A House Disappointment
There’s some sentiment out and about that Democrats “underperformed” in House of Representatives races, perhaps because (in Mickey Kaus’ words) “swing voters compensated for the bold, hopeful risk they took on Obama (including for overcoming any race prejudice) by gravitating back toward Republicans in their local Senate and House races.” In reality, as Andrew Gelman shows, there was a strong uniform swing toward the House Democrats:
As he notes, this was actually larger and more uniform than the fairly large and fairly uniform swing toward Obama:
The House Democrats will presumably have outperformed the presidential ticket in terms of swing because they can individually tailor their message to the different natures of politically and socioeconomically diverse congressional districts.
All told, 56 percent of Americans voted for a House Democrat whereas only 52 percent voted for a Republican in 1994. That’s a larger majority than Obama got and, indeed, would have been considered a pretty crushing landslide on the presidential label. It’s difficult to interpret what voters “want” based on their voting behavior, but it’s difficult to see this outcome as consistent with the idea that voters became nervous that congressional Democrats would rubber stamp an Obama agenda. There was both a strong level of support for House Democrats and a strong trend toward supporting them.
61 Responses to “A House Disappointment”
Register
RSSLogin
Topic Cloud
ArchivesBlog RollAbout Matt YglesiasContact Me
Donate
November 11th, 2008 at 11:00 am
“There’s some sentiment out and about that Democrats “underperformed” in House of Representatives races … In reality, as Andrew Gelman shows, there was a strong uniform swing toward the House Democrats”
Well, if you’re writing propaganda, they overperformed. However, in reality…
Before the election, Stu Rothenberg said the Dems were going to pick up 27 - 33 seats, with a good shot at hitting 40. Charlie Cook said the Dems were going to pick up 25 - 30 seats, with a decent shot at hitting 35 - 40.
But the Dems ended up with ~22 seats.
The Democracy Corp polling of swing districts topped out about a week before the elections, the fact that we lost the last week of the national campaign cost us 5 to 10 seats.
That’s underperformance. And you can see the same thing in Senate races. We didn’t close the campaign properly, and it cost us.
November 11th, 2008 at 11:14 am
Clearly the Democrats can only consider their overwhelming victories for 2 elections in a row and capturing all branches of government and a near “filibuster” (were such things actually required) proof majority as a failure.
This is clear evidence that Barack Obama cannot close the deal. If only some sage had warned us in advance about that.
November 11th, 2008 at 11:17 am
Why does anyone read Kaus? He is mostly wrong - often intentionally. He just likes to throw memes out there to see what sticks.
If you believed Kaus you’d think Americans were gonna freak out about Ayers - and legitimately so! You would have thought Al Gore spent all his time making up stories about Love Canal and that “The Bell Curve” correctly governs American life and welfare mamas compete with General Dynamics for pork and that Obama should really clear the air about his birth certificate because Instapundit is smart and libertarian and Clinton may have killed people and Gore tried to claim he invented the internet and Obama is friends with terrorists etc
Yeah - Kaus is really that dumb.
November 11th, 2008 at 11:24 am
Btw - have you ever seen Kaus on Bloggingheads? Everything he offers up is frivolous trivia. He revels in those fake issues that coincidentally happen to help Republicans.
He helps to legitimize phony topics that would otherwise be stuck in a Limbaugh ghetto. Often he dwells on racial topics dear to the talk radio set.
November 11th, 2008 at 11:26 am
You don’t have to be very bright to see that, if Obama wins 60% of the vote in a safe red state, he gets all the EVs but the Dems only pick up 60% of the seats in the House (even with a raft of generous assumptions that don’t actually hold). And Kaus has in the past noted how few congressional seats are actually competitive (by way of arguing for re-districting reform in CA without insisting on it in TX, naturally). So he isn’t stupid enough to really believe that Obama’s electoral margin should have been replicated in the 435 House races, he just hopes that his readers are.
November 11th, 2008 at 11:27 am
You’re telling me that Mickey Kaus didn’t know what the fuck he was talking about? My whole world is shattered.
