A surfer rides a wave in Nazare, central Portugal, on Dec. 15, 2017. (Francisco Leong/AFP/Getty Images)

There are a lot of positive signs for the Democratic Party’s chances in the 2018 midterms. On the generic congressional ballot, asking Americans if they prefer the Democrats or the Republicans to win the House, the Democrats have a 6-point lead in the RealClearPolitics average. President Trump continues to be unpopular, and a president’s unpopularity tends to mean his party does worse in congressional elections. Then there are all of those off-year state and congressional elections, in which the Democrats have picked up a slew of seats and, more broadly, seen huge gains over their performance in 2016.

The most recent example of that last effect is in Missouri, where, on Tuesday, there were four special elections to fill vacant seats in the state legislature. All four were held by Republicans; Trump won all four by an average of 49 points in 2016, according to analysis from DailyKos.

On Tuesday, though, the Democrats did an average of 32 points better than they had in the districts in 2016. In one district, the 97th, Democrats actually picked up the seat. In another, the 144th, the Democrats gained 54 points over how they had performed in the 2016 presidential race.

(In addition to Tuesday’s results and those from 2016, we’ve included the most recent contested election in each district.)

This is the pattern we were used to seeing in 2017 and 2018: Big shifts to the left.

Before the Democrats start assuming a Democratic tsunami is bearing down on Washington, though, it is important to add context to the good news the party is seeing. That congressional ballot lead has narrowed considerably over the past 90 days or so. Trump is still unpopular, but his approval ratings are ticking upward. Those victories in special elections? They may not tell us as much as it at first seems.

Cook Political Report’s Dave Wasserman had an elegant solution for why the big swings that have been seen in legislative races did not result in massive swings in congressional races that were enough to carry the day for the Democrats.

We can see that effect in Missouri. Instead of looking at percentages, let’s look at actual vote counts.

Suddenly, the picture is very different.

Each of the candidates running on Tuesday saw a big drop-off in the number of votes they received relative to how the presidential candidates did in those districts in 2016. The closest candidate was Democrat Jim Scaggs in the 144th District; he got 95 percent of the number of votes Hillary Clinton got two years ago. The other three Democrats got an average of 31 percent of the number of votes Clinton received in 2016. The Republicans in those races got 16 percent of the votes that Trump did. In two of those three races, that drop-off did not change the result. In one, the 97th, it did.

These are heavily Republican districts which, in elections with significant turnout, have consistently backed Republican candidates. In very-low turnout elections, the more-energized Democratic base could take a bigger bite out of the Republican lead. Take that 39th District. Add 1,000 Democratic votes to the 2016 election, and the results go from an 71-24 Trump win to a 67-29 Trump victory. Add 1,000 votes to Tuesday’s results, and the outcome shifts from a 64-36 Republican win to a 52-to-48 Democratic one. Same number of votes; much bigger swing on Tuesday.

November 2018 will not see the turnout of 2016, since turnout in midterm elections is consistently lower than in presidential years. It will be higher than in the special and off-year elections we have seen since Trump was inaugurated. That will likely muffle the effects of Democratic enthusiasm.

The Democrats are still well-positioned. But 54-point swings are not going to happen with regularity in House races this November. The Democratic wave washing over state houses now will have a tougher time overcoming higher-turnout elections.