Reading the Returns: Election Night Analysis

By now, everyone's seen the results. As expected, Obama swept today's Democratic contests, winning in Washington, Nebraska, and Louisana. Less anticipated was his margin of victory. In the caucus states, he won better than 2/3 of the votes; in Louisiana, by double-digits.

So how do we make sense of those margins, and attempt to figure out their significance? Three ways spring to mind.

1) Polling:

Washington State: Only one polling firm released results from Washington in February: SurveyUSA. It found Obama ahead among likely caucus-goers 63-33. So score one for the much-battered reputation of the robo-polling firm, and mark these results 'as expected.'
Nebraska and Louisiana: I'm not aware of recent polling in either of these states.

2) Internal expectations:

Obama Campaign: An internal campaign spreadsheet that was leaked to Bloomberg News gives us a rare view of what the campaign itself expected. The first of the projected scenarios had Obama winning 60-40 in WA and NE, and 54-46 in Louisiana. At this hour, Obama's actual margins in all three states exceed those counts (although in Louisiana, due to Hillary falling short). That scenario had Obama winning the pledged delegate count 95-63 tonight, and 1,647 to 1,580 overall. We'll see how that works out - one projection has him winning Nebraska delegates 14-10, while CNN puts him at 15-9, just like the spreadsheet. In WA and LA, the situation is still too confused to hazard a guess. But with the margins bigger than he expected, you've gotta believe that David Plouffe is going to be a happy man tonight.

Clinton Campaign: The first rule of expectations management is that it's better to be pleasantly surprised than suddenly dissappointed. If that applies to the Obama camp's internal projections (a good reason to take them with a grain of salt) it applies as well to the Clinton camp's memorandum to reporters as well. To quote:

The Obama campaign has dramatically outspent our campaign in these three states,
saturating the airwaves...we will continue to compete in [February states] and
hope to secure as many delegates as we can...

That about sums it up. The name of the game for the Clinton camp was to keep these races close enough to amass some delegates. We'll have to wait till morning (at the earliest) before we get reliable delegate counts. But when expectations start that low, it's tough to be dissappointed.

Edwards campaign: Yes, you read that correctly. In the big surprise of the evening, the moribund campaign of John Edwards roared back to life. Well, not really. But confounding expectations, it seems that thousands of (white?) Louisiana voters who were left to choose between a black man and a woman chose...neither. He didn't break the 15% threshold, so he won't walk away with any delegates, but I think we can faily say that he exceeded expectations. For Edwards, it must be a tantalizing reminder of what might have been. And for the remaining two contenders, it's a troubling indicator for November.

3) Exit Polls: Alas, caucus states don't get much love from the consortium, so we're stuck with the exits from Louisiana. Here, there's plenty for both camps to enjoy:

Obama campaign: There have to be some smiles in Illinois tonight. Obama won among both men and women, among voters in every age cohort up to 65, among religious voters (and ran almost even among the more religious Catholics), with those earning less and more than $50k, and among both Democrats and Independents. But there was plenty to frown about, to, including this statistic that's likely to get some play: 35% of primary voters said they would not be satisfied if Obama wins the nomination. (But then again, 12% of them voted for Obama - so much for polling.)

Clinton campaign: If the tally from Louisiana was discouraging, the exit poll bodes well for Clinton's chances, particularly in the crucial states of Ohio and Pennsylvania. White voters supported Clinton (70-26) by margins almost as large as blacks went for Obama (82-18). She won Catholics, including 73% of white Catholics. Voters thought she "cares about people," has "experience" and "electability" - she lost out only on "can bring change." The poorer the voter, the gloomier about their personal economic future and the nation's, the better Hillary did. And she brought home her bedrock supporters - 63% of voters over the age of 65 cast their ballots for Clinton. If she can replicate those margins in states with different demographic compositions, she can win in a landslide.

So there you have it. On the whole, the evening went about as expected. Three dramatic wins for Obama, whose lead among pledged delegates continues to mount. Clinton can take solace in the thought that the national media expected these wins, and so is unlikely to play them to her disadvantage, and that the biggest prizes left on the calendar are rust-belt states full of aging, white-ethnic voters.

If you find this worthwhile, please click the 'recommend this' link, so that other readers can share it. More election and polling analysis is available on my blog (click my name). And, as always, I welcome your comments and corrections.