Why are your numbers in NY-23 so different from Siena and Research 2000?
We're getting a lot of that today.
Two main reasons:
-We find, and were finding even before the Scozzafava dropout, that Hoffman has a much bigger lead with Republicans than those polls were showing. Research 2000 had Hoffman leading 41-34 with GOP voters. Siena had it at 50-29. We find it at 66-17 so far.
-We are anticipating a much more GOP friendly electorate than the other two polls. Siena found Obama's favorability in the district at 59%. Research 2000 had it at 50%. So far we find Obama's approval rating with likely voters at 39%. We're finding an electorate that voted 50-43 for John McCain in comparison to last fall's narrow Obama victory in the district. That drop off in Democratic voters is consistent with what we're finding in much of our 2009 polling but obviously we won't know if we were right or not until Tuesday.