Gillibrand could be in serious trouble.

[cinyc]

I haven't seen any polls showing her in a dead heat; I only recall the Rasmussen polls that listed her leading by double digits.

They are coming... probably tomorrow.

Quinnipiac's poll will definitely be released at 6:30AM tomorrow morning.  I suspect SUSA's poll will be released some time after the late local newscasts have had the opportunity to put results on air tonight - but we'll see.

[cinyc]

SUSA: Gillibrand +1 (no specific numbers yet) - losing in all regions but NYC proper;
Schumer 54%- Sacrificial Lamb (R) 33% - Schumer is winning all but Upstate outside of Western NY

http://www.wgrz.com/news/local/story.aspx?storyid=87726&catid=37

Edit: Gillibrand leads DioGuardi 45%-44%; Other is at 8%.

http://www.surveyusa.com/client/PollReport.aspx?g=ddf78dd6-bef1-44e3-af71-225f2d2da637

[cinyc]

Another very Republican sample... but still... what the hell...

Indeed. 2008 exits had it at 50D-26R-25I while this is 44D-35R-20I. This is going to be a different electorate from 2008 of course, but I find it difficult to believe it's as different as SUSA seems to think.

Why is everyone, including Nate Silver, comparing turnout for Presidential election years to the recently released New York polls?  2006 would be better for comparison.  Those exits were 47D-27I-25R.   The Dem percentage is off by a whopping 3 points, which is perfectly reasonable given 2006 was a Democratic year and this year isn't.

[cinyc]

Another very Republican sample... but still... what the hell...

Indeed. 2008 exits had it at 50D-26R-25I while this is 44D-35R-20I. This is going to be a different electorate from 2008 of course, but I find it difficult to believe it's as different as SUSA seems to think.

Why is everyone, including Nate Silver, comparing turnout for Presidential election years to the recently released New York polls?  2006 would be better for comparison.  Those exits were 47D-27I-25R.   The Dem percentage is off by a whopping 3 points, which is perfectly reasonable given 2006 was a Democratic year and this year isn't.

To be honest I only went with 2008 numbers because I had them lying around and didn't want to look up 2006 ones.

I could believe 44D but I think 35R and 20I are a little fishy.

It depends how the question is worded.  In 2006, the exit polls asked "No matter how you voted today, do you usually think of yourself as a: [Democrat/Republican/Independent/Something Else]".  That methodology allows for more people to self-identify with a particular party than if the pollster asks something like: "Are you registered to vote as a [Democrat/Republican/Something Else/Not in a Party]."  During a wave, more "independents" are going to self-identify with the party riding the wave than declare themselves independents.

[cinyc]

Just FYI - 2004 was 45D, 29R, 26D

There was no exit poll in 2002.  

In 2000, it was 45D-28R-27I.   (Sen race; Hillary Clinton beat Lazio.)
In 1998, it was  41D-34R-25I.   (Gov race; Pataki won re-election in a landslide. )
In 1996, it was 42D-29R-28I.    (Pres race; Clinton won Reelection.)
In 1994, it was 39D-32R-29I.    (Gov race;Pataki squeaked by Cuomo.)
In 1992, it was 42D-32R-26I    (Presidential Race)/ 43D-32R-25I (Senate Race; D'Amato won reelection.)

Shifts happen.  The mid-term electorate is usually less Democratic and more Repubican than the Presidential year electorate.  Republicans ending up in the 30s in NY mid-term elections isn't unheard of.

[cinyc]

Quinnipiac has Gillibrand up by 6, 48-42. 

http://www.quinnipiac.edu/x1318.xml?ReleaseID=1505