Department of “Huh?”

Latest polls look OK for Obama. But CBS/NYT has him even with Clinton nationally, but down 18 in the Super Tuesday states. Bzzzzzzzzztttt!!! Does not compute.

The polling is all over the map. Obama looked as if he were about to catch HRC in the two big tracking polls (Rasmussen and Gallup). Then both showed her pulling away again yesterday. Today, Rasmussen shows her pulling away still more &#8212 to an 11-point lead (49-38) &#8212 while Gallup shows the race tightening again, with Obama down by only two at 46-44.

The conventional (non-tracking) polls show:

Pew: Clinton +8 (46-38) [In the field Weds-Sat]

ABC/WaPo: Clinton +4 (47-43) [In the field Weds-Fri]

USAToday/Gallup: Clinton +1 (45-44) [a different poll from the Gallup tracker; in the field Weds-Sat]

CBS/NYT: Even (41-41) [In the field Weds-Sat]

So everything except Rasmussen seems to be consistent with a tight race, with Clinton ahead but movement toward Obama, especially among women; Rasmussen seems to be on a different planet.

Overall, not bad news for Obama fans, especially with the Field Poll showing California almost even (36-34 for Clinton) and the Zogby state-level polls full of good news (Obama +4 in California, +20 in Georgia, within a point in New Jersey and Missouri).

But, as Colombo used to say, there’s just one thing. The CBS/NYT results make no sense whatever. Nationally, the poll shows an even race. But in the Super Tuesday states it shows Clinton up by 18, at 49-31.

That can’t possibly be right, can it? If they’re even nationally, and HRC is way up in the Super Tuesday states, Obama must be ahead by a comparable margin in the non-Super Tuesday states, including Michigan and Florida (already voted) and Ohio, Pennsylvania, and Texas (still to come). It’s not as if Super Tuesday represents a homogeneous bloc of states; on the map, they look competely random. And the Pew poll shows the reverse pattern: Obama down 6 in Super Tuesday states, down 16 on post-Super Tuesday states. That makes slightly more sense, if intensive campaigning helps the underdog.

The obvious guess is that this is just a small-numbers problem; with only 472 Democratic primary voters in the overall sample, the Super Tuesday-state subsample must have been well under 300. But it really can’t be the case that Obama is even nationally and down 18 in the states that will vote on Tuesday. Can it?

Afterthought Everything would fall into place if a programming error at CBS had simply flipped the two groups of states. But that’s pretty hard to believe.

Second afterthought Pew shows Obama with a respectable 36% of the white vote, up 10 points over a month ago and 14 points over two months ago. But it also shows him with only 47% of the black vote, even with two months ago and down five points since early January. That surely can’t be right. Again, a small-numbers problem; the latest poll had only 97 black respondents. If we use Pew’s numbers for white voters but give Obama the same 2/3 or more of the black vote he gets in other polls (and got in South Carolina), then Clinton’s lead would be cut roughly in half, to about 4 points (assuming that 20% of the Democratic electorate is black), consistent with the other three national conventional polls released today.

Update The Rasmussen state-level polls don’t seem consistent with the Rasmussen national tracking poll showing HRC +11. Obama is now +1 (and gaining) in California, + 36 points in Illinois, +15 in Georgia. Clinton is only +6 in Alabama, having lost ten points in a week; even in New York she’s only ahead by 18, half the size of Obama’s lead in Illinois. Her only other double-digit leads (+12 in NJ, +14 in Tenn) are in polls taken January 30.

If I were in a betting mood, Obama at 40 cents on the dollar would look to me like an excellent bet.

Footnote Here’s a cheerful finding from the Pew poll: neither Obama’s unfavorables among Clinton voters (now 30%) nor Clinton’s unfavorables among Obama voters (now 31%) have been rising noticeably . So it looks as if (so far) the bitterness of the battle is largely restricted to the political junkies who read and write blogs.

Author: Mark Kleiman

Professor of Public Policy at the NYU Marron Institute for Urban Management and editor of the Journal of Drug Policy Analysis. Teaches about the methods of policy analysis about drug abuse control and crime control policy, working out the implications of two principles: that swift and certain sanctions don't have to be severe to be effective, and that well-designed threats usually don't have to be carried out. Books: Drugs and Drug Policy: What Everyone Needs to Know (with Jonathan Caulkins and Angela Hawken) When Brute Force Fails: How to Have Less Crime and Less Punishment (Princeton, 2009; named one of the "books of the year" by The Economist Against Excess: Drug Policy for Results (Basic, 1993) Marijuana: Costs of Abuse, Costs of Control (Greenwood, 1989) UCLA Homepage Curriculum Vitae Contact: