top of page

Post-Midterm Postmortem: That Was the Election That Was

Now that the midterms are ten days in the rear-view mirror, even as a handful of House races are still undecided, it’s clear that the overall results were surprisingly (though not astonishingly) good for the democrats. And so, everyone is asking what happened, i.e., why the red wave (which, to be fair, was not the most likely outcome) never occurred? And lots of people are answering more or less the same way, generally citing four inter-related causes:

Cause #1: Trump. Need we say any more? While the republican establishment would have preferred to define the election as a referendum on the fairly unpopular Biden administration, Trump instead made it about the re-litigation of the 2020 presidential election and/or the opening salvo of the 2024 election. The American people had rejected Trump in the two previous electoral cycles – three, if you count that he lost the popular vote in 2016 – and they were happy to reject him yet again.

Cause #2. The Dobbs decision. Basically, if the five (or six) horse-persons of the SCOTUS apocalypse had feverish visions of remaking the country into a neo-fascist theocracy, they chose the wrong time to start dismantling American freedoms. Almost immediately after the overturning of Roe, voters (especially younger ones) recognized the concrete costs of republican overreach and penchant for minority rule. And they’re still fired up.

Cause #3. Weak republican candidates. If the senate minority leader starts making bitchy asides about his own party’s “candidate quality,” then you know the republicans weren’t using their sharpest pencils to write this cycle’s script. And it’s not just in congress. Delusional election-deniers did pretty poorly in state-level races in most of the purple battlegrounds. It used to be the democrats who habitually shot themselves in the foot.

Cause #4. A repellent opposition party. The American public may have soured on Biden, but the republicans did not provide a particularly attractive alternative, in no small part because they don’t seem to stand for anything except American exceptionalism and legislative obstruction. You never quite know these days whether they’re championing isolationism or muscular militarism, fiscal conservatism or runaway budget deficits. In the meantime, the dominant party voices are those of xenophobes, conspiracy theorists, and Trump sycophants. What human being with an ounce of brains can vote for them?

Of course, one could argue that the last three of these talking points are simply corollaries to or variations of the first point, i.e., the Trump factor. It was, after all, Trump who stacked the Supreme Court with reactionary hacks, who championed incompetent loyalists, and who tapped into the republican rank and file’s lowest instincts. In any event, these are all perfectly reasonable explanations, but as you might expect, the Toteboard has a couple more hypotheses to stir into the stew. But remember, this is not for the purpose of rewriting the emerging narratives. Rather, it is to introduce some very simple subtexts that don’t yet seem to have caught anyone’s ear.

And so here we go, one chamber at a time:


It’s important to note that the final senate configuration pending the Georgia runoff (which will be the subject of a future post) stacks up quite will with what the Toteboard had been anticipating (and repeating) since early 2001, i.e., “a midterm election cycle that may not have a whole lot of movement at all, at least in the Senate.” That’s not exactly unprecedented, but it is unusual.

So what independent variable was the best predictor of that result? Again, to quote the Toteboard from twenty-one months ago: “For all the talk of historical patterns, favorable maps, and upside-down mandates, all of this may end up being trumped by the simple fact that because of the country’s hyper-partisan mood . . . it might just be that voters end up choosing senators in accord with their state’s respective colors.”

This isn’t rocket science. It doesn’t employ any fancy algorithms or probabilistic models. The simple truth is that in almost every state that had a senate election, the state voted the same color they did in the 2020 presidential election. The one exception was Wisconsin, the only blue 2020 state that had an incumbent republican on the ballot in 2022, and both races were decided by less than a single percentage point. In fact, when other states were close in 2020, most of them ended up pretty close in 2022 as well. There’s no better symbol of how this pattern is holding than the results in the state that had the narrowest percentage margin of victory in 2020, i.e., the Toteboard’s home state of Georgia. It seems only fitting that the state that Biden won by 0.23%, without reaching 50% of the total vote, should have an almost identical result in 2022, with the democrat leading by 0.93% without hitting 50%, but now heading to a runoff because of the state’s regressive election laws.

If we were to have made predictions based on the “state-color model,” we would have done at least as well as or better than the main big-money analysts like Larry Sabato (who missed Pennsylvania), FiveThirtyEight (who had Pennsylvania and Nevada both leaning red), and Race to the WH (who had Nevada leaning red). And of course, one or more of them are going to get Georgia wrong too (Sabato and FiveThirtyEight guessed red, Race to the WH guessed blue). Still, it’s interesting to note that while we’re seeing less ticket-splitting between presidential and senate races, that doesn’t quite carry over at the state level, as red Kansas elected a democratic governor, while blue New Hampshire and Nevada elected republicans.


