As we have all observed, Nate Silver and other purveyors of “data journalism” have gotten a lot of flak early in this election cycle, some of it warranted, but much of it the kind of taunting schoolyard kingpins typically inflict on tyros they instinctively understand will be rich and powerful someday.
Close to the line between rational objection and special pleading is a column from National Journal‘s house conservative, Josh Kraushaar, who begins with this hackish complaint about 538’s early Senate projections:
[C]ount me underwhelmed by the new wave of Senate prediction models assessing the probability of Republicans winning the upper chamber by one-tenth of a percentage point. It’s not that the models aren’t effective at what they’re designed to do. It’s that the methodology behind them is flawed. Unlike baseball, where the sample size runs in the thousands of at-bats or innings pitched, these models overemphasize a handful of early polls at the expense of on-the-ground intelligence on candidate quality. As Silver might put it, there’s a lot of noise to the signal.
So? Has Nate somehow failed to observe that the projections will become more reliable the closer we get to November? Or is there something else he’s missing? Yeah, that’s it:
The models also undervalue the big-picture indicators suggesting that 2014 is shaping up to be a wave election for Republicans, the type of environment where even seemingly safe incumbents can become endangered. Nearly every national poll, including Tuesday’s ABC News/Washington Post survey, contains ominous news for Senate Democrats. President Obama’s job approval is at an all-time low of 41 percent, and public opinion on his health care law hasn’t budged and remains a driving force in turning out disaffected voters to the polls to register their anger. Public opinion on the economy isn’t any better than it was before the 2010 midterms when the unemployment rate hit double-digits. Democrats hold only a 1-point lead on the generic ballot in the ABC/WaPo survey—worse positioning than before the GOP’s 2010 landslide.
Now if cherry-picking the most bleak of national indicators and then comparing them to indicators that largely proved wrong in 2010 proves another Republican “wave” is on the way, then it will always, always seem apparent just on the horizon to those who want to see it. National indicators, BTW, are just as subject to change as state polls, and Silver, BTW, does factor in Obama’s approval ratings and economic conditions.
But then having done the journalistic equivalent of “trash-talking,” Kraushaar eschews said practice and offers his own, quasi-empirically based projections, which (with the exception of a strange, wonder-if-they-are-related paean to Iowa Senate candidate Joni Ernst) follow pretty much the same sorts of micro-indicators Silver uses.
So it’s tough to figure out when a guy like Kaushaar is spinning or telling us what he really thinks. That’s generally not a problem for Nate Silver.
In case anyone reading this blog hasn’t figured this out, I”m proud to say my own election musings are reasonably free from suspicion that I’m shilling for any “team.” Yes, I may occasionally seem to be cheerleading for aggressively weird Republican candidates, but if I’m biased that’s probably because it’s a lot more fun to write about rabid ideologues than buttoned-down automatons, and I also may be defending a general perspective that the GOP is in the grip of an ideological bender that is the closest thing we have right now to a single key to American politics.
As it happens, even as a lot of observers from both “teams” have begun more highly rating the odds of Democrats surviving this November with a majority, I’ve been moving in the opposite direction–not due to any sense of a “wave,” but because of verifiable state-by-state numbers, and not just current polls.
For example: Nate Cohn’s recent stunning analysis of the midterm-presidential turnout “gap” and partisan preferences of old folks and young folks in North Carolina makes me wonder how, mechanically, Kay Hagan is going to build a majority from an electorate that would have given Mitt Romney a ten-point victory if 2012 were re-run in 2014. Similarly, I look at Louisiana and wonder if Mary Landrieu can equal her 33% performance among white voters in 2008 (as compared to 14% for Obama), and if that will be enough given the strong likelihood the white percentage of the electorate will be closer to 2010’s 71% than 2008’s 65%. I also wonder if Democrats can equal their astonishing post-November turnout effort in Louisiana back in 2002 (the last time Landrieu had to run under “jungle primary” procedures). Maybe I’m just not feelin’ a wave, eh?
Truth is, as we do approach November, you should mistrust anyone who is still talking about “waves” and “momentum” and “enthusiasm” and other intangibles instead of just looking at the gradually accumulating data. I know a lot of readers dislike polls, but they are more reliable than talking out of one’s hat and placing thumbs on the scale for “the team.”