November 11th, 2008 at 11:27 am
A lot of blue-state Dems got complacent in the last week, and some House races reflected it with much smaller than expected D percentages (Adler in NJ3, Shulman in NJ5, Stender in NJ7, Massa in NY26, Madia in MN3, and CA4 to name a few).
Meanwhile, in tossup states, energized Dems did very well, flopping 3 seats in Virginia, a seat and a Senator in NC, getting Chabot and possibly the 15th in Ohio, winning all 3 NM seats, and knocking off Porter in NV.
I also wonder if the House Ds outperforming Obama are a function of all those urban seats that go 80%+ to the Dems while even the reddest of the red districts usually top out around 65%-ish. Very few disricts reach a PVI of R+20, while districts that are D+30 are much more common.
November 11th, 2008 at 11:28 am
Before the election, Stu Rothenberg said the Dems were going to pick up 27 - 33 seats, with a good shot at hitting 40. Charlie Cook said the Dems were going to pick up 25 - 30 seats, with a decent shot at hitting 35 - 40.
And why should the predictions of Stu Rothenberg and Charlie Cook represent the baseline against which the question of over- or under-performance is measured?
November 11th, 2008 at 11:29 am
It’s all about your baselines.
Petey thinks that the baseline should have been the 57-58% that was sorta-kinda projected by Cook et al.
I think that’s insane. Underperforming a projection that called for one of the largest landslides in congressional history is simply not an issue to me. We merely followed up one landslide (2006) with a small landslide. We just didn’t follow it up with a historic landslide. Darn.
Also, Kaus’ argument is about the relation between Obama’s vote and the House vote. Kaus is obviously wrong because we did better in the House vote.
November 11th, 2008 at 11:35 am
You know, before the election I had a vision (must have been those mushrooms) that the Dems would pick up 49 seats. They really underperformed my vision badly. Damn, we must have lost that last week of the campaign.
November 11th, 2008 at 11:45 am
Since the blue line represents the average, this is hardly surprising. The thinner black line represents 2004 performance. The differing size of the gaps between the two lines shows that Democrats made more gains in House vote share from 2004-2008 than they did in Presidential vote share.
Also, I agree that using talking-head predictions as a standard is dumb. Talking heads often overdramatize their predictions to make them sound more interesting.
November 11th, 2008 at 11:45 am
It looks to me as though there about as many states below the blue line on the first chart as there are above it.
Yes, because, amazingly enough, the blue line on each chart represents the average. And you’ll see that the blue line on the House chart is higher than the blue line on the Obama chart.
Try again.
November 11th, 2008 at 11:46 am
Chris beat me to it, dammit.
November 11th, 2008 at 11:49 am
I’m convinced that Matt actually has a Peteybot that is programmed to repeatedly insert the same unsupported assertions in every thread.
November 11th, 2008 at 11:50 am
So why is Kaus wrong? Why does he make up stuff that seems to be unimportant?
Maybe he wants to boost the fake “we are a center-right country” talking point that is all the rage with pundits.
Who knows - but he pulls this crap all the time and it has become tiresome since he is still quoted in the press and on blogs.
Just listen to him spout his garbage on bloggingheads - almost everything wrong or irrelevent or stupid or immature. Then ask youself why there is a market for this kind of crap coming from that person.
November 11th, 2008 at 11:55 am
“Underperforming a projection that called for one of the largest landslides in congressional history is simply not an issue to me.”
A 22 seat pickup is “one of the largest landslides in congressional history”?!?!?
Ummm…
(In his defense, DivGuy is usually not wrong on this kind of comic level. I’ll attribute it to him not having had a second cup of coffee.)
—–
“And why should the predictions of Stu Rothenberg and Charlie Cook represent the baseline against which the question of over- or under-performance is measured?”
The issue isn’t Rothenberg and Cook. They are merely aggregators of all of the data available.
What their projections are a clue of is that Dems would have done better by a non-trivial number of seats had the election happened a week earlier, given the timeliness of the data streams they were employing.
You can see the Dems peaking too early yourself by reviewing the publicly available data, with Democracy Corp’s files being a good place to start.
—–
“A lot of blue-state Dems got complacent in the last week, and some House races reflected it with much smaller than expected D percentages (Adler in NJ3, Shulman in NJ5, Stender in NJ7, Massa in NY26, Madia in MN3, and CA4 to name a few).”
