Hillary Clinton and Donald Trump are running neck and neck, according to a new election poll out from Quinnipiac University. The presumptive Democratic nominee leads the race 42 to 40 percent, and the poll has a margin of error of plus or minus two percentage points.
The most remarkable thing about this particular result is that it has somehow commanded full stories from the New York Times and CNN, among others. I thought we were beyond this. It has been eight years since Nate Silver started beating into the media’s collective consciousness that tracking the often random fluctuations of individual polls is a singularly terrible way to follow an election. Pollsters frequently have house effects that bias their results toward one candidate or another. Their results move around in weird ways. The most exciting polls, like this Quinnipiac production, are often just outliers. You can get around this issue by averaging together the results of lots of different polls. Lo and behold, Real Clear Politics and HuffPost Pollster still have Clinton comfortably in the lead. None of this is new.
I get why news organizations still trumpet their own polls. They paid for them. They want to reap a little web traffic and attention from the effort—which is probably a worthwhile social trade-off, since it ensures that we have a consistent amount of polling for the averages. And yes, sometimes polls ask interesting questions aside from “who are you voting for” that are worth exploring. Reuters recently found that Trump supporters aren’t too fond of blacks. That result probably deserves some coverage. The first poll after a massive news event like the Orlando shooting? OK, fine, people are going to be curious, might as well write something up.
But random horse-race results? Come on. It is way past the point where we ought to have given up this journalistic crutch—which, by the way, Slate is occassionally guilty of leaning on as well. There’s plenty of other content for us to churn out.
