I'm surprised, but I do see a poll for NPR by POS/Greenburg-Quinlan with a tie last week. Is that the one you saw?

I'd be mighty surprised if that held up. Not many people are asking the "generic ballot" question yet, but in favorability ratings, polls are consistently showing the Congressional Dems as about +5 to +10 favorable (steady to slightly improving) and the GOP about about -15 to -30 unfavorably (and declining).

What makes that generic ballot finding so unexpected is that this is the biggest partisan gap in Congressional favorability in the 20 years of data I've seen. That is, usually both parties are relatively popular, or (more often) both unpopular. Usually the favorable numbers are within 10 points of each other, and often within 5. For there to be a 20 to 40 point gap favoring one party is simply unprecedented -- even in 1994 the gap was smaller.

There are some possible explanations since the questions are slightly different, like people dislike the GOP generically by like their own GOP rep (or flipped for Dem incumbents). Or it may just be an outlier.

Most available data suggest the Democrats would win bigger today than they did in November 2008, but at this point in the cycle data is sparse.