It wasn’t a failure of those particular candidates. It was a failure of the national campaign in the final week.
November 11th, 2008 at 12:01 pm
Voters obviously thought McCain was riskier - Recall his shakey performance following the attack on Ossetia - Kaus regards Obama riskier soley because he assumed swing voters to be somewhat afraid of a black man as President. But no stats bear this out.
So Obama was able to pull more Dems along. He had coattails.
November 11th, 2008 at 12:02 pm
Shorter Petey: Making the perfect (as defined by me, of course) an enemy of the good and antagonizing natural allies in the name of picayune ideological distinctions are not just annoying tics, they’re a lifestyle!
November 11th, 2008 at 12:04 pm
“But in the House, there should be diminishing returns to any such shift following the first subsequent election. So, it is utterly unsurprising the Democrats did not gain as many seats in 2008 as 2006.”
FWIW, the last time a party won more than 20 seats in back-to-back elections, the wave was actually larger in the second election - 1932 was bigger than 1930.
November 11th, 2008 at 12:08 pm
I’m convinced that Matt actually has a Peteybot that is programmed to repeatedly insert the same unsupported assertions in every thread.
This! Matt enjoys torturing his commenters.
Kaus and Petey are professional concern trolls, spinning any fact in a negative way.
Obama had coattails obvious, in part b/c of turnout, and if you consider your baseline as if Hillary or Edwards had run, Obama did really well.
November 11th, 2008 at 12:26 pm
I agree with Petey. Folks, after EVERY election, we should be looking to analyze our own weaknesses. Just ’cause he was divisive in the primary season is no reason to dismiss him now.
Introspection is healthy. Call off the firing squad and crunch the numbers. BTW, Petey, which districts do you think we should have picked up? After all, national campaigns are run on a state-by-state basis.
November 11th, 2008 at 12:30 pm
Just ’cause he was divisive in the primary season is no reason to dismiss him now.
No, the reason to dismiss him is that he has no evidence to back up his claims. The fact that he’s a jackass is just gravy.
November 11th, 2008 at 12:31 pm
“Petey are professional concern trolls, spinning any fact in a negative way.”
If you think the House results aren’t a negative for us, you’re really not paying attention.
The ~258 seats we’re going to end up with is likely going to be around our ceiling for the foreseeable future, given the coming cycles. And we left a lot of low-hanging fruit on the table during the final week of the campaign.
Incumbency is a real asset, and maxing out your gains in good cycles is crucial to establishing long-term House dominance. Those low-hanging fruit are going to be much higher-hanging in coming elections. And given that I think the country is essentially governed from the House, such things matter to me.
But if you want happy talk, Peter K., as seems to be your inclination, if you’d told me in 2004 that we’d have ~258 seats in four years, I’d have been ecstatic. Happy now?
November 11th, 2008 at 12:38 pm
“Petey, which districts do you think we should have picked up? After all, national campaigns are run on a state-by-state basis.”
Well, I think the problem in the last week was a national problem, not a district-by-district problem. A rising tide lifts all boats and vice versa.
I think we lost the national television “air war” in the final week, and that resulted in losing a non-trivial number of cliffhanger election seats.
November 11th, 2008 at 12:55 pm
“Judged by those standards, this doesn’t look so bad”
This is the first time in 76 years that a Party has had two genuinely good House elections in a row. Of course that’s not so bad. But we knew that was coming by early October.
My repeated point in this thread is that this election does look so bad compared to how it looked a week before the vote. And I think that’s a result of some poor decisions.
(And you should group ‘94 in with ‘92, not ‘96…)
November 11th, 2008 at 12:58 pm
Petey is a wannabee.
November 11th, 2008 at 1:03 pm
“Petey is just on a perpetual quest for his contrarian views to be right about something”
FWIW, my views were highly popular, not contrarian, in places like this in 2007. My views are contrarian, not popular, in places like this in 2008.
Fashion.
November 11th, 2008 at 1:19 pm
A 22 seat pickup is “one of the largest landslides in congressional history”?!?!?
