COLUMBUS, Ohio – Clout Research polling in several political races around the country on Tuesday showed another round of perfect work. Several state senate races in Michigan, including some field races with several candidates, came through right on the mark set by Clout in pre-election polling. What is most remarkable in these races is that, even in races where large chunks of the electorate tells Clout they are undecided, on Election Day, the results mirrored Clout’s polling in not only the same order of candidates finishing, but also by about the same margins. Of course, the precise numbers aren’t the same as in the polling because the undecideds, obviously, have decided. But the margins and the ratios are right in line. We are very pleased with that work.
Perhaps the most notable for Clout was the polling in KS CD1, where incumbent Tim Huelskamp lost by 13 points. Clout conducted two surveys there in the run-up to the election – one about a month before the election, and at the start of some very intense campaigning by both campaigns and by outside groups. That poll showed the Congressional Republican Primary Election in a dead heat, but a subsequent survey a couple weeks later – fielded a couple of weeks before Election Day – caught the wave that was building in favor of challenger Roger Marshall. We found Marshall had a seven-point lead, and in the following two weeks, he continued to build on that momentum, ending up with a 13-point victory.
The eventual success of the Marshall campaign was indicated in one often-overlooked factor: Huelskamp’s unfavorability rating was almost double that of Marshall. The incumbent clearly had a part of the Republican electorate that did not like him, and he could not take advantage of the fact that his challenger was much less well-known. Usually, an incumbent can paint an unknown challenger as “too risky” or “too inexperienced”, but it is difficult to paint a negative picture of your opponent when your own negatives are too high – and that is a key factor that likely doomed Huelskamp in yesterday’s election.
Nevertheless, Clout was right on the mark in first identifying the standing of the race heading into the final month, and then in identifying the trend toward Marshall when the election had finally started to take a firm shape.
There is risk in releasing polls a month or several weeks before an election, when campaigns are just ramping up into their fastest gears and the political nastiness has yet to hit its peak intensity. Many pollsters – including Clout Research – are hotly criticized because of polling made public weeks before an election in very fluid races don’t match the final results on Election Day, but it is a risk we take in the interest of public information. The more that people understand about the ebbs and flows of public opinion in political campaigns and the factors that can change electorates in a heartbeat, the better off the entire industry will be. At Clout, we will continue to explain our findings and how we reach them in as many races as possible.