WE DON’T KNOW who will win the presidential election, but it seems clear who is getting tagged with a big loss: pollsters.
“We still don’t know much about this election — except that the media and pollsters blew it again,” reads the headline over Washington Post media columnist Margaret Sullivan’s early-morning dispatch posted at 5 a.m.
After failing four years ago to pick up on movement toward Donald Trump in the closing days, especially in key swing states that decided the race, pollsters said they learned their lesson.
One upshot of that: lots of state-level polls carried out through last weekend so that late movement in the race could be detected and crucial state findings didn’t get lost in the broader narrative of national polling numbers. Pollsters also said that despite sophisticated modeling that adjusts for a host of demographic factors that may be linked to voter preferences, they had not sufficiently accounted for educational attainment level in 2016, something they made sure to correct for this year.
With all of that said — and with pronouncements there would need to be a polling error greater than that seen four years ago for Trump to prevail — we woke up this morning to a race still standing, as several headlines put it, “on a razor’s edge.”
The election is coming down to the outcome in a few key states — Michigan, Wisconsin, Pennsylvania, Nevada, and Georgia.
Many political junkies had been obsessively checking the poll-crunching site FiveThirtyEight, and its final forecast gave Joe Biden an 89 percent chance of winning while Trump only prevailed in 10 percent of thousands of plausible scenarios run through their model.
Without final results, we still don’t know how far off the state-level polling averages were, but it seems clear they were off in almost all the key states in ways that undercounted Trump’s support, just as they did four years ago.
The headline on FiveThirtyEight editor Nate Silver’s final pre-election write-up on Monday declared Biden the clear favorite, but with the giant-sized caveat that “it’s a fine line between a landslide and nail-biter.”
Many are now probably now down to gnawing on cuticles.
There will be no end to the election and polling post-mortems, but first we do need to get to a mortem.
Read in the context of this morning’s uncertainty, one passage from Silver’s pre-election piece seems pretty on the money right now:
“[W]hat’s tricky about this race is that — because of Trump’s Electoral College advantage, which he largely carries over from 2016 — it wouldn’t take that big of a polling error in Trump’s favor to make the election interesting. Importantly, interesting isn’t the same thing as a likely Trump win; instead, the probable result of a 2016-style polling error would be a Biden victory but one that took some time to resolve and which could imperil Democrats’ chances of taking over the Senate.”
A Democratic Senate takeover looks increasingly unlikely this morning, and Biden could certainly still pull this out — though there’s no landslide in sight.
Silver may have tried to cover all the bases, but the polling world seems to be in for some serious reckoning.
According to Sullivan, “we should never again put as much stock in public opinion polls, and those who interpret them, as we’ve grown accustomed to doing. Polling seems to be irrevocably broken, or at least our understanding of how seriously to take it is.”
One wild card that will surely be scrutinized is the role of massive mail-in voting and how that might have influenced polling.
“We’re, in some sense, always fighting the last war,” said Steve Koczela, president of the MassINC Polling Group, on The Codcast two weeks ago. “We know we fixed what happened in 2016, but it’s just something worth keeping in mind that we don’t really know exactly what 2020 is looking like just yet.”
Koczela added that he was not engaged in extensive presidential race polling this year, “which I have to say is somewhat of a mental health relief for me.”