The results of election night 2020 were not as conclusive as many anxious Americans had hoped. But, for me, one long-running debate was settled for good: it’s time to retire the New York Times’ needle—the graphic display that shows the likely winner of an election.
I’ll say right up front that I have been a strong partisan in this particular fight. In 2016, I found that the whipsawing gauges, which were programmed to be extra bouncy, accomplished nothing but to induce high anxiety. (Others disagree, of course, because it’s the internet. I also pronounce “gif” with a soft G. Don’t @ me.)
On Monday, the Times’ polling guru Nate Cohn announced that the needles were returning for 2020, but only sort of: instead of a running gauge predicting the ultimate winner of the presidential election, the Times limited it to just three states: Florida, Georgia, and North Carolina. Why those three states? Because they were releasing the kind of data that the needle’s algorithms require. Cohn wrote that they were restoring the needles because they allowed for a simple presentation of complex electoral analysis, in real time.
When Florida began releasing votes soon after polls closed, all three swung toward Trump. A little after 9pm, Cohn explained on Twitter, in a fairly complicated walk through the models used, that he thought they were being too pessimistic.
Throughout the night, Cohn continued to interpret results in terms not of the election outcome, but of the needle: “Trump’s lead in Georgia down to 2.5 points; needle unmoved on the news.”
This seemed a strange use of his time as, in the run-up to the election, Cohn’s polling blog was an exceptional resource for making sense of the deluge of survey data released each day, and was nuanced and hedged enough in its explanations that it is difficult to criticize even in the wake of results that have again left people questioning the reliability of polling.
The Times did, in fact, turn off this year’s needles at 6am on Wednesday morning. It removed all traces of them from its election results pages.
It should be for good. Instead of educating readers on how to understand its needle, the Times should continue to educate readers on how to understand the election results themselves. They have exceptionally skilled human beings, like Cohn, who can do a much better job than a faux dial.
Gabriel Snyder is a contributing editor to CJR.