| Obama | Romney |
| 100.0% probability of winning | 0.0% probability of winning |
| Mean of 334 electoral votes | Mean of 204 electoral votes |
My previous analysis showed President Barack Obama leading Romney by 327 to 211 electoral votes, and with a 99.9% probability of winning an election held then.
Since then, eleven new polls covering nine states have been released:
| start | end | sample | % | % | % | |||
|---|---|---|---|---|---|---|---|---|
| st | poll | date | date | size | MOE | O | R | diff |
| CA | Field Poll | 21-Jun | 02-Jul | 848 | 3.4 | 55 | 37 | O+18 |
| FL | Rasmussen | 09-Jul | 09-Jul | 500 | 4.5 | 45 | 46 | R+1 |
| ME | Critical Insights | 20-Jun | 25-Jun | 615 | 4.0 | 49 | 35 | O+14 |
| NM | WeAskAmerica | 09-Jul | 10-Jul | 1295 | 2.8 | 51 | 40 | O+11 |
| NC | PPP | 05-Jul | 08-Jul | 775 | 3.5 | 47 | 46 | O+1 |
| NC | PNA | 01-Jul | 08-Jul | 500 | 4.4 | 48 | 49 | R+1 |
| ND | Rasmussen | 10-Jul | 11-Jul | 400 | 5.0 | 36 | 51 | R+15 |
| PA | WeAskAmerica | 09-Jul | 10-Jul | 1227 | 2.8 | 47 | 40 | O+7 |
| VA | PPP | 05-Jul | 08-Jul | 647 | 3.9 | 50 | 42 | O+8 |
| WI | Marquette | 05-Jul | 08-Jul | 810 | 3.5 | 50.6 | 43.3 | O+7.3 |
| WI | PPP | 05-Jul | 08-Jul | 1057 | 3.3 | 50 | 44 | O+6 |
California (+18%) and Maine (+14%) for Obama and North Dakota (+15%) for Romney are strongholds.
I probably shouldn’t be surprised any longer on how solid New Mexico is for Obama at +11%. The polling history backs the new poll up:
Weeks after the unsuccessful recall of their Republican Governor, Wisconsin seems to be holding for Obama. Both new polls have Obama’s lead just outside the margin of error:
Romney goes up by +1 in this week’s Florida poll, but Obama still takes three of the four current polls for the state. Obama would be expected to win Florida right now with a 91% probability.
Obama and Romney split North Carolina this week at one poll apiece. In the past month of NC polls, Romney takes three and Obama takes two. The simulation analysis suggests that Romney would win the state (now) with a 59.5% probability.
Once again, a Pennsylvania poll puts Obama up. Obama has lead in both Pennsylvania polls taken over the past month. In fact, Obama has led in the past 14 consecutive PA polls—all the way back to early February.
Virginia gives Obama a +8% lead over Romney. But Romney lead by a smaller margin in a much bigger current poll. The two polls, taken together, have the race in a dead tie (Obama won 50,112 times, Romney, 49,888 times). It is difficult discern a solid trend. One could argue Obama maintains an advantage, and one can argue that Romney has turned the state around:
So, what would happen if the presidential election was held today?
A Monte Carlo analysis of state head-to-head polls, using the rules of the Electoral College suggests that President Barack Obama would almost certainly beat Mitt Romney.
Now, Obama would receive a mean of 334 (+7) electoral votes to Romney’s 204 (-7). Of the 100,000 simulated elections, Obama won 99,976 times and Romney won 24 times.
Here is the distribution of electoral votes [FAQ] from the simulations:
Ten most probable electoral vote outcomes for Obama:
- 328 electoral votes with a 3.43% probability
- 335 electoral votes with a 3.32% probability
- 341 electoral votes with a 3.27% probability
- 337 electoral votes with a 2.79% probability
- 329 electoral votes with a 2.69% probability
- 342 electoral votes with a 2.63% probability
- 336 electoral votes with a 2.57% probability
- 343 electoral votes with a 2.48% probability
- 323 electoral votes with a 2.46% probability
- 338 electoral votes with a 2.43% probability
After 100,000 simulations:
- Obama wins 100.0%, Romney wins 0.0%.
- Average (SE) EC votes for Obama: 333.9 (16.8)
- Average (SE) EC votes for Romney: 204.1 (16.8)
- Median (95% CI) EC votes for Obama: 335 (297, 364)
- Median (95% CI) EC votes for Romney: 203 (174, 241)
Each column of this table shows the electoral vote total aggregated by different criteria for the probability of winning a state (Safe=100%, Strong=90%+, Leans=60%+, Weak=50%+):
| Threshold | Safe | + Strong | + Leans | + Weak |
|---|---|---|---|---|
| Safe Obama | 155 | |||
| Strong Obama | 131 | 286 | ||
| Leans Obama | 42 | 42 | 328 | |
| Weak Obama | 13 | 13 | 13 | 341 |
| Weak Romney | 16 | 16 | 16 | 197 |
| Leans Romney | 11 | 11 | 181 | |
| Strong Romney | 135 | 170 | ||
| Safe Romney | 35 |
This table summarizes results by state. Click on the poll count to see the individual polls included for the state.
| 0 | 0 | EC | # | Total | % | % | Obama | Romney | |
|---|---|---|---|---|---|---|---|---|---|
| 8 | 4 | Votes | polls | Votes | Obama | Romney | % wins | % wins | |
| AL | 9 | 1 | 484 | 41.5 | 58.5 | 0.4 | 99.6 | ||
| AK | 3 | 0 | (0) | (100) | |||||
| AZ | 11 | 1 | 475 | 43.2 | 56.8 | 1.8 | 98.2 | ||
| AR | 6 | 1* | 679 | 36.8 | 63.2 | 0.0 | 100.0 | ||
| CA | 55 | 1 | 780 | 59.7 | 40.3 | 100.0 | 0.0 | ||
| CO | 9 | 2 | 1699 | 52.8 | 47.2 | 95.0 | 5.0 | ||
| CT | 7 | 1* | 1239 | 56.8 | 43.2 | 100.0 | 0.0 | ||
| DE | 3 | 0* | (100) | (0) | |||||
| DC | 3 | 0* | (100) | (0) | |||||
| FL | 29 | 4 | 4012 | 51.5 | 48.5 | 91.0 | 9.0 | ||
| GA | 16 | 1* | 404 | 43.3 | 56.7 | 2.9 | 97.1 | ||
| HI | 4 | 1* | 517 | 64.8 | 35.2 | 100.0 | 0.0 | ||
| ID | 4 | 0* | (0) | (100) | |||||
| IL | 20 | 1* | 546 | 61.5 | 38.5 | 100.0 | 0.0 | ||
| IN | 11 | 1* | 447 | 45.0 | 55.0 | 6.5 | 93.5 | ||
| IA | 6 | 1 | 967 | 50.6 | 49.4 | 60.3 | 39.7 | ||
| KS | 6 | 1* | 442 | 45.0 | 55.0 | 7.0 | 93.0 | ||
| KY | 8 | 1* | 528 | 45.5 | 54.5 | 6.8 | 93.2 | ||
| LA | 8 | 1* | 542 | 41.1 | 58.9 | 0.2 | 99.8 | ||
| ME | 2 | 2 | 931 | 58.4 | 41.6 | 100.0 | 0.0 | ||
| ME1 | 1 | 1* | 488 | 64.8 | 35.2 | 100.0 | 0.0 | ||
| ME2 | 1 | 1* | 421 | 53.7 | 46.3 | 85.5 | 14.5 | ||
| MD | 10 | 1* | 792 | 62.4 | 37.6 | 100.0 | 0.0 | ||
| MA | 11 | 1 | 848 | 58.5 | 41.5 | 100.0 | 0.0 | ||
| MI | 16 | 5 | 3419 | 51.2 | 48.8 | 83.9 | 16.1 | ||
| MN | 10 | 1* | 904 | 58.1 | 41.9 | 100.0 | 0.0 | ||
| MS | 6 | 1* | 717 | 40.0 | 60.0 | 0.0 | 100.0 | ||
| MO | 10 | 1* | 455 | 46.2 | 53.8 | 12.4 | 87.6 | ||
| MT | 3 | 1 | 372 | 45.2 | 54.8 | 9.2 | 90.8 | ||
| NE | 2 | 1* | 553 | 43.4 | 56.6 | 1.4 | 98.6 | ||
| NE1 | 1 | 1* | 389 | 45.5 | 54.5 | 10.7 | 89.3 | ||
| NE2 | 1 | 1* | 252 | 49.6 | 50.4 | 46.9 | 53.1 | ||
| NE3 | 1 | 1* | 284 | 35.9 | 64.1 | 0.0 | 100.0 | ||
| NV | 6 | 1* | 450 | 53.3 | 46.7 | 83.5 | 16.5 | ||
| NH | 4 | 3 | 1725 | 51.4 | 48.6 | 78.3 | 21.7 | ||
| NJ | 14 | 1* | 947 | 62.9 | 37.1 | 100.0 | 0.0 | ||
| NM | 5 | 1 | 1178 | 56.0 | 44.0 | 99.8 | 0.2 | ||
| NY | 29 | 1* | 758 | 62.8 | 37.2 | 100.0 | 0.0 | ||
| NC | 15 | 5 | 3107 | 49.7 | 50.3 | 40.5 | 59.5 | ||
| ND | 3 | 1 | 348 | 41.4 | 58.6 | 1.1 | 98.9 | ||
| OH | 18 | 2 | 1663 | 53.9 | 46.1 | 98.9 | 1.1 | ||
| OK | 7 | 1* | 448 | 30.4 | 69.6 | 0.0 | 100.0 | ||
| OR | 7 | 1 | 631 | 54.4 | 45.6 | 93.7 | 6.3 | ||
| PA | 20 | 2 | 2119 | 53.8 | 46.2 | 99.1 | 0.9 | ||
| RI | 4 | 1* | 495 | 59.4 | 40.6 | 99.8 | 0.2 | ||
| SC | 9 | 1* | 1833 | 51.7 | 48.3 | 85.4 | 14.6 | ||
| SD | 3 | 1* | 442 | 44.3 | 55.7 | 4.9 | 95.1 | ||
| TN | 11 | 1* | 654 | 46.0 | 54.0 | 7.7 | 92.3 | ||
| TX | 38 | 1* | 460 | 38.9 | 61.1 | 0.0 | 100.0 | ||
| UT | 6 | 1 | 1149 | 27.7 | 72.3 | 0.0 | 100.0 | ||
| VT | 3 | 1* | 528 | 67.8 | 32.2 | 100.0 | 0.0 | ||
| VA | 13 | 2 | 1606 | 50.0 | 50.0 | 50.1 | 49.9 | ||
| WA | 12 | 2 | 1386 | 56.2 | 43.8 | 99.9 | 0.1 | ||
| WV | 5 | 1* | 373 | 40.8 | 59.2 | 0.7 | 99.3 | ||
| WI | 10 | 3 | 2300 | 53.4 | 46.6 | 98.9 | 1.1 | ||
| WY | 3 | 0 | (0) | (100) |
* An older poll was used (i.e. no recent polls exist).
Details of the methods are given in the FAQ.
The most recent analysis in this match-up can be found from this page.
I know I said this last time, but you gotta fix South Carolina. It undermines the rest of your analysis to call SC a likely win for Obama based on a single outlier poll from December.
Speaking as someone who works in Democratic politics, I can tell you that it’s SO much closer than what you present here. Not exactly helpful to have people thinking we’ve got this in the bag…
Clara,
“I know I said this last time, but you gotta fix South Carolina.”
Sorry…I don’t “fix” things by making shit up.
“It undermines the rest of your analysis to call SC a likely win for Obama based on a single outlier poll from December.”
So be it. The rules are the rules, whether I like the outcome or not.
The most recent real poll in SC has a small lead for Obama (45% to 42%). It was a very large poll, and so the sampling error is small. Even so, S.C. only “goes for” Obama 85.4% of the time in the simulations.
“Speaking as someone who works in Democratic politics, I can tell you that it’s SO much closer than what you present here.”
Fine. Produce a poll and demonstrate it.
“Not exactly helpful to have people thinking we’ve got this in the bag…”
Anyone who thinks “we’ve got this in the bag” misunderstands the analysis, which is not a projection to November. I make that amply clear in the write-up.
Wow – it’s nice to see new polls in PA, VA, and WI with good numbers for the President. On top of that, FL and NC are close to a dead heat, they are both flippable to the Democrats if enough money and effort is spent there.
Clara @ # 1 is concerned that these poll analysis make it look like it’s “in the bag”, which might discourage enthusiastic participation. But we are rather committed partisons here, and don’t need to be motivated by fear – that’s what the other side does. Besides, we regularly get hammered by the trolls claiming that polls show that Romney is winning and has momentum, which is just a variation of the old “peer pressure” strategy of encouraging people to jump on whatever bandwagon they think will win. It’s nice to see some hard numbers, coupled with electoral vote strategy, which shows where the strong and weak areas are.
I am also dubious that S.C. will go blue, but if we can pull in VA, N.C., and FL, that leaves S.C. alone among a sea of blue on the East Coast.
rhp6033,
“…if we can pull in VA, N.C., and FL, that leaves S.C. alone among a sea of blue on the East Coast.”
VA and NC are both extremely close, but right now FL is Obama’s to lose (91% probability of winning based on current polls). On the other hand, the state, more than any other, has a habit of oscillating between the candidates across the election season.
So, with the polls definately trending blue in an electoral-college match-up, what strategies can we expect from the Republicans in the coming months:
(a) October Surprise – They’ve probably got some ad campaigns already put together which will pull some “issue” out of their nether-regions to air in September/October, probably funded by the incredible amount of money in pro-Republican PACS. It will be lies, of course.
(b) Voter suppression – already well into effect throughout states where Republicans hold a majority in the state legislature, governor’s office, or Secretary of State’s office. Expect more childish tricks close to the election, like telephone calls which claim that the voter isn’t properly registered and will be arrested if he tries to vote, or claiming that the voting date has changed.
(c) Denial of Service attacks – hacking attempts to shut down “Get Out the Vote” communications immediatly before and during the election.
(d) vote count rigging – like the 2004 election where the vote count differed substantially from the exit polls – but only in selected, crucial areas.
I’m sure they will come up with a few more, but these are must some of the strategies the Republicans have used over the past few national elections to make sure the outcome was different than othewise reliable polling predicted.
I understand and agree with Clara’s concern, but she misunderstands this data and Darryl has not been cheerleading the data. They are what they are.
Darryl and his data are clarity in a shit storm of spin and as close to unbiased in its presentation that I’ve encountered — pretty much anywhere.
Thanks again, Darryl.
Of course, we can expect Romney to get a bump in August after the Republican convention. Likewise, President Obama will get a bump after the Democratic convention. But both bumps will be temporary.
It will be interesting to see if the Republican convention in Tampa will have more of a lasting impact on Florida. Assuming, of course, that they don’t act like it’s an out-of-town playhouse for bad behavior.
That would be Zotz @6. Sigh.
Gee, I would love to see a picture of a bunch of Caddies sporting Romney bumper stickers all parked at a Tampa strip club.
Zotz @ 6,
Thanks! I do aim for an unbiased analysis, and unlike most of my other posts on this blog, I strive to be neutral in the presentation.
The apparent S.C. problem will resolve itself soon enough….
Clarity in a shit storm.
Someone should trademark that.
Thanks, Darryl.
rhp6033,
“Of course, we can expect Romney to get a bump in August after the Republican convention. Likewise, President Obama will get a bump after the Democratic convention. But both bumps will be temporary.”
Yep…you can see that from the 2008 analyses (from this post).
McCain got the bigger bump, but then a much bigger post-bump crash. That probably was a mix of the economy tanking and The Vetting of Sarah Palin.
