Iowa’s U.S. Senate campaign has been stuck in a holding pattern for most of the summer. Seven straight opinion polls showed either a tied race between Bruce Braley and Joni Ernst, or one candidate ahead by 1-2 percentage points, well within the margin of error. For weeks, I’ve seen negative ads against both candidates almost every day on television, with a positive spot occasionally sprinkled in. I keep hearing the same anti-Braley or anti-Ernst ads again and again on radio too. Since no major external event has occurred to change the dynamic of the race, I was expecting to see more statistically tied polls at least until the first of three debates to which the candidates have agreed.
Instead, last week Loras College released a poll showing Braley ahead by 45.3 percent to 40.5 percent. Braley had better favorability ratings than Ernst.
Today Quinnipiac released a poll showing Ernst ahead by 50 percent to 44 percent. Ernst had better favorability numbers, led among independents, and had a much bigger lead among men than Braley’s lead among women.
The Loras poll of 1,200 likely voters had a margin of error of plus or minus 2.82 percent. The Q-poll of 1,167 likely voters had a margin of error of plus or minus 2.9 percent. At least one of these polls is way off. Neither Loras nor Quinnipiac have polled in Iowa before this election cycle, so we don’t have a track record to judge them by. For what it’s worth, the available evidence hasn’t convinced me that either Braley or Ernst has a significant lead, and here’s why.
UPDATE: Fox News is out with their latest Iowa poll: Braley and Ernst are at 41 percent each. Notably, the sample includes 36 percent self-identified Democrats, 34 percent Republicans, and 25 percent independent/other.
Continue Reading...