No, the predictions that you cited, had they come true, would have meant one of the largest landslides in congressional history. So
(a) the fact that a couple people predicted one of the largest landslides in congressional history
plus
(b) 2008 did not produce one of the largest landslides in congressional history
does not equal
(c) the Democrats underperformed.
November 11th, 2008 at 1:20 pm
Petey:
But if you want happy talk, Peter K., as seems to be your inclination, if you’d told me in 2004 that we’d have ~258 seats in four years, I’d have been ecstatic. Happy now?
Yeah, that’s better. Too bad Edwards went all Clintonian on us, or he could have been Attorney General.
November 11th, 2008 at 1:30 pm
“No, the predictions that you cited, had they come true, would have meant one of the largest landslides in congressional history.”
You’re nuts.
Cook and Rothenberg were projecting around 30 seats, which wouldn’t have been anywhere near “one of the largest landslides in congressional history”.
The table I have handy only goes back a century, but there have been seventeen House elections in the past century of congressional history with shifts larger than 30 seats, with several involving shifts of more than twice as many seats as that.
Guestimating to go back to the beginning of the republic, I’d guess a 30 seat pickup would be more like the 45th largest landslide in congressional history - in other words, right around the median House election…
November 11th, 2008 at 1:42 pm
“Guestimating to go back to the beginning of the republic, I’d guess a 30 seat pickup would be more like the 45th largest landslide in congressional history - in other words, right around the median House election…”
And of course, guesstimating back to the early days of the Republic, when the party system was weaker and less entrenched, and state parties were not using computer software to gerrymander safe seats for their incumbents, would be utterly retarded.
This are the largest back-to-back wins in the postwar era. That’s pretty goddamned good. If there is a single positive development that could possibly come from kvetching about Congressional Democrats underperforming Charlie Cook’s projections, other than providing some small measure of ass-covering for blog-commenters who have been relentlessly negative toward Obama for the past 8 months and are desperately cherry-picking factoids to make themselves appear slightly less foolish, I have no idea what it would be.
November 11th, 2008 at 1:50 pm
30 seats on top of a 30-seat majority would have been a crazy landslide. With House elections all happening every two years, the size of a landslide can only be measured when you know the underlying division of House seats.
When you only measure the size of the shift, you fail to account for context. It would be like shouting about the incredible Denver Nuggets offense without considering the pace of the games.
This is borne out by the basic popular vote numbers. Democrats got 56% of the vote in House elections. To win a Cook-size victory, we’d be looking at 58%, maybe more. That’s just a massive, massive victory.
November 11th, 2008 at 1:56 pm
“This are the largest back-to-back wins in the postwar era. That’s pretty goddamned good.”
I believe I’m repeatedly in agreement with this idea above.
“If there is a single positive development that could possibly come from kvetching about Congressional Democrats underperforming Charlie Cook’s projections, other than providing some small measure of ass-covering for blog-commenters who have been relentlessly negative toward Obama for the past 8 months and are desperately cherry-picking factoids to make themselves appear slightly less foolish, I have no idea what it would be.”
It’s odd. You think the only possible lens to view what I’m saying here is how I voted in the Democratic primaries.
Instead, I was originally stating that we lost the last week of the election pretty decisively, which is why most folks saw the final House results as a disappointment.
Then, a couple (!) of commenters decided that a 30 seat pickup would’ve been the among largest pickups in congressional history, and I corrected them.
You actually have a problem with any of that, LaFollette Progressive? It all seems pretty non-controversial to me.
—–
Look, I guess if you are still too busy celebrating the Obama victory to worry about how the campaign ended, more champagne to you. But the last week was lousy, and it cost us rather dearly.
November 11th, 2008 at 2:05 pm
“If there is a single positive development that could possibly come from kvetching about Congressional Democrats underperforming Charlie Cook’s projections, other than providing some small measure of ass-covering for blog-commenters who have been relentlessly negative toward Obama for the past 8 months and are desperately cherry-picking factoids to make themselves appear slightly less foolish, I have no idea what it would be.”
