In "What's in a Wave," I set out an admittedly arbitrary marker for what would count as a "blue wave" in the midterm elections: Democrats gain at least 40 seats in the House of Representatives (17 more than needed for a majority and about 10 more than the out party usually picks up in an off year election) while gaining at least two seats in the Senate, which would give them a majority in that body, too. The Senate map is so stacked in favor of Republicans that the latter is pretty plainly the higher hurdle, and it should probably count as a wave if, for example, the Democrats gain 50 House seats without losing any ground in the Senate. But let's just stick with the plus-40, plus-2 formula. The prospect of gaining 40 House seats is well within the Democrats' reach. What about netting two seats and a majority in the Senate?
When people talk about a favorable map for Republicans, they only mean that every two years about a third of the Senate seats are up for election, and this year's roll of 35 happens to give a considerable advantage to the party of Trump. For one thing, Democrats currently hold 26 out of the 35 seats, so they start out with almost three times the exposure as Republicans. Of course they aren't in any real danger of losing many of those 26 races: Elizabeth Warren will be reelected in Massachusetts, as will Bernie Sanders in Vermont, Angus King in Maine, Sheldon Whitehouse in Rhode Island, Tim Kaine in Virginia, Amy Klobuchar in Minnesota, Dianne Feinstein in California--enough others to make compiling a list sort of tedious. The problem arises from the fact that, of those 26 seats Democrats are defending, 10 are in states carried by Trump. Let's list them and at the same time place them in two categories: five that Trump won by a single-digit margin (Pennsylvania, Ohio, Michigan, Wisconsin, and Florida) and five others he won by a double-digit margin (West Virginia, Indiana, Missouri, North Dakota, and Montana).
It's easy to see why, in the immediate aftermath of Trump's election, Senate prognosticators thought the Democrats in 2018 would be looking up the south end of a north-bound horse. Back then, Republicans held 52 Senate seats, and there was a very real prospect they could this year attain a filibuster-proof majority of 60--just hold the seats they already had, sweep the five states Trump won in landslides, and win three of the five in the other Trump states. But since then their position has deteriorated substantially. They went from 52 to 51 seats when Democrat Doug Jones won a special election in Alabama [sic], so it's already too late for them to hold the seats they already had. In the five single-digit Trump states, four Democratic incumbents appear to be safely ahead--only Florida looks close. Meanwhile, in the five Trump landslide states, two Democratic incumbents--Jon Tester in Montana and Joe Manchin in West Virginia--are looking like strong favorites to win reelection. It's not too much to say that, while Americans have taken the measure of Trump in office, the number of Republican-held seats in the Senate has declined from 52 to 51, and the number of ripe prospects for Republican Senate pickups in 2018 has declined from "about 10" to "maybe 4." Consider that if you're asked the right-track, wrong-track question!
Still, 51 is more than 49, and 4 is greater than zero. Even if Democrats suffered nothing even resembling an upset, and went 4-for-4 in Florida, Indiana, Missouri, and North Dakota, they'd still need to win two seats currently held by Republicans. Recall that only 9 Republican seats are on anyone's ballot. Alas, these 9 include Nebraska, Utah, Wyoming, and, on account of a retirement, both seats in Mississippi. Forget about those five. Let's list the other four in order of ripeness with an appropriate handicapping descriptor:
Arizona (Democrat favored)
Nevada (Democrat more likely than not to win)
Tennessee (knife's edge)
Texas (this would be a Democratic upset)
Maybe the first thing to be said about this list is geographical. It's true that Democratic fortunes have ebbed in the older, whiter Midwest. But they are rising in the sunny and growing Southwest. New Mexico has completed the transition from red to blue, Arizona and Nevada are en route, and Texas is next in line. In Tennessee, a crimson state, the Democrats have a very strong candidate, Phil Bredeson, who, when running for reelection for governor in 2006, won two-thirds of the statewide vote and all 95 counties.
The sum of it all is that, right now, it looks like a Senate majority hinges on the outcomes across eight states--four seats that the Democrats are trying to defend (in Indiana, North Dakota, Missouri, and Florida) and four where Democrats are on offense (in Nevada, Arizona, Tennessee, and Texas). For a majority, Democrats need to win most of them--probably six of the eight. That sounds like a tall order, and it is, but here's something: the likelihood of winning six isn't much less than the likelihood of winning four. Today, it looks like Republicans have at least a two-thirds chance of retaining their majority. But suppose we could know today that, say, Democrat Claire McCaskill was going to win reelection in Missouri. The Democrats' chances of flipping the Senate would then jump from less than a third to better than fifty-fifty--on the principle that if the tide rolls in, the tide rolls in. The converse would be that if we knew McCaskill was going down, we could be reasonably sure that Republicans would win most of these tight races and add to their Senate majority--because if the crest of the wave doesn't bring McCaskill to shore, it likely will fail also in Texas, Indiana, North Dakota, and Tennessee.
If you're similarly obsessed with this kind of stuff, click on the youtube video above to get the assessment of the Cook Political Report.