Regression to the mean?

Drum takes a look at the long-term polling numbers via Sam Wang and essentially makes an argument for regression to the mean.

Roughly speaking, Hillary Clinton has had a steady 3.5 percent lead over Trump all year. Then she got a boost from the Democratic convention, followed by a few bad weeks for Trump. That wore off and she dropped back to a little below where she’s been all along. In the last few days, Clinton has started rising again, and my guess is that over the next few weeks she’ll meander back to her longtime lead of 3.5 percent. Pollwise, the single most remarkable thing about the Clinton vs. Trump race is how stable it’s been ever since the day Trump took his famous escalator ride down to the ground floor of Trump Tower to announce his candidacy.

To the extent that Democrats are panicking, I think it’s because a few weeks ago Clinton was ahead by 7 percent or so, and everyone was licking their chops and wondering if a landslide was possible. It was deflating when that turned out to be a mirage. I got caught up in that a bit too, and it was probably foolish. In reality, it was just a temporary bump and was never likely to last.

We’ll know soon enough, but, honestly, I think an expected regression to the mean makes far more sense than any other prediction for the polls in the coming weeks.  And it strikes me as far more likely than the liberal freak-out scenario of Trump on a continuing rise.


About Steve Greene
Professor of Political Science at NC State

3 Responses to Regression to the mean?

  1. R. Jenrette says:

    While it is truly horrifying to think of Trump becoming President, there is something worse.
    That there is a dangerously large minority of American voters enthusiastically prepared to vote for such a man is terrifying.

  2. rgbact says:

    Or voters started focusing more after Labor Day and pollsters implemented their likely voters screens, which favor Trump alot more than just registered voter polls. And Trump got a new campaign manager who writes his speeches so he talks more substance. I like my theory better.although Wang is always there to make liberals feel better.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: