I wrote a couple weeks ago a couple days after the third presidential debate about momentum and how Mitt Romney's momentum from the first debate was very much real and very much continuing in his direction. The margin for Barack Obama in swing state polls had slipped from an average of 4% to about 1%.As if to prove me wrong, that week Romney's momentum stopped. Ever since the third debate, there has been no movement towards Romney in the swing state polls: Ohio has held firm for Obama, and the Great Lakes states have all shown constant or improving poll trends for Obama (except for my weird Michigan numbers). Still, the fact that Obama is still only up 1% in the swing states means that it will still be a close election despite the fact that it is clear Obama has the advantage.
The following are the median poll spreads in the nine battleground states plus Romney's "extended map" states of Michigan, Pennsylvania and Minnesota, in order of how they would swing right now.
(I'll have more on the U.S. Senate races in a separate post).
Mitt Romney starts with a baseline of 191 electoral votes including Montana, Arizona and Georgia, states where Obama has a good shot at being within single digits. His likeliest battleground state pickup is ...
North Carolina (R+3) (based on 9 latest polls from PPP, Survey USA, Elon, Rasmussen, Civitas, Gravis, ARG, Marist, and Purple Strategies) - The Survey USA poll that showed Romney up by 5 was the most effectual poll of the week, and while PPP and Elon call the race tied, no polling outfit has shown a result with Obama on top since the previous Survey USA poll one month ago.
Florida (R+1) (based on 15 latest polls from WeAskAmerica, Gravis, Quinnipiac, Survey USA, CNN/ORC, PPP, Rasmussen, Susquehanna (VSS), Pharos, ARG, Fox News, Mason-Dixon, Marist, Suffolk and WaPo) - There were 9 polls this week, and Mitt Romney was at or above 50% in 6 of them. Florida still could go either way though; every single poll published in the last two weeks by firms that have house effects less than R+2 has shown the margin no more than 1% either way.
As of right now, this is all my model shows Romney getting, 235 electoral votes. But there are ties.
Colorado (tied) (based on 12 latest polls from PPP, Survey USA, CallFire, CNN/ORC, Rasmussen, WeAskAmerica, ARG, Purple Strategies, Marist, Gravis, Quinnipiac and the University of Denver) - Prior to today my model had shown Colorado in Romney's tally for 11 straight days, but in the past three days six polls were released, and five of them showed leads for Obama. The most meaningful may be the poll from Wednesday from the conservative firm WeAskAmerica that showed Obama up by three. The polling results from Colorado this week are what one could expect from a candidate with momentum, and that candidate is not Mitt Romney.
Virginia (tied) (based on 12 latest polls from WeAskAmerica, PPP, Roanoke College, Quinnipiac, Gravis, WaPo, Purple Strategies, Fox News, Rasmussen, ARG, Marist and Suffolk) - 7 new polls this week, and only one of them showed a Romney lead (that crazy Roanoke College poll). For Obama the best poll might have been the WeAskAmerica poll today that showed him up by 1 despite having an unusually rightward tilt shown by the senate race numbers (George Allen up on Tim Kaine by 0.8).
Even if you assume Romney stops the Obama momentum and wins these ties, he only has 257 electoral votes, 12 shy of victory. But it is not looking good for him anywhere else.
New Hampshire (O+2) (based on 8 latest polls from New England College, Gravis, Marist, PPP, Rasmussen, ARG, UNH and Suffolk) - Four new polls this week all showed Obama with leads, including today's poll from New England College showing Obama up 6, the largest lead a non-UNH poll has shown since a Marist poll on September 25th had him up 7. There were two days last week when my model showed Romney tied with Obama in New Hampshire, but now the polls are indicating that the momentum is with Obama.
Michigan (O+2) (based on 7 latest polls from PPP, Rasmussen, Epic-MRA, Detroit News, Baydoun/Foster, Gravis and MRG) - The biggest numerical anomaly. Despite having only one poll show the race any tighter than 3 points, the median Obama number (47, 48, 48, 48, 50, 52, 52) minus the median Romney number (42, 42, 45, 46, 46, 47, 47) gives a result of O+2.
Nevada (O+3) (based on 9 latest polls from Survey USA, CallFire, Gravis, Marist, PPP, Rasmussen, ARG, Suffolk, WeAskAmerica) - There were only three new polls this week, and they've shown the same results as every other Nevada poll I have in my database: Romney not leading, but Obama not pulling huge numbers. Nevada is a state with a tight race, but it is not a swing state.
Iowa (O+3) (based on 8 latest polls from Gravis, CallFire, Rasmussen, Marist, WeAskAmerica, PPP, ARG and the Des Moines Register) - There were six polls that came out in Iowa in the last three days and only one of them (Rasmussen) showed a Romney lead. Gravis is the Republican-leaning pollster showing the largest margin for Obama right now, 4 points. This is another state where Obama seems to have momentum.
Ohio (O+4) (based on 17 latest polls from WeAskAmerica, Rasmussen, CNN/ORC, U. of Cincinnati, PPP, Quinnipiac, Survey USA, Pharos, Gravis, Purple Strategies, ARG, SBSI, Suffolk, Fox News, Marist, Columbus Dispatch and WaPo) - Ohio gets the most attention from pollsters this year, but it is not swinging right now. 12 new polls came out this week, and Romney was only up in one of them, a Rasmussen poll on Monday that showed him up 2 points. On Friday Rasmussen showed the race tied. Obama doesn't have momentum here only because he's been leading by the same 3-5 point margin for the last two or three weeks.
Pennsylvania (O+4) (based on 10 latest polls from Franklin & Marshall, Philadelphia Inquirer, Rasmussen, Pharos, Gravis, Muhlenberg, Susquehanna, Quinnipiac, PPP and Siena) - There was only one poll released this week after Hurricane Sandy made landfall, and it used data from before the disaster. Still, prior to the hurricane Pennsylvania was like Nevada; a close-ish state but definitely not a swing state.
Wisconsin (O+4.5) (based on 8 latest polls from WeAskAmerica, St. Norbert, Marist, Rasmussen, Marquette, PPP, Mason-Dixon and Quinnipiac) - Six polls released in the past three days, and none of them showed Romney with a lead. Wisconsin had looked like a possible non-Ohio alternative route to 270 electoral votes for Mitt Romney at one time, but Wisconsin provided Obama with arguably his best results from any state this week. With maybe the exception of the Marist poll that only showed Obama up 3, every single poll this week was a pleasant surprise for Obama fans. First PPP showed Obama up 5, then Marquette showed him up 8, then Rasmussen couldn't even tweak the numbers enough to show a Romney lead, then the NPR station showed Obama up 9, and finally the Republican WeAskAmerica showed an amazingly large 7 point lead.
Minnesota (O+7) (based on 5 latest polls from PPP, Survey USA, Mason-Dixon, Rasmussen and St. Cloud St.) - I only started charting Minnesota because the Romney campaign has claimed it can win there based on Mason-Dixon's polling. But Mason-Dixon is the only firm that has shown Obama with less than 50%. This state has been in the bag for Obama since the time Minnesota let Mondale supporters into the electoral college.
There are other states where Romney will keep it in single digits (Oregon, possibly New Mexico, possibly Connecticut).
No comments:
Post a Comment