Friday, January 20, 2017

Polls Were NOT "Wrong"

Stop blaming polls or calling them "wrong" because only the Huffington Post polling was seriously flawed. Every other model actually offered accurate *ranges* of potential outcomes.

The polls were not wrong. Polls give probability not certainty. They were accurate. If I tell you Hillary Clinton has an 85% chance of winning... hello? She still has a 15% chance of losing. People didn't want to accept that. They assumed 85% = she can't lose. Sean Trende has attempted to explain this with the example of Pennsylvania. The commonwealth was a close election: so close that one percent in both directions did change the winner, but that does not make the polls incorrect.
It wasn't the polls: It was the pundits
What occurred wasn’t a failure of the polls. As with Brexit, it was a failure of punditry. Pundits saw Clinton with a 1.9 percent lead in Pennsylvania and assumed she would win. [Note: Margin of error was 3% in most polling!] The correct interpretation was that, if Clinton’s actual vote share were just one point lower and Trump’s just one point higher, Trump would be tied or even a bit ahead.  
The best polls were within margins of error. I wish people would stop claiming all polls were "wrong" since they were better than most previous elections. This was a close election, as a 3% to 5% margin of error is just what it says -- a margin. People assumed a lot, but Nate Silver and Sean Trende have explained repeatedly the polls were accurate within their margins. We're not going to get 0.1% error from polling.

People simply did not want the "worst case" scenario, and the press didn't explain with maps: "Here are the two extreme outcomes."

No comments:

Post a Comment