You know, I thought he’d stop his trashing of Obama once the man met any realisitic electoral expectations. I’ve asked Petey to make a plausible argument in favor of another Democrat performing better than Obama did on November 4. After all, wasn’t he the one constantly talking about how “Scottsdale was eating Chicago’s lunch” or some such nonsense daily since June?
The fact that he refuses to do so indicates that there are concerns other than the future success of the Democratic Party or progessive politics being worked out in his hysterical (in both senses of the word) rants.
November 11th, 2008 at 2:07 pm
People also vote for other reasons than to kick the other party out. Why do you think Romney won in MA? It wasn’t because he was a Republican. People didn’t really like his opponent Shannon O’Brien. I didn’t (but I held my nose.) Point being, many people can vote more for personalities than causes. IMHO, I personally believe its one of the main reasons why likable incumbents almost always win.
November 11th, 2008 at 2:25 pm
To somewhat support Petey:
The comparison of national vote share for House Democrats to Obama’s national popular vote result is very relevant to the falsity of Kaus’s claim and somewhat relevant to the question of how much or little Obama helped House Dems. It’s totally irrelevant to the question of whether the House Dems performed as well as projected and whether or not the House Dems performance was a disappointment. Essentially every projection had them performing better, if they come out of this with a gain of 25 (isn’t that still possible?) it will get them within the range of projections, but it’ll be the bottom of the range.
November 11th, 2008 at 2:27 pm
dems would have won more seats if Randy Scheunemann had Georgia fire more missiles in South Ossetia during the last week.
November 11th, 2008 at 2:47 pm
It’s totally irrelevant to the question of whether the House Dems performed as well as projected and whether or not the House Dems performance was a disappointment.
The problem is that Petey considers the former to be conclusively probative of the latter — sounds like you might as well, although I am loathe to impute agreement with Petey to anyone. There is, of course, a sense of the word “disappointment” which might make that true, but not any sense that most people in the real world would adopt. I mean, I always project to win huge at the blackjack table, so when I break even or only win a little, it’s a “disappointment.” But hardly disappointing in any real sense.
November 11th, 2008 at 3:06 pm
No, I don’t think it’s conclusively probative, I think we either need to figure out was wrong with the projections or conclude that they should have done at least slightly better and something or other caused them not to. Also, Darcy Burner lost again, which is disappointing.
Besides Petey’s cites to Rothenberg and Cook, there’s also Larry Sabato calling +26 and Pollster.com giving a range, if we assume that no “lean” seats change hands, of +10 to +33.
November 11th, 2008 at 3:08 pm
This are the largest back-to-back wins in the postwar era.
In 1994 the GOP picked up 54 seats in the House and 8 in the Senate, causing them to gain control of both from the Democrats. It was the first time in 40 years that the GOP had gained control of the House. The GOP also gained 12 governorships and 472 state legislative seats, gaining control of 20 state legislatures from the Democrats. This was the first time the GOP had held a majority of governorships in more than 20 years, and the first time the GOP had held a majority of state legs in 50 years.
In the 2008 election, the Dems had a huge incumbency advantage from the retirement of so many GOP congressmen.
November 11th, 2008 at 3:08 pm
Well, having established that the Dems have actually made historic gains (if looked at over two cycles instead of one) I’m left wondering what the point is of Petey’s “lousy last week” hypothesis is.
Is it… don’t screw up, ever? That is sound advice, although the execution can be difficult.
Is it… about a particular tactic that Petey thinks failed, and predicted so in advance, and which he is now cautioning against using in the future?
Is it… if we would have done better, we would have done better?
I mean, what is the purported value of this observation, here?
November 11th, 2008 at 3:23 pm
GOP retirements automatically create Dem incumbents?
November 11th, 2008 at 4:19 pm
GOP retirements automatically create Dem incumbents?
No, GOP retirements means the GOP loses the incumbency advantage in those seats.
Well, having established that the Dems have actually made historic gains
Sorry, what does this mean, exactly? “Historic gains” in what?