The house of representatives is a little bit more complicated, as there were some unpredictable results, with a few legitimate upsets in both directions. As it stands right now, the democrats are on track to lose anywhere from six to ten seats overall, probably on the higher side of that range, though still considerably less than most analysts were predicting. The republicans will hold a narrow majority; it remains to be seen how narrow.

So how does the party that has a president whose approval rating is hovering around 42%, that is suffering through record-high inflation, and that has engaged for two years in more infighting than messaging – how does that party manage to survive a restless electorate with only minor damage?

The Toteboard’s take on this begins by looking not simply at the 2022 congressional races in isolation, but expanding the conversation to consider the 2020 presidential election and the concomitant down-ballot races. Historically, when the presidency changes partisan hands, especially when the election is decided by a substantial margin (which hasn’t happened much in the last twenty-odd years), the victor’s party normally over-performs in the House elections. Some of this is what pundits call “coattails,” but a lot of it is also just the result of a general (and likely temporary) shift in the public’s partisan mood. And so, when Obama sailed into office in 2008 (succeeding Bush 43), the democrats gained 24 House seats, as well as 8 senate seats. When Reagan ousted Jimmy Carter in 1980, the republicans gained 35 House seats, as well as an astounding 12 senate seats.

Now when the incumbent’s party loses seats in the subsequent midterms, political analysts normally attribute that to a repudiation of the sitting president, and sometimes that is at least partially accurate. But a large piece of that dynamic is that it provides a compensation for the party’s over-performance two years earlier, that it is basically bringing back down to earth numbers that had been inflated by the previous presidential election. Reagan and Obama may have gained three-dozen and two-dozen House seats during their respective first-term elections, but those numbers were simply not sustainable. They were temporary partisan blips, not durable realignments.

But Biden’s party-shifting election two years ago deviated from the pattern. Despite winning the popular vote by a fairly impressive 4.5% – bettering Trump’s 2016 margin, Obama’s 2012 margin, and Bush’s margins in both 2000 and 2004 – Biden’s party actually lost 14 House seats that year. To get a sense of how truly unprecedented that was, how truly bizarre that was, you have to go all the way back to John Kennedy’s election of 1960 to find a presidential cycle where the victor’s party lost more seats than that (and that was itself a compensation for Eisenhower’s glandular-case 50-seat loss in 1958). The Toteboard had initially characterized this as a “backasswards mandate.” Biden won big, and he lost big.

Biden’s party didn’t just fail to over-perform in 2020, as history says it should have, it seriously under-performed. What this means is that for whatever reasons – e.g., misgivings about Biden, uneasiness about democrats controlling both branches of congress, etc. – the voters already checked Biden’s party back when they voted for him in 2020. That is to say, the compensation that normally occurs with the midterms was already baked into the 2020 House results – and so there was simply less space to do so again. Thinking of this numerically, the starting point for this election cycle – democrats controlling the House by about ten seats – was actually kind of a “false baseline.” The democrats “should” have started with a much stronger majority, but didn’t, and so the apparent baseline actually inflated existing democratic electoral weaknesses and exaggerated the head-start that precedent says republicans should have had going in. So while history, and Biden’s sagging popularity, and inflation and all that stuff, may have suggested a theoretical across-the-board democratic vulnerability in 2022, there just weren’t that many democratic seats that had over-performed two years earlier. To put it in a somewhat playful way, democrats didn’t lose all that many House seats last week, because they already lost many of the seats they would have lost in 2022 . . . back in 2020! Maybe it's just a coincidence, but if you add the 14 seats the democrats lost in 2020 to whatever their final 2022 seat loss turns out to be, it will come real close to those two-dozen seats Larry Sabato and others were anticipating this time around.

Now we know that if you are expecting a shellacking, well, you can only put so much shellac on a table that already has multiple layers of varnish on it.

Whether or not you buy these hypotheses, here we are again (or still), with a bitterly divided government reflecting a bitterly divided country. And in this political climate, every vote, every seat, really matters. That’s why for the next three weeks, Georgia will be on everybody’s mind.


bottom of page