Regarding VA, this NYT blog post is rather interesting. Obama was there today:
http://thecaucus.blogs.nytimes.....r-of-bain/
He wasn’t talking about Bain. He wasn’t talking about outsourcing.
The president, who is barnstorming around southern Virginia on a two-day campaign swing, did not touch the questions about Bain Capital in his stump speech. He even dropped a line he has used in recent appearances about how Mr. Romney’s buyout firm earned a reputation as a “pioneer” of outsourcing.
That’s scut work for the minions, apparently. Those minions are quickly backtracking from yesterday’s ‘felony’ idiocy. Fortune and CNN saw to that.
FYI, mostly.
On a related note, VA’s retiring Senator Webb won’t support Obama’s re-iterated tax plan.
http://hamptonroads.com/2012/0.....a-tax-plan
“That’s a no,” Webb told Reuters when asked if he’d vote for the plan.
Webb has called for taxes to be increased on dividends and capital gains instead, and his spokesman, Will Jenkins, told POLITICO: “Sen. Webb has consistently stated that he opposes raising taxes on ordinary earned income.”
Romney’s margin in the much larger current poll is only a widdle bit smaller than Obama’s margin in the older poll little more than half its size.
There isn’t much in VA to support a blue result. Obama’s hanging on by a thread but it doesn’t look like Webb will throw him much support, he can’t count on Kaine to say anything useful if Kaine wants to beat Allen, and McDonnell’s been a very effective surrogate.
@1
This sort of thing ☝is why, despite being a member of the left and working on Democratic and environmental campaigns, I rarely listen to people who “work in Democratic politics.”
The only way for Darryl to “fix” South Carolina is for more polling data to be released. You gotta dance with them that brung ya.
South Carolina will go to Romney, but that’s not a problem: the Milk Chocolate Messiah will get enought electoral votes to get another term.
No worries!
“…fix South Carolina…”
Kind of a bad choice of words, Clara. Makes you sound like the-end-justifies-the-means kind of gal. You’re not from King County Elections, are you?
Thanks Darryl! You put a LOT of work into these which helps to give me perspective.
New poll by Mason-Dixon covering FL:
http://www.nationaljournal.com.....a-20120714
Virtual tie. Obama up 1, quite similar to Darryl’s Rasmussen poll included above that has Romney up 1.
What’s particularly interesting about this poll are two things:
1. Gary Johnson got 2%. Another spoiler election?
2. It’s a likely voters poll, and Hispanic voters only went for Obama by 49-42.
I don’t think Obama is going to get the Hispanic vote bump out of the immigration policy shift that he is hoping he might get. Either that, or FL Hispanics are a whole lot more different from Hispanic elsewhere than I had previously thought.
Serial Conservative,
“Either that, or FL Hispanics are a whole lot more different from Hispanic elsewhere than I had previously thought.”
1. FL Hispanics are VERY different from Hispanics in other states.
a. Looking at the partisan makeup of all registered Florida Hispanic voters in 2012, is 55.5% Democratic and 44.5% Republican.
b. If you normalize (i.e. make it add up to 100%) the 49%-42% result to compare it to part a above, you get 53.8% Obama, 46.1% Romney.
2. The “Hispanic or Cuban” sample in the poll was only 105 people. The normalized poll results (part b above) are statistically not different from the partisan composition (part a above).
Finally, it is sketchy enough making any inference for the poll itself from that small of a subsample. Making inference to the U.S. Hispanic population projected four months from now is absurd!
@15,
Thanks for sharing your sickening racism with the rest of us.
Now please put out the burning cross on your lawn and get to your Klan meeting.
In the bag? No, but we could lose Florida and a few other swing states and still win. Mittens has to capture *all* of them to even have a prayer, and it’s obvious that the American people aren’t buying what he’s selling. This is before (1) the full extent of all of his crooked dealings becomes known and (2) he nominates a kook who makes Sarah Palin appear statesmanlike in a frantic effort to win back the base. Oh, and don’t forget – he has the McCain “not really a conservative” problem times ten.
i guess daryl knows more than rassmusen