Nate Silver believes Democrats will not suffer from low turnout in the 2022 midterms

[politicallefty]

I think it's far too early to say. We really can't even begin to imagine the environment for the midterms until next year. That was certainly true of 2010. Remember, Democrats actually gained a seat in a special election in Upstates NY in November 2009. It wasn't until Ted Kennedy's seat fell that I think serious alarms were going off in Democratic circles. It's also worth noting that the unemployment rate was bottoming out around then, but it was still a very slow recovery. The unemployment rate going into the 2010 midterms was 9.5%.

I'm not going to make a prediction because it's too early. Yes, I agree that the House is at extreme risk. The Senate is completely different because of the staggered terms and the map that's up at the time. There are years when both Houses experience a wave, years when it's one or the other, and also years where it's neither. The past four midterms have seen 2 in the first category and 2 in the second category (and of those 2, one of each: Senate wave in 2014 and a House wave in 2018). For this purpose, I would consider a wave to be +5 in the Senate and +25 in the House. To break it down by year since 1960 (the year colour shows the party in power at the time):

DOUBLE WAVE
1994
-House: R+54
-Senate: R+8
2006
-House: D+31
-Senate: D+6
2010
-House: R+63
-Senate: R+6

HOUSE WAVE ONLY
1966
-House: R+47
-Senate: R+3
1974
-House: D+49
-Senate: D+4
1982
-House: D+26
-Senate: D+1
2018
-House: D+41
-Senate: R+2

SENATE WAVE ONLY
1986
-House: D+5
-Senate: D+8
2014
-House: R+13
-Senate: R+9

NO WAVE
1970
-House: D+12
-Senate: R+1
1978
-House: R+15
-Senate: R+3
1990
-House: D+7
-Senate: D+1

PARTY IN POWER WINS
1962
-House: R+1
-Senate: D+4
1998
-House: D+5
-Senate: No Net Change
2002
-House: R+8
-Senate: R+2

Interestingly, this categorization almost divides the elections evenly. That was not by design. When I started sorting the years, I found that a couple didn't quite fit the profile of a non-wave year. They were really outright wins for the party in power. That includes 1962, 1998, and 2002. If you went back, 1934 would be in this category, but even further as a Senate wave for the party in power (and one could argue for a new category if I was going back that far). Both 1962 and 2002 had national security explanations. In 1962, the Cuban Missile Crisis had basically ended very shortly before the midterms and 2002 had an incumbent President riding on high approvals after 9/11. And as we all know, impeachment backfired on Republicans in 1998. Even if they hadn't moved on impeachment, it probably would've been at worst a "No Wave" election for Clinton and the Democrats.

It's not surprising that the Senate-only waves happened during the 6-year midterm of a two-term President. Both Reagan and Obama had significant Senate coattails which were enough to hold for 6 years.

I'm not going to make a prediction this far out, but there's a lot of reason to believe that 2022 could fall into one of the latter categories. Even a "No Wave" scenario could topple both Houses of Congress though, considering the tight margins. For Democrats to hold the House, they probably need to fall into the last category, although passage of HR1 could save their majority under a "No Wave" year. There's no reason to discount Senate Democrats out yet either. Next year is extremely unlikely to be a Senate-only wave. If the economy starts booming and the pandemic is largely behind us, there's reason to hope for a relatively good Democratic year. That said, they're going to have to be on the lower end of a rather ugly bell curve. But as I said before, it's way too early to predict how the midterms are going to go.