 likely voters in Michigan, fielded September 21-23, has the open Senate race tied, with Democrat Abdul El-Sayed and Republican Mike Rogers each at 50 percent. A second version of the same question from the same sample, offering another option, puts both candidates at 45 percent, with other choices at zero.

Our model now gives El-Sayed a 69 in 100 chance of winning the seat, down from 71 percent in our previous run earlier today. The estimated mean margin narrowed from D+3.2 to D+2.9. Our 5th-to-95th percentile range runs from R+6.6 to D+12.4, a spread wide enough to include comfortable outcomes for either candidate.

That is a small adjustment, and it does not change the shape of the race our average has been describing. Our published estimate has sat between 65 and 71 percent through September. The new co/efficient numbers land on the closer end of that picture rather than redirecting it: a tie in a single poll is consistent with a modest Democratic edge in the average, and our model treats it that way.

Michigan matters to the chamber arithmetic. Our model gives Democrats a 54 percent chance of winning Senate control and puts their mean seat count at 50.7, just short of the 51 seats they would need for a majority. Republicans need 50, because the vice president is a Republican and a 50-50 tie goes to them. With a mean projection that close to the threshold, a single competitive state carries real weight, and Michigan is one of them.

Forty-one days remain before the November 3 election. Our estimate here is a probability, not a call: a 31 percent chance for Rogers is a regular outcome, not a remote one.