Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
Don’t Fool Yourself — The Blue Wave Is Probably Real
On August 13, Ricochet member @bloodthirstyneocon shared a post about entitled “Nope, the Blue Wave Is a Myth.” The one datum pointed to in that post was the Real Clear Politics average of generic ballot polls which showed the Democratic advantage at a mere 3.9 points. The prevailing wisdom (as cited in the post) is that Democrats need a seven point advantage to retake the house, due to factors including gerrymandering and “wasted votes” in Democratically dominated urban areas.
August 13 was about the only time anyone could make that point, because there were two times this year that the Democratic advantage dipped that low, and one of them was on August 13. It immediately rebounded. So if you were pinning your hopes on something other than a blue wave to that one datum, you should know that the current RCP average is D +8.9.
However, the RCP average is not the only thing pointing to a blue wave.
The website FiveThirtyEight recently released their congressional forecast model which looks at all 435 congressional district. It’s … very complex. (If you want to hear Nate Silver explain how it works, you can find that here, and they made some changes which they talk about here.) They actually have several models that look at a whole slew of factors, including polls, weighted based on history. But since there aren’t quality unbiased polls for most districts, they do rely on other data.
So what does their model show?
Currently they give Republicans a 25.7% chance of keeping control of the house, or, as they phrase it, one in four. They give Democrats a three in four chance of keeping control.
If you’re curious about their forecast for your particular district, you can find that here. Since they launched their model a couple of weeks ago, those numbers have remained stead, as has their estimate of the popular vote margin, which they estimate at D +7.8.
Traditionally, the party that controls the White House loses seats in the midterms. It’s happened in almost every midterm year going back at least half a century. It happened in 2010. It happened in 2006. It happened in 1994. (The exception was 2002, which took place a 14 months after 9/11.)
Let’s look at 2010. That year, if we look at President Obama’s job approval rating for this week (8/30-9/5), it was 45 percent approve, 47 percent disapprove. On Election Day, it was the same, 45-47. Democrats lost 63 House seats that year.
For comparison purpose, Gallup currently has President Trump at 41 approve, 53 disapprove. That is in line with FiveThirtyEight that has Trump’s numbers to be 40.3-54.4 and RCP which shows at 41.8 to 54.0. In other words, Donald Trump is dramatically more unpopular than Barack Obama was at this point in their respective midterm cycles.
Is this set in stone? Could things change? Sure. A lot can happen in two months, and FiveThirtyEight still gives the GOP a one in four chance of holding onto the house. Weird stuff can and does happen in politics.
But if you think the Blue Wave is a myth, with all due respect, you’re deluding yourself. This isn’t a media creation. This isn’t a deep state conspiracy. This isn’t Democratic propaganda. There is data that strongly points to a victory for Democrats on November 6.
Addendum
Here are a few responses to common objections:
Yeah, yeah, but all the polls said Trump wouldn’t win either.
I know that’s what some Trump supporters claim, but it doesn’t apply to everyone. FiveThirtyEight still has their 2016 election page up. You can look at it for yourself. They gave Trump a 28.6 percent chance of winning. Not zero percent, not 10 percent, 28.6. What they’ll tell you is that they took crap from people before election day for having it that high, but it was what their model predicted.
All these polls are biased against Republicans.
Not all pollsters are created equal. There are polling outfits that lean toward Democrats and there are some that lean towards Republican. But the incentive structure in polling favors accuracy. If you’re interested in the quality of pollsters, their predictive values, and how they lean, FiveThirtyEight keeps a list of pollster ratings.
Isn’t it interesting that it swung from D+4 to D+11 in one poll!
Yeah. That’s why we look at the rolling average. An individual poll is going to vary. The rolling average smooths that out.
This poll is imperfect because of X! I disagree with its methodology.
All polls are imperfect. That’s why we take an average, and look at it over time. That mitigates the imperfections of individual polls.
You’re only talking about this because the RCP average shows a high Democratic advantage right now.
It has consistently showed a large Democratic advantage. The aberration, the cherry pick, would be to post about it when it’s at the second lowest point of the year.
You’re clearly rooting for the Democrats!
Whether I am or I’m not is completely irrelevant to what the RCP average is or what the FiveThirtyEight model says.
You’re so blinded by hatred of Trump that…
That’s not an argument. Again, whether I am or I’m not is completely irrelevant to what the RCP average is or what the FiveThirtyEight model says.
