1. An Unsustainable Strategy. It’s become increasingly clear that one strategy of the White House this midterm is to amp up African-American and Hispanic turnout, and this has occasionally gone so far as to include pitting ethnic groups against each other.
This might help the White House stave off some losses next week, but sooner rather than later President Obama is going to have to pull back from this approach, and try to get back to the “post-racial” dynamic of his campaign. At least, he will if he wants to be reelected to a second term.
The latest Gallup job approval numbers by racial and ethnic groups finds President Obama’s job approval among whites at just 35%. This is no outlying result, as this graph indicates:
[img nocaption float=”center” width=”640″ height=”463″ render=”<%photoRenderType%>”]12155[/img]
His average this month (so far) is 36%. In comparison, he won 43% of the white vote in November 2008. If he had instead won 36%, with everything else being constant, he would have won just 48% of the overall popular vote, and we’d be humming Hail to the Chief whenever John McCain walks into the room. Relatedly, if you re-run the 2006 midterm election with the Democrats winning 36% of the white vote (rather than the 47% they actually carried) and performing the same with all other groups, they would have gotten 45% of the vote, the party’s worst showing since 1946.
2. Hulkamania! Yesterday, I made the case for why the 2010 midterm could be very, very bad for House Democrats. I called it my “Incredible Hulk” argument. I don’t buy it just yet, but I have to acknowledge it. It seems that I am not the only one.
Nate Silver. Silver projects a 52-seat pickup. He adds: “[E]rrors could also work in Republicans’ favor, potentially enabling gains in excess of 60 or even 70 seats. And much of the data released within the last day suggests that, if anything, they are strengthening their position.”
Cook Political. Charlie Cook projects a range of 48-60 pickups. He adds: “It would be a surprise if this wave doesn’t match the 52-seat gain on Election Night in 1994, and it could be substantially more.”
Stuart Rothenberg. Rothenberg projects a range of 45-55 pickups. He adds: “With a week to go until Election Day, House Democrats face the potential of a political bloodbath the size of which we haven’t seen since the presidency of Franklin Delano Roosevelt. The largest midterm House loss for the president’s party during the last 50 years was 52 seats in 1994. The previous largest losses were 55 seats in 1942 and 71 seats in 1938.”
What’s going on here? Everybody seems to cluster around roughly the same numbers, then offers a caveat like, “but it could get a hell of a lot bigger!” Why is that?
I can only speak for myself, and my thinking is that there is just so much uncertainty. There is the uncertainty of turnout. There is the uncertainty of the independents and how they break. There is the uncertainty of the generic ballot and whether it will be next week what it is today. There is the uncertainty that comes with analyzing 435 discrete congressional districts with precious little data. And this uncertainty is multiplicative. If I’m 80% certain that Event A is going to happen and I’m 80% certain that Event B is going to happen, I’m only 64% certain that Event A and B are going to happen.
What’s interesting is that the general consensus seems to be that this could get much worse for the Democrats, but not much better. I think it’s like a football game where one team is simply better than the other – like Super Bowl XX, the Bears versus the Patriots. By halftime, the Bears had a 23-3 lead over the Pats, and anybody worth his salt would have predicted that they were going to win. The only real question was whether the final score would be by 14 points or more. (Final Score, 46-10 for Da Bears).
I think that is what is going on here. This midterm election is one of those Super Bowls that is over by the third quarter. The real question now is how many points does the GOP put on the board before the ref blows the whistle and puts the Democrats out of their misery. Could be 50, could be 60, could be…?
One other note about my estimate is that it is much more informal than any of these three (or Sabato, for that matter). I’m using what I’d call a “Keep It Simple, Sam” system. I’m just taking the independent vote, plugging it into the 2004 results, and comparing that to 1994 (when the GOP won the popular vote by 7 points). Really, anybody could do what I’m doing. No “propiertary prediction models.” No insider scoops. Just the generic ballot results interpreted in light of postwar electoral history.
3. Is Pot Helping Boxer? That’s the suggestion from SurveyUSA:
In California, the opposite. Uniquely motivated 2010 voters are more Democratic…[I]n the Senate contest: the incumbent Democrat trails by 4 points among habitual voters, and nominally leads by the 2 points that SurveyUSA reports here only when these uniquely motivated voters are included…Are the “Yes” on marijuana voters the tail wagging Barbara Boxer’s dog? This is more plausible to SurveyUSA than the reverse, which would be that 3-term incumbent Boxer has a unique tractor-beam in 2010 that is drawing to the polls otherwise disaffected voters who just happen to be pushing Proposition 19 over the top.
The outfit’s latest survey shows Boxer expanding her lead to five points, but that movement is due entirely to a decline in Fiorina’s standing. In their latest poll, regular midterm voters back Fiorina, but the non-regular midterm voters who pass through the likely voter screen because of high enthusiasm back Boxer 2:1. They also back Prop 19 by 14 points.
4. The Latest Sign of the Dempocalypse. This is from Jeremy Jacobs at the Hotline OnCall:
Some of the polls are so striking that next Wednesday, the day after the midterm election, observers may turn to these surveys as a symbol of when the bottom fell out for Democrats.
In New York’s 20th District, Rep. Scott Murphy (D) has gone from holding a 17 point lead over Republican Chris Gibson last month to trailing him by nine — 51 percent to 42 percent…
Similarly, Rep. Walt Minnick’s (D-Idaho) race against Republican Raul Labrador (R) has narrowed dramatically to Minnick leading by only three points, 44 to 41 percent, in a new Mason-Dixon poll…
Even more telling, the Democratic Congressional Campaign Committee released an internal poll that showed Rep. Jim Costa (D-Calif.), a Democrat who wasn’t on many vulnerable lists weeks ago, leading by only six points and under 50 percent — 47 percent to 41 percent. That was in response to a public, automated SurveyUSA poll showing the congressman trailing by 10 points, 52 to 42 percent, despite representing a district that gave President Obama 60 percent of the vote.
And in New Mexico’s 1st District, Republican Jon Barela (R) has taken the lead over Rep. Martin Heinrich (D) for the first time, according to Barela campaign polling obtained by Hotline On Call…
