(Check the bottom of this item for a 5 p.m. update.)
I blogged last week on the intricacies of counting the vote in the Democratic nominating contest, and proposed a better estimate – a count in which Barack Obama led, even with Michigan and Florida included. Hillary Clinton apparently didn’t get the memo.
Clinton’s continued to proclaim herself ahead in the popular vote; apparently she thinks it helps in her effort to win over superdelegates. (The recent direction of superdelegate preferences suggests otherwise, but that’s another blog.) So we've updated with the Kentucky and Oregon results to see if the picture's changed.
Answer: Yes, in one scenario. Given her net of 154,706 votes yesterday (in the count so far - it's not done in Oregon) our estimate now has Clinton ahead, by just over 52,000 votes out of more than 36 million cast, if we include both Florida (where neither candidate campaigned) and Michigan (where Obama wasn't on the ballot). Exclude Florida and Michigan, or just exclude Michigan, or give Obama a chunk of the Michigan uncommitted vote, and he's ahead.
Here are our estimates, with 88 percent of the estimated vote in Oregon reported:
With Without With FL, MI and FL MI and FL Without MIObama 18,148,061 17,571,847 18,148,061Clinton 18,200,357 17,001,062 17,872,048
Cl +52,296 Ob +570,785 Ob +276,013
As noted, the Oregon count isn't completed; we'll update when that happens. But our decision desk director, Dan Merkle, says it doesn't look like Obama will make up the more than 52,000 voters there that he'd need to surpass Clinton in the "all states" category.
You can see how we arrive at these calculations in the original blog. To recap briefly, the official counts you’ve seen elsewhere are based on delegate counts, not voter turnout, in five caucus states where voter preferences weren't tallied. We’ve adjusted as follows:
-In Iowa and Nevada we are using turnout reported by the state parties, allocated by vote preference in the entrance polls.
-In Maine we are using turnout reported by the state party, allocated by initial delegate proportions.
-In Washington we are using the beauty contest primary results.
-In Texas we are using both the primary results, and caucus turnout reported by the state party (given as “just under a million,” calculated at 900,000) allocated by initial delegate proportions.
Note, the estimate including Michigan gives Obama zero, since he wasn't on the ballot. There could be an argument for giving him all or some of the uncommitted vote there, since it included, for instance, 68 percent of African Americans, 50 percent of postgraduates and 50 percent in $100,000+ households, all among his core support groups. ("Uncommittted" beat Clinton in those groups, and a few others.) But we're not going there; this calculation sticks with our estimate of actual votes for actual candidates. Who knew they'd be this tricky to count.
5 p.m. update:
The AP today updated its vote totals in Idaho, Indiana and North Carolina, as well as in Oregon, where the count is now an estimated 96 percent complete. These have inched things in Obama’s favor:
With Without With FL, MI and FL MI and FL Without MIObama 18,176,329 17,600,115 18,176,329Clinton 18,214,506 17,015,211 17,886,197 Cl +38,177 Ob +584,904 Ob +290,132
In the all-states count, Clinton’s lead, if you want to call it that, is a smidge over one-tenth of a percentage point. Obama’s is 1.7 percentage points without Michigan and Florida, eight-tenths of a point with Florida but not Michigan.
A Clinton staffer wrote in to ask how we can include the Texas caucuses, since that double-counts primary voters who also went to the caucuses. We discussed this in Friday's item. But it doesn’t make a big difference; take out our estimate for the Texas caucuses and Clinton gains 108,000 votes – still slightly ahead with Michigan and Florida, still a little less slightly behind without.
And as our political director, David Chalian, keeps saying - it's supposed to be about delegates.