Many Democrats want to believe....but they just can't bring themselves to do so. I mean (gulp) what about 2016? Don't wanna get fooled again, etc, etc.
In that regard, there are a number of fears that are typically expressed about why things are not really as good as they look. Here are some of the most common, with recent evidence to the contrary.
1. So the Democrats have over-performed in special elections. Bah, humbug, that doesn't predict anything. Ah but it does, it does. The latest evidence is a nice analysis by the good folks over at Daily Kos Elections (not following them?--you should!). In brief, they created an index of special election results in each interim period between federal elections and compared them to the House popular vote in the next federal election. They find a very strong predictive relationship.
And that's not all. They also looked at their index for just the year immediately following a federal election to get a sense of how predictive recent special election results, which just cover one year (2017), might be. They also find a strong relationship (though not quite as strong as with the two-year index). How strong? The one-year index does about as well in predicting the House popular vote as the generic Congressional ballot just one month before the actual election. That's pretty good and tells us that the 2017 special election results really are sending us a strong (and very positive) signal about Democratic prospects in the 2018 House election.
2. But won't the Republicans' traditional midterm turnout advantage neutralize all the Democrats' advantages? Nope, not likely. In fact, it's not clear that Republicans are likely to have much of a turnout advantage in this election anyway. As Harry Enten notes on 538, in elections where they occupy the Presidency, so their candidates are not running as the opposition, the Republican turnout advantage has been minuscule; they turn out a a level barely higher than Democrats and do not succeed in significantly improving their margin over that among registered voters as a whole. That means there's not really much of GOP thumb on the scales in terms of turnout in 2018 since the President is a Republican (and a very unpopular one at that).
3. Ah, but what about the economy? It's doing well and that will make Trump and therefore the GOP more popular, undercutting the Democrats. Well sure, if Trump does become a lot more popular because of the economy, that would really help the GOP, given the well-established relationship between a President's approval rating and the midterm electoral performance of the President's party. But that just doesn't seem to be happening. As Nate Cohn shows in a recent analysis, the extraordinary thing about Trump is how massively he's under-performing the state of the economy in terms of job approval. Going by economic performance alone, his job approval by this point in his term should be somewhere in the 50's not in the high 30's where it continues to languish. Cohn notes:
It remains possible that Democrats could snatch defeat from the jaws of victory. Or that perhaps a really huge economic boom could finally manage to elevate Trump's approval ratings and therefore his party's prospects in 2018. But right now a Democratic wave still looks like a pretty good bet.Since 1950, no party has held the House through a midterm election when the president’s approval rating is less than 40 percent. The Republican Party’s considerable in the House would at least give them a shot to survive this time, but the growing Democratic and the G.O.P.’s weak showings in this year’s special congressional elections suggest that the president’s approval rating is weighing on the party in exactly the way one would expect.