In today’s WaPo, contributor Daniel W. Drezner writes:
Nate Silver says I should be nervous about the election. Here’s why I’m not too nervous.
My model of this election is that Trump has a rigid core of supporters but also a hard ceiling on that support. Clinton has more voter support but also more “soft” support. These are voters who become easily disaffected when she has a bad news cycle or two. (It’s also possible that those on the left get disaffected when she appeals to moderate Republicans and vice versa.) So when the race looks close, it’s not because Trump is attracting Clinton voters, it’s because possible Clinton voters are not feeling all that good about Clinton and might choose not to vote — or answer a pollster.
This was also a pattern that was seen briefly in the 2012 election when everyone freaked out about Obama appearing unprepared for Romney turning into a human lie machine about the positions he had taken up until that night. The fallout then was swift and dramatic. A race that was always in Obama’s favor, but was never able to break out of the margin of error suddenly broke out of the static pattern that had been holding all election long. Romney catapulted ahead and the freakout began. Pundits fed the frenzy with stories about how bad this was looking for Obama.
All along, team Obama had internal polling that told them not to flip out because they weren’t seeing any real change. The public perception was that Obama came back and clobbered Romney in the ensuing debates and restored his standing and went on to win the election, but the hindsight wisdom is that nothing truly changed through all of that mayhem. There was just a lot of noise from Democrats who momentarily lost their enthusiasm.
What is believed to have happened after the first debate is that Obama’s supporters were shell shocked and so depressed by the abrupt reversal of fortune that the likely voter screens skewed hard to the Republican side and resulted in bad polling numbers for Obama, which further depressed his base and led to more bad polling numbers. Sound familiar?
Let’s review what’s happened since Labor Day:
- Pollsters switched from the Registered Voter model to the Likely Voter model. This results in tightening naturally.
- Trump suddenly stops being Trump against all odds and his numbers start improving… more tightening.
- Clinton has a health scare and some mis-steps… squeeze.
- Her already nervous base gets more uncertain. People start pointing fingers. The downers start crawling out of the woodwork and become emboldened. There’s infighting and that’s worth a little bit of tightening too. It’s getting hard to breathe.
- It’s psychologically hard to suddenly think you’re losing after being way ahead for so long and we start seeing examples of loss aversion behavior with people prematurely trying to rationalize life under President Trump. That’s great for dystopian fiction writers, but not good for morale. Tighten it up a little more please.
- The tightening leads to more of the above which gets us... even more tightening.
Depressed voters equals more voters who either don’t pass the LV screens or avoid answering the phone if they see that it’s a pollster. A lot had been made out about the impossibly White and Republican crosstabs of some of the polls that came out right after Hillary’s pneumonia. People called out the pollsters as having lost it.
The polling was probably legit. They just detecting a change in the behavior of their respondents. There was that one A- 538 pollster that went so far as to say they expected the voter model in Ohio to resemble the 2004 where Kerry lost to Bush. That matches a electorate where millenials and minorities stay home and a healthy chunk of Latinos go Republican, which makes no sense in the grander scheme of things nor does it jive with the previous 9 months.
But this tightening has a silver lining according to Drezner:
There is a bevy of voters who are not jazzed by Clinton but are petrified by a Trump presidency. Once polls start to show that it’s close, they will decide to vote for Clinton or say so in a poll. When the lead expands, they get more complacent and disaffected by Clinton’s flaws.
The bottom line is that Hillary had some of her worst weeks exactly when Trump had a few of his best and yet she’s still ahead in all of the aggregate pollster rankings. Keep your heads up people and if you’re not volunteering or donating, then your job to is talk people off of the ledge and arrest the self perpetuating loss in enthusiasm that we’re seeing.
The fundamentals are still on our side. Obama is enjoying favorable approval ratings. Middle class income is starting to rebound sharply and everyone did better economically under Obama even if they don’t realize it. Trump is still Trump and has a piss poor team of surrogates that aren’t fit to carry water for the likes of the POTUS, FLOTUS, VPOTUS, Bernie, Warren, and a whole bunch of high ranking Republicans who are willing to speak out against Trump.
For crying out loud, even the The Dallas Morning News, came down on the side of Hillary Clinton with an endorsement. You have to go back 75 years for their last Democratic endorsement and back then the Democrats were the party of racism and White grievance. We did not suddenly begin to fall in love with Trump as a nation.
Yes, this is a closer race than any of us could have thought, but the fundamentals are still on our side. Don’t hide. Don’t hang your head no matter where the polling drifts from week to week. There’s no room for complacency, but we’re in better shape than the numbers look right now. Buck up and straighten your spine and fix the posture of those around you.
---
Last thing to add… the 538 Podcast mentioned a couple of days ago that polling was heavy right after Labor Day, but has since tapered off as the pollsters are holding their horses for the first debate, which means that Hillary’s bad week is baked in pretty good and her rebound is getting muted.