November 11th, 2008 at 5:02 pm
Does anyone else think the Dem vote was under-reported? Many “competitive” races, including open seats, happened to go red. It was reported that Minnesota’s two most closely contested races used a significant number of machines of the kind that were previously challenged in Michigan. Both Republican victors, Bachmann in MN-6 and Paulsen in MN-3, ran 7% stronger than where they last polled. The “close” races weren’t. Musgrave in Co and Goode in VA, maybe Doolittle in CA, however, lost uber-”safe” districts. Maybe in districts where little vote “fixing” was expected to be necessary? I think the true Dem vote was so great that it swamped machine fraud and even made “safe” GOP seats competitive.
November 11th, 2008 at 5:04 pm
“No, GOP retirements means [sic] the GOP loses the incumbency advantage in those seats.”
which is still not the same as “the Dems had a huge incumbency advantage”
November 11th, 2008 at 5:11 pm
which is still not the same as “the Dems had a huge incumbency advantage”
It provided a huge incumbency advantage to the Dems.
November 11th, 2008 at 7:29 pm
I predicted before the election that Dems would not only win every House and Senate seat (including the ones that weren’t up for a vote this year), but that they would, in fact, expand Cheney’s fourth branch of government by getting 600 more Vice Presidents elected, each with their own shadow government and undisclosed location. And so in that respect, we got totally slaughtered and should be ashamed of ourselves.
It’s obvious America wants Republican leadership, so we better just give it to them. Otherwise, our defeat next time might be even worse. Hell, we might even lose a seat. Yeah, it could be that bad.
November 11th, 2008 at 10:10 pm
If you set Charlie Cook’s “there is no way Dems pick up less than 24 seats” from the weekend before the election as the baseline then Dems underperformed. Maybe it was a lousy last week (surprised late deciders picked McCain since usually they break for the non-incumbent or non-incumbent party) but I think there is a different reason.
Obama didn’t make the election about Dems vs Repubs. This helped him in red states but didn’t help Dem challengers. There were more than a few Republicans who ran House campaigns echoing some of Obama’s rhetoric about bi-partisanship.
I think this is all for naught. In 1980 Republicans picked up 12 Senate seats and 35 House seats but Reagan still had to go through a House that was +50 Democrats. If Obama can convince or threaten the Republicans in states or districts he carried to get on board the size of the Dem pick up will not matter.
November 13th, 2008 at 2:25 pm
Okay, putting aside whether 22 is a big or a small number, remember that this cycle there was far less polling of individual house races than in 2006. As a result prognosticators were forced to rely on broad trends as opposed to race by race data. I doubt if Democrats were ever ahead in 30 or 40 races, but there is no way of knowing.
Furthermore, if you look at Rothenberg’s race by race predictions and assume the Tossups would be split evenly, the Dems would have won 26 seats. Cook by the same metric predicts a Democratic pickup of only 20.5 seats. The Dems did about as well as predicted by race by race analysis, the only way we “underperformed” was against the prognosticators’ dreamier guesses.
February 11th, 2009 at 1:14 pm
This is very interesting information. I just bookmarked it now.
March 1st, 2009 at 6:57 pm
cialis
I want to say - thank you for this!
March 11th, 2009 at 4:52 am
Very interesting site. Hope it will always be alive!
March 22nd, 2009 at 10:33 am
buy viagra online
Great site. Good info
March 22nd, 2009 at 7:55 pm
the comments here are having a laugh - i’ve added your blog to my netvibes account, keep up the good work
March 23rd, 2009 at 4:28 am
Wicked blog - the guys commenting here are full of it! I’ll be keeping up with your posts
April 2nd, 2009 at 5:29 am
Very interesting site. Hope it will always be alive!
buy cheap viagra
April 2nd, 2009 at 8:23 pm
Just wanted to say I loved the blog!
April 3rd, 2009 at 4:19 am
Excellent site. It was pleasant to me.
cheap brand pfizer viagra
April 3rd, 2009 at 8:41 pm
Yeah, Sometimes it will happen.
April 9th, 2009 at 6:50 am
I rarely comment on blogs but yours I had to stop and say Great Blog!! viagra
April 16th, 2009 at 10:08 pm
Just want to say i`m glad i found this site.
I am from Sweden and also now teach English, give true I wrote the following sentence: “Cruises and vacation packages, travel insurance online quote airline toll free numbers.”
Regards
Barnabas.