CNN  — 

House Speaker Paul Ryan’s announcement Wednesday that he will not seek re-election is the latest sign that the Republican House majority is in trouble.

The writing has been on the wall for a while now. President Donald Trump’s low approval rating, Republicans’ poor standing on the generic congressional ballot and Democratic performance in special elections since Trump took office all point to a bad outcome for Republicans in November.

And now Ryan, the leader of the GOP House majority, has announced his exit, dealing a major symbolic blow to the party as it heads into a tough campaign season.

Just over the last week, the prominent Cook Political Report has shifted their projections of 15 House seats in favor of the Democrats. After Ryan’s announcement, there are now 80 Republican-held seats that are or have the potential to be competitive compared to just 16 seats for Democrats in Cook’s ratings.

When looking at Cook ratings specifically at this point in the cycle, there’s been a fairly clear pattern since 2006:

  • Solid seats are won more than 99% of the time by the party favored in the solid seat at this point.
  • Likely seats are won about 85% of the time by the party favored at this point.
  • Leans are won about 70% of the time by the party favored at this point.
  • Tossups are won only 45% of the time by the party who currently holds the seat at this point. (That is, they aren’t true tossups. They’re actually slightly more likely to go the other party!)

Given the number of seats in each of these categories right now, the current Cook projections suggest about 220 seats go Democratic to 215 going Republican after the election. That suggests close to a true tossup for control of the House.

In other words, Democrats pick up a net of 24 seats. Democrats only need to net 23 seats to win the House majority.

This distribution, however, misses a rather key point. The side that eventually picks up seats tends to do a lot better than the seat ratings suggest at this time. And usually, it is pretty clear which side is going to pick up seats. This year it’s almost certainly the Democrats.

When a party is going to pick up seats, they usually not only to do a very good job at holding the seats that are rated to go to them but also do a good job at picking up seats that the other side is actually favored to win. Put another way, seat-by-seat estimates at this point tend to underestimate the extent of the wave.

Here’s how the party that picks up seats on average does compared to the ratings at this point:

  • They win 99.8% of the seats they are solid in and even 1.3% of the seats that are solid for the other party.
  • They win 99.2% of the seats they are likely to win and 24% of the seats the other party is likely to win.
  • They win 90% of the seats that lean in their direction and 42% of the seat that lean in the other party’s direction.
  • Finally, the party that ends up gaining seats in the midterm win on average 68% of the seats rated as tossups at this point.

If apply this distribution to the current state of play, you’d end up with Democrats winning 233 seats in the House to just 202 for the Republicans. That’s a net gain of 38 seats for the Democrats.

But Democrats shouldn’t be popping the champagne just yet, even with Ryan exiting the House. Although most years follow a pattern similar to the average, there’s enough variability in how race ratings at this point translate to eventual seat gains over the last five cycles to be exactly sure how many seats Democrats will gain.

Republicans actually won most of the toss-up races in 2016, even though they lost seats. If the distribution follows the 2016 pattern, Democrats would only win around 208 seats or a net gain of 13 seats.

Of course, it’s plausible that Democrats win even more seats than 233. If the different categories break as they did in 2006, Democrats would win about 248 seats or a net gain of 53 seats.

Neither the low-end or the high-end seat gains look like the most likely outcomes at this time, but they give you an idea that there are a lot of different potential outcomes given we’re still months before the election.

Indeed, there’s even more reason to be cautious: different race ratings suggest potentially different stories about the state of play for the House. CNN produces its own ratings, which are slightly less bullish on the Democrats. There are also the folks at Inside Elections who see things somewhat differently.

Inside Elections has over 20 more seats that are solid (or safely) Republican than Cook and over a dozen more than CNN. Part of this difference may have to do with the fact that Cook and CNN view the likely category as uncompetitive at this time (though could become competitive), while Inside Elections views the likely category as competitive. Inside Elections also has about 15 seats more overall than either CNN or Cook in which Republicans are at least nominally favored, though much of that has to do with the fact that Inside Elections has an extra category between lean and toss-up known as “tilt.”

Applying how races have broken historically to the Inside Elections’ ratings would suggest that Democrats are only slightly more likely than Republicans to gain control. Of course, that’s only if races break towards the party that gains seats in the fashion they normally do. If Republicans hold the seats they’re favored in at this time, then they probably would hold on to control of the House.

Still, the fact that the Democrats are still in a decent position to take back the House even in the more pessimistic race ratings gives you an idea that things are probably leaning their way and gave Ryan good reason to leave.