All polling is all broken.
It’s really not though. When it’s done well and used in the correct way, it’s actually a very useful tool. Even in 2016, well done polls were pretty accurate.
A lot of Trump supporters refuse to talk to pollsters. We’re too busy living our lives.
That would be interesting … if there was any data to back it up. (Note: the plural of “anecdote” is not “data.”) I’ve heard Republicans trot this chestnut out to explain inconvenient polls for at least 25 years. Not for nothing, but the idea of shy Trump supporters seems … odd to me. If you think Trump supporters are too busy to share their opinions, obviously you’re not on Twitter.
Published in General
I’ll offer one more objection, one you didn’t mention: there’s a difference between “victory for Democrats” and a Blue Wave. I think Democrat control of the House is likely following the election — likely and, in some respects, advantageous to the right (though I still hope it doesn’t happen).
But the arguments about urban concentration diminishing the Democratic preference advantage have legitimacy. And there are anomalies in the summer polling that suggest the out-party advantage may be different — and weaker — than in past years. (Guest Henry Olsen has interesting things to say about that in this week’s Powerline podcast — well worth hearing.)
I am going to assume the Democrats take back the House and the Republicans keep the Senate. I don’t expect a wave election, merely something closer to the typical off-year gains of the out-party.
Sure. And what’s your basis for that assumption?
Good points Fred. All GOP should just stay home and say screw it.
Okay, we are in total disagreement.
I’m not playing the game of why-for’s and what/where is the data.
Are you willing to wager on it?
Y’know, put some skin in the game, a la Ricochet?
If so, we’ll define terms and skin up.
If not, well, then…
Well, for one thing, the polling you presented: it shows a range of D+14 to D+55 seats, but anticipates only D+34. In 2010, a legitimate “wave” year, the Republicans gained 63 House seats and a half dozen Senate seats. FiveThirtyEight is predicting barely half that House gain — in my book, that falls short of a wave.
Again, that urban concentration factor is real: Trump animus is concentrated in the populous but already-left urban areas: in most of the country, Trump is actually above 50% in the polls. The pollsters will try to control for that, and they generally do a pretty good job, but polling urban voters is different from polling rural voters, as we saw in 2016.
Inventory numbers suggest a strong third quarter. Consumer confidence is high and rising. The Kavanaugh hearings are likely to go well for conservatives. Frustration with a left-tilting Democratic party is, I suspect, stronger than people realize. Socialism is a stupid idea and I don’t think it’s going to play well. Democrats are losing the crucial white millennial vote, particularly among males — something hitherto unexpected.
And, as Mr. Olsen points out in that Powerline piece I linked in #1, something isn’t normal about this year’s out-party trending. Historically, the out-party (the Democrats) should have received an end-of-July bump in the polls. They didn’t, and no one knows why.
Trump’s popularity fluctuates, there are big stories going on — Kavanaugh, Mueller, press and DOJ corruption — and the economy is booming. The Democrats are shifting further left by the day.
None of that would make me feel comfortable if I were predicting a blue wave, versus more modest and historically normal Democratic wins.
I note in passing that you can barely contain your glee.
I also note your repeated assertions that Donald Trump will not win in November 2016.
So I guess we know how good your predictive powers are.
Let’s make it interesting, Fred. I said in a post back in May (which I cannot find right now, because of the appalling limitations of the Ricochet post archiving system) that I was willing to make the ultimate wager on the outcome of this election.
If we lose the House, I will unsubscribe to Ricochet. Will you take me up on that wager, Fred?
You misspelled “straw men.”
This is a fair post. It’s not a welcome one, but sometimes it’s useful to sober up and realize we could lose this thing. In 2010, the Dems didn’t see it coming, that’s true…and in 2006, we got a “thumpin” (in President Bush’s word) that we didn’t expect. As Fred says in his intro, neither this post nor the earlier one has any secret new information. It’s impossible to tell what’s going to happen.
Pollsters do a pretty good job on candidate match-ups. The pollster activity that I dislike, and that I think most conservatives dislike, is a different type of polling, where framing the question frames the answer: “Do you think rents and prices should be fair?” “Would you be more or less inclined to support a SCOTUS nominee with a savage anti-woman agenda?” “Do you support mothers controlling their own wombs?” That’s where you make the numbers sing any song you want them to.
