(CNN) Results from the Washington state primary on Tuesday are another bad sign for Republicans.
We already know that Democrats lead in the House polls, and Republicans have a lot more seats at risk. Republicans therefore are hoping that the polling and the race ratings are overstating their vulnerability. They may be.
The outcome in Washington's primary however, suggest the opposite may be true. The polling and race ratings might be understating the Democrats' advantage. In other words, Democrats may actually be in slightly better shape than we think.
Washington, like California, uses a top-two primary system. That means all the Democrats and Republicans run in the same primary (with the top two vote-getters, regardless of party, advancing to the November general election).
Historically, the primary vote all the Democrats receive relative to all the Republicans is a good indicator of the fall result. As the New York Times' Nate Cohn pointed out in June, the top-two primary results are about as predictive as a poll taken within the final three weeks of the campaign. The catch, though, is that Democrats tend to do slightly better in the general election than in the primary.
With that in mind, look at the results in these three districts: Washington's 3rd, Washington's 5th and Washington's 8th. Not all the votes are in, but if the picture doesn't shift too much, then Republicans could be in big trouble.
Washington 3rd: President Donald Trump won here by 7 points. CNN, the Cook Political Report and Inside Elections currently rate the 3rd, represented by Jaime Herrera Beutler, as either likely or solid Republican. That is, the belief is that she has a very good chance of winning reelection.
The cumulative Democratic percentage of the vote so far is actually outpacing the Republican percentage of the vote by a little less than a point. That suggests this race may actually be a toss-up in the fall, not one where Republicans should be heavily favored.
Washington 5th: Trump took this district by 13 points in 2016. CNN, the Cook Political Report and Inside Elections currently put the 5th, represented by Cathy McMorris Rodgers, as either lean or likely Republican. Put another way, these ratings indicate that she's thought to be a favorite, though not as heavy as one as Beutler.
The results tonight suggest that this race, like the 3rd, may also be a tossup. Republicans are running about three points ahead of the lone Democrat. If you apply the average shift from the primary to fall results (with Democrats doing better in the fall), this race is forecasted to be razor tight.
Washington 8th: Unlike the 3rd and the 5th, Trump lost here by a three-point margin. That's part of the reason why CNN, the Cook Political Report and Inside Elections all believe the 8th District, where Rep. Dave Reichert is retiring, is a toss-up. This is the equivalent of thinking that a race is 50/50 or close to it.
The Democrats, though, are up by 3 percentage points in the primary. Given how Democrats usually improve upon the primary result in the fall, the forecast would be for them to win by around 5 points in November. That's more like a race that is closer to leans Democratic than it is a toss-up.
The results in Washington wouldn't be the first sign that the national polls may be somewhat underestimating Democrats. Democrats have been outperforming the partisan baseline based on the last two presidential elections by greater than 15 points in the average special election. A forecast based solely on the specials has them winning the national House vote by low double digits. Most of the national polling has been suggesting a margin closer to high single digits.
But these Washington results aren't a guarantee of anything nationally. Polling in special elections has been fairly accurate (though it too has underestimated Democrats a little bit). Democrats may still outperform expectations in the Evergreen State, while not doing so in the rest of the nation. The California top-two primary results were good for Democrats, though not anywhere near as good as Washington's for example.
It's also possible that Washington Democrats will underperform what the primary suggests will occur in the fall. All these November predictions from the primary come with a margin of error. The average error is in the neighborhood of 6 percentage points. It's large enough for Republicans to win in all three districts.
And even if Democrats do better nationally than believed, they won't do better in every district. Republicans would probably win some toss-up seats.
On the other hand, there may be a district or two (like Washington's 3rd) that aren't on many people's radar right now that will end up being on Election Night. Most wave elections (if this turns out to be a wave) end up with a few incumbents going down to defeat who most never thought would.