Potential 'popular vote nightmare' just three days away

>>> with the aftermath of sandy possibly affecting voter turnout in the traditionally blue northeast another storm forecast in the area election forecasters fear we may have a popular vote electoral college split. you've heard that possibility first floated by someone who is wearing gray sweaters all week on this show a while back. nate cohen is taking it a step further. he laid out a detailed scenario, vote turnout nightmare. let me ask you first about the popular vote issue. what we've talked about on this show before is the idea that obama is sort of underperforming in democratic states, he's going win anyway, he's had more of a falloff in blue states than elsewhere and that could drag down his national popular vote total. do you think there's potential then when you add in the hurricane and the logistical issues that come out of that that further depresses turnout in new jersey or something and affects the national popular vote ?

>> i do think the blue states would be responsible for obama loss in the popular vote if he was winning the electoral college . there are recent polls out of massachusetts, connecticut and california in the last 24 hours that shows obama losing nine to ten points even more than that in connecticut off of his 2008 performance and that would represent a pretty substantial decline that's greater than that elsewhere in the country. in terms of the hurricane it's hard to say. i'm not sure about what areas are affected. the atlantic coast of new jersey is not in a democratic area. staten island will vote for romney . suffolk county , new york is divided. i don't know how much turnout will fall on the city of new york . if i want does it would make a difference.

>> democrats are contesting on the shore, there's talk maybe the republican areas hardest hit there. i want to ask you another thing about the battleground. i took a look this morning and it seems there's basically a fire wall here that obama has on tuesday that involves three states, ohio , wisconsin , and nevada . if romney won basically everything else that's up for grabs in not these three states he would be short of 270. he would have 267. if you look at the clear averages in these three states these are persistent leads. ohio , wisconsin , nevada . are we at a point where the only hope for romney there's something wrong in the method oflg these state polls?

>> i think we're about at that point opinion he has a couple of days left for him to make a come back. there isn't a single poll, a single nonpartisan poll in ohio , wisconsin or nevada that shows romney in the lead. and, yes, you know the polls have been wrong before and it's conceivable that things can be off. you can ask about sharon engel if polls are right. the overwhelming majority of the time when someone holds a lead of this magnitude, this consistently, they wine up winning. and, you know, there are reasons to think the polls could be off. it's tough to talk to people on cell phones. assumptions about turn out are more important than ever. but, yeah, you're right i think the polls have to be wrong at this state level.

>> polls ask who are you going to vote for but there's data around polls who do you think will win and that's extraordinarily predictive. why is that question so predictive when we make people into mini anthropologists.

>> it's what people are hearing. people are hearing stuff about romney . the average person, you know, has an idea that the president is still pretty popular. that makes them think that, you know, most voters or swing voters will tilt a little more in obama 's direction. that has an effect on swing voters . not in a direct sense that they think obama will win but the balance of information that people are hearing out there is the same information that will sway swing voters . it makes sense it would be a pretty decent relationship between the actual results and what people think.

>> nate , i got to be honest. i care little for polls. let's gossip instead. i'm sure you've seen some of the blow back against the other nate for some of his predictions. i want to put up a tweet who said the degree to which people put their faith in nate silver goes against everything nate silver stands for. she's a fan. you guys have taken on something of a mythic quality like something out of oz or lord of the rings . when did you number crunchers become like rock stars ?

>> well, i don't think there's anything mythical or magical about it. the poll says obama will win in the swing states . nate silver says obama is ahead in the swing states and win the electoral college . the polls are usually right. the model acknowledges the polls could be wrong. that chance is at 15% based on the same sort of analysis i gave earlier that look there are a few examples when the polls are off this much. ken buck, sharon engel, hilary and obama in new hampshire but more the exception than rule therefore obama is the favorite. if obama wins i don't think that makes, you know, nate silver or myself or anyone else who stairs at the polls and says obama is favored, you know, being incredibly insightful. polls are right and pollsters get the credit. if obama loses i don't think nate silver should be panned out. i don't think either.

>> never pan you.

>> you could use your wizardry to answer this question. isn't it quite possible at this point in terms of electoral college when you're looking at the key battleground states the president has a lead in most them. isn't it possible the president will win the electoral college in a landslide.

>> i don't think it would constitute a landslide. but it's easy to see how grab gets to 303 or 332. by bush-gore standards it's a landslide. pretty middle of the road .

>> in speaking of that you have reviewed all of our map. we have made your own electoral college picks and you chose a favorite using your wizardry.

>> yeah. i thought that your map actually was the best.

>> hey.

>> sorry guys.

>> landslide prediction.

>> don't necessarily think this is the best map because i agree wildfirery state here. i don't think that obama is favored in florida and colorado and virginia strike me as coin flips. but there's a consistency that's, you know, evident in your map. obama is probably likely to win both colorado and virginia if he wins either of them and if obama wins florida he's probably going to win say virginia and that i think complicates some of the other maps and steve's instance i think he showed obama losing colorado but winning virginia and i think that's possible but unlikely. i think s.e. was right focus on new hampshire and iowa as potential areas of weakness for obama but gosh the polls are getting out of hand for romney in wisconsin . he's down by like seven or eight points in wisconsin .

>> i'm counting on cheese heads across the state to pull it out for romney those.

>> nate , i've been sticking up for your but after that if you guys are wrong next week you are discredited. thanks for joining us nate