On Ricochet we say “The Prez has topped 50%!” if it’s on Rasmussen. Remember when we all loved Zogby, because he told us just what we wanted to believe?
Does that mean if we retain the House, Fred leaves Ricochet?
Oh and BTW, 538 gave Trump <30% chance of winning. I guess it’s easier to to apologize about getting it wrong “like everyone else did” (not me) than it is to give people an accurate forecast. BTW 2, “registered voter” and “likely voter” ain’t the same thing.
The LA Times/USC poll got it right in 2016–a Trump victory. People in town hated the paper for being right. So what’s this gutsy “rebel” poll say about 2018? It says, basically, that we’re going to get creamed.
Does that mean it’s certain? Of course not.
A diffrent Nate, Nate Cohn of the NYT, paints a different picture.
https://www.nytimes.com/interactive/2018/07/19/upshot/democrats-midterm-elections.html
Should I add that to the objections section?
Sure. Can you give me an example?
That’s a terrible wager. That’s lose-lose. No matter what happens, everyone loses. Why would I take a wager like that?
A couple more data points: 4.2% Q2 economic growth and 4.1% estimated Q3 economic growth. If the actual growth rate is anywhere near this number, it will help Republicans. Oh, and the numbers will come out on October 26, right before the elections. There won’t be enough time for that story to die down.
Yeah. I mentioned that and linked to it in the OP.
Concur, wholeheartedly. The one thing we know is that no one knows nuthin’. So let’s wait; we’ll all know soon enough.
If you were really confident, you would. I’ll just tell you now, after we hold the House this year, I’m gonna be insufferable. I’m not gonna let anyone forget how wrong you were.
Just like the data strongly pointed to a Hillary landslide.
One way I keep my stress level low when it comes to elections is to ignore the polls, then vote when the day comes. If I didn’t know any better, it sounds as if you want a blue wave . . .
( By the way, it isn’t about Fred. It’s about ideas, facts, interpretations. Making it about Fred just increases the urge to engage in trolling behavior — on his part and ours. )
Historically speaking, we’re likely to lose the House. If one goes by the popular press, we’ll lose the House, Senate, and Oval Office upon impeachment — but the press lie in service to their prevailing ideology, which is left.
We’re likely to lose the House, but there’s a decent chance that we won’t, and a much better chance that this will fall short of a “wave” election: we’ll continue to dominate at the state level, we’ll keep our Senate majority, and we’ll take back the house in two years when Trump is reelected. That’s what I expect to happen.
I’m sure you also explained how this year it’s different. The polls will behave this time.
Depends on what you mean by “we” Gary. If you are a GOP coalition supporter like you and I, a creaming in the House for “us” seems likely.
If however one is principally fueled by Trump derangement, there’s little comfort in a midterm shellacking.
1982, 1994 and 2010 are staring back at you. 1990 and 2002 also testify to the weak predictive value of midterms.
A loss of the House while picking up strength in the Senate would make a very interesting 2019.
Trump and Cocaine Mitch approving judges, exercising shoe-on-the-other-foot executive action, slashing regulations and, unfortunately, dealing with Schumer to make 2019 the year of infrastructure.
It’s going to be an ugly year.
Trump’s going to love it.
Nope, I am a Democrat so I am on your side.
No . . . not at all.
You’re right, @henryracette. This is about ideas. My original post and the one back in May discussed many ideas. The RCP numbers fluctuate wildly, like the ramblings of a mad man. They narrowed considerably in late Spring, followed by a considerable widening in Summer, followed by another narrowing last month. Now, they look good for Democrats. Based on the pattern of the past few months, the numbers will be back below D+5 on Election Day. With our built-in advantage, that’s not enough for the Dems to take the House.
But these aren’t ideas. This is just horse race, which skirts the fundamental question: with such a man in the White House and structural midterm trends what they are, why have the polls been this close? Economic growth! No one has been able to explain to me what is so bad about 4% economic growth?
It’s not the economy, stupid? I don’t buy it. No one has explained to me why the President’s personality matters so much more than jobs and the amount of money in people’s paychecks. To me, economy good = incumbent party advantage is the null hypothesis. You’ve gotta provide a lot of evidence to reject the null hypothesis.
He doesn’t have the courage of his convictions. He’s afraid he’ll lose.
In other words, for this election they are predicting the same thing as the 2016 election.
2016 Trump win – 28%
2018 – Republican house 25%