Wavy Gravy

I spent some time looking at congressional districts and their Cook Partisan Voting Index.  The CPVI is basically an attempt to measure whether a district or state leans one way or another and by how much.  It's not flawless, but it's a metric people use.  I then looked at ten states to see how many Republicans represented districts with a CPVI or R+10 or better.  In other words, which districts tilted to the GOP by ten points or less.

The reason I did so is because midterms elections are usually referenda on the sitting president and Trump's numbers are atrocious.  This sets up the conditions for a "wave election."  Roughly speaking a "wave" overrides both the CPVI and the advantages of incumbency.

Here are the 12 states: CA, VA, PA, MI, OH, WI, CO, FL, NY and NJ.

Here is what I found if we just take seats in D+ or even CPVI:

California flips four seats to Democrats, New York flips three, Colorado, Virginia and New Jersey all flip one for a total of ten seats.  This is significant, because CA, NY and NJ are strong lean Democratic states and Virginia and Colorado are toss-ups.  The overall culture of the state can matter a great deal.  I would expect most if not all of those ten seats to flip in 2018, unless Trump's fortunes reverse by a huge amount.  I mean a YUUUGE amount.

What about the "lean Republican" districts with a CPVI of R+5 or lower?  That adds 3 more to CA, 5 more to NY and 2 more to VA and 1 more each to NJ and CO.  However, it also adds the following states: OH flips 2 seats, but FL and WI flip 4 each, PA flips 5 and MI flips 7.  That's a total of 32 more seats.

So a wave that wipes away the 5 point partisan/incumbent advantage puts 42 House seats in play.  Go up to a 7 point wave and that adds 2 more from PA and MI, 3 more from NJ and 5 each from FL and OH.  That's 17 more seats.

Democrats need to flip 24 seats in 2018 to hand the gavel back to Nancy Pelosi.  If they flip those first 10 seats, half the "lean Republican" districts and a third of the "strong lean" Republican districts, that's 31 seats.

This is very inexact, as it fails to account for the shifting demographics of WWC and suburban, college educated whites, who are moving in opposite partisan directions under Trump.  But it's better than saying "Democrats have to win the 'National Popular Vote for the House' by 7%."  Because what Democrats have to do is create a wave overrides the partisan and natural gerrymandering present in today's House.  That is best measured district by district.

But one thing it definitely should suggest is that Democrats have a real shot - at this point in time - of taking back the House.  This is especially true because they have some very winnable seats in CA and NY, where the cultural tide is working against the GOP.  There are additional <R+5 seats in AZ (2), IL(5), IA (2), ME, MN (2), NE, NV, NM, TX, WA (2).  That adds 18 more seats in play, some of which are in fertile ground, like Illinois, Washington and Minnesota.

And this is where getting WWC voters to turn on Trump could make for an historic wave election.  Because right now, they are the only group still supporting him.  Secondly, Democrats need to start recruiting and training candidates NOW.