Three weeks out is a good time to take stock of how things are looking in this year’s election.  Voter registration has closed in about half the states and will close in a good chunk of the remaining states in the next seven to ten days.  (Some of these states do permit “same day voting” where you can register in person when you vote, but most do not.)  Meanwhile, early voting has started in about one-third of the states.   (One-third of the states do not technically have “no excuse” early voting.)

The 2016 elections should serve as a reminder of the limitations of polling.   By definition, undecided voters have not decided and they can break in any proportion.  While national polling tends to be more accurate than district-level polling (and more common), elections are decided on a district-level basis.  And, all polling uses turnout assumptions that may or may not reflect who actually votes in an election.  The one advantage that Democrats have over the poll numbers is that groups that tend to vote Democratic tend to have low turnout, especially in the mid-terms.  Thus, if we can get a high turnout from these segments of the population, we should exceed the current poll numbers.

With that caveat, this year’s election appears to be a creature of the calendar.  Because 2014 and 2016 were good years for the Republicans, it looks like Democrats are poised to make significant gains in the U.S. House and in state legislative races.   It is clear that Trumpism is redefining the Republican party putting many suburban voters who used to lean Republican up for grabs.  That has created a decent number of “Romney-Clinton” districts and even more districts that Romney won solidly that Trump barely won.   A classic example of the districts that will decide who controls the House is one of the districts from the area where I grew up, the 7th district of Texas.  For the past fifty years. dating back to the first President Bush, this district has been a Republican district.  Forty years ago, the minority population in this district was non-existent.  It is now minority-majority, but just barely.  It also includes several very highly educated parts of Houston including the area around Rice University.    It also includes some of the wealthiest neighborhoods in Houston.  Will the Democrats win this district this year?  Maybe.  But the fact that Republicans are having to defend this district is a problem for the Republican Party in the near future.

There are other districts in areas that barely went to Trump in 2016.  As Democrats were taught repeatedly during the Obama era, your voters expect that winning an election will bring results.  They tend to stay home when winning doesn’t bring the type of results that they expected.  On the other hand, the party out of power is likely to be upset at what is getting done.    That change in the expected voters can make a difference in who wins swing districts.

On the other hand, as we have known since the day after the election,  the calendar works against us in the Senate.  With the exception of the two special elections, the Senators up for re-election this year are the one elected in 2012 when Obama was beating Romney.  That means a lot of Democratic Senators and very few Republican Senators (the next two maps of Senators elected in 2014 and 2016 should favor Democrats) and the expected lower turnout in mid-term elections will favor Republicans.  Control of the Senate is going to come down to Florida, Indiana, Missouri, Montana, and North Dakota (currently held by Democratic Senators) and Arizona, Nevada, Tennessee, and Texas (currently held by Republicans).  With North Dakota looking like a long-shot (even more so with new rules making it harder for Native Americans to vote), Democrats will need to hold all of the rest of their seats and pick-up at least three of the four vulnerable Republican seats.  There is an outside chance that control of the Senate could come down to the Mississippi special election.  That election is a free-for-all general election with a likely run-off (what is typically called a “jungle primary”).  It is likely that Mississippi will go to a run-off and Democrats chances will depend on which candidates make the run-off.  (There are three major candidates, two Republicans and one Democrat.  There is not a lot of polling, and it looks like a close three-way race.  Given the typical breakdown in Mississippi, the Democrat should ultimately make the top two, but that is not a sure-thing.  Assuming that the Democrat does make it through, which Republican makes it through will determine whether the Democrat can win.)

The next twenty-one days are about working hard locally.  The good news is that we are doing well in enough areas that the Republicans are having to decide which Republican members of Congress will have to fend for themselves because there is not enough money to go around.  The Republicans are playing defense this cycle.  For all the brave words from the liar-in-chief about a Republican wave, the Republicans are really hoping to keep their losses down.  The key number is twenty-three seats.  At this point, a win for the Republican is a very narrow House majority.  (In the perverse reality of politics, a 220-215 Republican majority may actually be good for Democrats running for President in 2020.  The Republicans would almost certainly have problems getting anything done, but they would still be nominally in charge of both house of Congress and the White House.)  On the other hand, we could be looking at Democrats gaining a record number of seats.  The next twenty-one days will tell the answer.