Hot Posts

6/recent/ticker-posts

Do the polls matter this close to the presidential election?


It’s hard to beat an addiction, and so like any other political junkie, I am assiduously checking RealClearPolitics and other polling sites constantly. But at this point, it feels almost like habit. Like a smoker who is experiencing declining pleasure from cigarettes but is simply used to holding on to a small stick between his fingers and taking drags throughout the day.

The overarching theme of the race for weeks has been that it is a tossup. It’s quite easy to go through the data and make the case for a Donald Trump victory or argue why Vice President Kamala Harris will win. Either could end up with a comfortable Electoral College margin, or the margin could come down to a few thousand votes in a single state.

According to the RCP average, Harris is up by 1 point nationally, which, given her expected margins in big states like California and New York, would be consistent with a Trump electoral victory. RCP also has Trump ahead by roughly a point in the battleground states. A Washington Post poll released this morning finds Harris with a marginal advantage in the battlegrounds — but the surveys were taken in the first half of October, and thus wouldn’t reflect any more recent momentum toward Trump. Either way, they are still statistically tied.

Looking at another metric — favorability — provides some evidence for the turn against Harris. She was an historically unpopular VP before the July switcheroo with President Biden and then benefitted from a partisan rallying effect and a media honeymoon. But now it appears her favorable ratings are trending back into negative territory — though still not as bad as they were before she was inserted as the Democratic nominee.

But all of this just brings me back to the original point. Which is, all the polls right now are demonstrating a race that is statistically tied. We can argue about whether that suggests a Trump advantage given he has over-performed polling in the past, or whether polls have already corrected (or even overcorrected) for past failures. On The Editors podcast, I have been giving Harris the narrowest of edges just based on the fact that I assume the more traditional and professional Democratic turnout machine will outperform the more ad hoc and experimental Trump operation. But I confess, it’s more of a hunch or educated guess.

Ultimately, as long as polls keep showing the race to be statistically tied both nationally an in all the key states, I find that they are less useful.

Though it won’t stop me from looking.

إرسال تعليق

0 تعليقات