Where the Election Sits Post-Labor Day

House Minority Leader Kevin McCarthy (R-CA)
and House Speaker Nancy Pelosi (D-CA)

Labor Day is frequently used as a prime benchmark in American politics, almost always a symbolic one rather than a practical one.  Generally, where the polls stand on Labor Day is a strong indication of where they might stand come Election Day, though every rule has its exceptions, and there are some doozies here (1980 & 2014 come to mind).  But I think it's worth noting how odd the political picture is this Labor Day, and why 2022 is shaping up to be the most unusual midterm election in a generation.

Generally, during a midterm election, the party out-of-power gains seats, particularly during a president's first midterm.  Bill Clinton saw the Republican Revolution in 1994, Barack Obama suffered agonizing losses in 2010, and Donald Trump saw the election of the House that would impeach him (twice) in 2018. The only recent exception to this rule was 2002, when George W. Bush won seats in both the House & the Senate (even gaining back a Senate majority) but did so with record-high approval ratings coming after the September 11th terrorist attacks.  Joe Biden is not popular, certainly not compared to Bush in 2002.  While his approval ratings have gone up markedly since a low-point in mid-July, he still has a 43% approval rating according to 538's aggregate tracker, hardly anything to write home about, and certainly not good enough to hold either house of Congress in a normal situation.

But as I'll say a few times in this article, 2022 is not shaping up to be a normal election, and one of the first ways you can tell this is by looking at Biden's 43% approval rating with the generic ballot numbers, which show the Democrats leading by about a point, and they've held that lead for almost a month now, it getting bigger after nearly a year of Republicans leading in that metric.  This becomes interesting because, despite some claiming otherwise, Democrats don't need a huge lead in the generic congressional margin to hold the House.  Frequently it's cited that Democrats need absurdly high numbers, like 4-7 points, to be able to win the House, but that's not really the case (I think this partially stems from people claiming this about the electoral college, but the electoral college is far more slanted toward the Republicans than the US House is).  The median seat using presidential numbers from 2008 is Michigan's 8th district, held by Democratic Rep. Dan Kildee, and it went for Biden by 2.1-points.  Essentially, if Democrats held every Biden seat, they'd not only win the House but would actually gain seats compared to 2020.  This still isn't likely to happen (for a variety of factors), but a generic ballot that is sitting at a 1-2 point win for the left is not something like 2012, where Democrats won the House popular vote by over a point but were still 17 seats shy of a majority.  In fact, the closer you get to a 2-point win (if that's truly what happens), the more likely it is that the control of the House is a jump ball.

Pundits are remiss to say this out-loud for two big reasons.  The first is that it's not what's supposed to happen.  2002 is an anomaly, and one that it's embarrassing to compare an election to...it's the equivalent to saying that Democrats are competitive in every red state because Doug Jones won, ignoring that Jones won for a variety of factors that are not repeatable (hence, Doug Jones losing in 2020).  Saying that the midterms will behave differently is going to take a lot of courage on behalf of pundits because of the second reason: they got 2016 & 2020 wrong.  While there are a variety of factors as to why 2016 was called incorrectly (the biggest being that late polls could not catch James Comey's actions against Hillary Clinton less than two weeks before the election) and why 2020 wasn't (the pandemic made polling more difficult than usual, and the lack of in-person campaigning for one side & not the other had an impact), the fact is that most pundits & polls indicated that Hillary Clinton would win in 2016 and that Joe Biden would win by a bigger margin than he did in 2020.  Going out on a limb for a third election, particularly in going against the fundamentals in a way that you didn't need to do so in 2016/20 (where the deck was stacked against Democrats, but not to a degree that it should be in 2022), feels like too much of an ask and I suspect if the Democrats are to pull off the upset of winning the House in November (and even by my calculations, it'd still be an upset), no one is going to call this until the networks do.

Senate Maj. Leader Chuck Schumer (D-NY) with
Senate Minority Leader Mitch McConnell (R-KY)

The same appears increasingly less likely with the Senate.  I'm still undecided on if I'll do one more State of the Senate before we do our Election Night Guides in October (they take forever to write, and I'm on vacation next week), but right now it appears that the Senate is still in reach for Republicans, but you'd bet on the Democrats if you were forced to place a wager.  There are for sure two tossup seats (Nevada & Georgia), but the rest of the map remains murky. Republicans seem to have given up in Arizona, which feels insane for a state that Joe Biden won by 0.3 points, but Blake Masters (R) is such a terrible candidate and is getting hammered by Sen. Mark Kelly (D) to the point that you have to favor Kelly even if you can't count out Masters.  The same is true in Pennsylvania, where Lt. Gov. John Fetterman is the mild favorite against TV personality Mehmet Oz (R), the latter of whom has ran the worst campaign since Christine O'Donnell in 2010.  Oz could win, but Fetterman has the momentum.

Fetterman would be a pickup, which means that Republicans need both of NV/GA in order to take the majority, but there's a third problem for the GOP-they have other vulnerabilities that Democrats don't have.  While Republicans will wax on about the possibility of taking Washington or Colorado or Connecticut, that's not happening this year, mostly because special elections since the Dobbs decision have basically made it impossible for Republicans to pick up strongly blue territory.  New Hampshire is about to join this list as R's seem like they'll waste an opportunity by nominating an unelectable candidate in Don Bolduc against incumbent-Sen. Maggie Hassan.  

Thus, the Republican opportunities are pretty bare (just Nevada, Georgia, & the waning opportunities in Arizona), but Democrats seem to have a lot more potential if they can make it happen.  Pennsylvania is favored to flip to the Democrats, but it's becoming increasingly plausible that Wisconsin should be in the same tossup category as Nevada & Georgia.  Lt. Gov. Mandela Barnes has led in every election poll this contest against Sen. Ron Johnson, a uniquely polarizing figure in Republican politics, particularly given his recent attacks on social security (the third rail of American politics).  The only reason that Barnes is not considered a tossup in the same way is that virtually everyone has been burned by Wisconsin poll numbers (particularly Hillary Clinton, Russ Feingold, & Joe Biden, the latter of whom won but by nowhere near the margins that he was promised heading into Election Night), and Johnson is like Frankenstein's monster, always coming back before the end credits roll.  Barnes, though, cannot be discounted, and the fact that Republicans are shoveling their limited cash supply on races in Ohio & North Carolina indicate that they are nervous about Democrats taking one of those seats, which would less complicate their chances in 2022 (if they lose Ohio or North Carolina, they have no shot at a majority, period), and more because it would make the majority in 2024 & 2026 considerably harder to maintain.  Republicans losing Nevada, Georgia, or Wisconsin would be bad for them...losing North Carolina or Ohio would be a catastrophe of cataclysmic proportions for their chances the remainder of the decade.

That's where the race for Congress sits post-Labor Day...basically, in chaos.  The Senate is mildly favored to go to the Democrats (but nowhere near a margin that they should breath easily with, at least not until Cortez Masto or Warnock look safe), and the House is still probably going to the Republicans (but that is based less on polling & special elections, and more on history, always a dangerous way to make decision in uncertain times when there's mounting evidence Democrats are gaining the upper-hand).  In the next sixty days or so, we're going to see how this holds up-it's possible that Republicans will revert to form & take both houses as the fundamentals say they should, but they are in uncharted territory, and perhaps it's time for us to start looking at 2002, not 2014 or 2006, for guidance on what will happen in November.