Search Results for: 10,000

Obama	Romney
100.0% probability of winning	0.0% probability of winning
Mean of 342 electoral votes	Mean of 196 electoral votes

The previous analysis had President Barack Obama leading Gov. Mitt Romney by 344 to 194 electoral votes. Obama’s victory margin was “statistically significant,” as the analysis suggested he would win an election now with certainty.

There were a plethora of new polls released in the last couple of days. So now, after a Monte Carlo analysis using 100,000 simulated elections, informed entirely by state head-to-head polls, Obama wins 100,000 times and Romney wins 0 times. Obama receives (on average) 342 (-2) to Romney’s 196 (+2) electoral votes.

This is two consecutive analyses in which Mitt Romney has gained electoral votes. Is this the end of the Convention bump for Obama? Will Romney be able to keep adding a couple of electoral votes a day?

If so, with 34 days to go, he’d still lose by five—with only 264 electoral votes. In other words, to win, Romney needs to shake loose more than two electoral votes a day, on average, through election day.

That’s the analysis. The big picture in this race can be seen from the time trends from a series of elections simulated every seven days using polls from 03-Oct-2011 to 03-Oct-2012, and including polls from the preceding 21 days (FAQ).

[Read more…]

Obama	Romney
100.0% probability of winning	0.0% probability of winning
Mean of 344 electoral votes	Mean of 194 electoral votes

Even Newer Update: There is a real North Dakota poll that has just been released. It shows Romney leading in N.D. by 51% to 39%. The poll fails my inclusion criteria because it was conducted on behalf of the North Dakota Democratic-NPL State Party. But there you have it.

Newer Update: Okay…stuff should be fixed now.

Update: OOPS! As was pointed out in the comment thread, I accidentally “invented” a new poll that turned N.D. blue! Alas, it was an Ohio poll that got entered for the wrong state (and then entered again for Ohio). I’ll do a new analysis soon.

The analysis a few days ago showed President Barack Obama leading Governor Mitt Romney by, on average, 345 to 193 electoral votes. Since then, we’ve had a plethora of new polls released, and the result is a small gain for Obama Romney.

Here are the new polls:

		start	end	sample		%	%	%
st	poll	date	date	size	MOE	O	R	diff
AZ	Moore	25-Sep	26-Sep	500	4.0	42	46	R+4
IA	Iowa Poll	23-Sep	26-Sep	650	3.8	49	45	O+4
IA	VCR	23-Sep	25-Sep	500	4.4	46	47	R+1
ME	Rasmussen	25-Sep	25-Sep	500	4.5	52	40	O+12
MA	Boston Globe	21-Sep	27-Sep	502	4.4	57	30	O+27
MI	Gravis Marketing	21-Sep	22-Sep	804	3.3	50.0	46.2	O+3.8
MI	PPP	17-Sep	19-Sep	2386	2.0	51	42	O+9
NV	Marist	23-Sep	25-Sep	984	3.1	49	47	O+2
NH	ARG	25-Sep	27-Sep	600	4.0	50	45	O+5
NH	Marist	23-Sep	25-Sep	1012	3.1	51	44	O+7
NM	PPP	17-Sep	20-Sep	3111	1.8	52	43	O+9
NC	Marist	23-Sep	25-Sep	1035	3.1	48	46	O+2
OH	Columbus Dispatch	19-Sep	29-Sep	1662	2.2	51	42	O+9
OH	PPP	14-Sep	18-Sep	2890	1.8	50	44	O+6
PA	Muhlenberg	22-Sep	26-Sep	427	5.0	49	42	O+7
PA	PPP	17-Sep	18-Sep	2051	2.2	52	40	O+12
VA	ARG	24-Sep	27-Sep	600	4.0	49	47	O+2
VA	Suffolk	24-Sep	26-Sep	600	4.0	45.7	44.0	O+1.7
VA	PPP	17-Sep	19-Sep	2770	1.9	49	43	O+6
WA	Rasmussen	26-Sep	26-Sep	500	4.5	52	41	O+11

Romney continues to out poll Obama in Arizona. The three current polls, taken together, has Romney with a 95% probability of winning the state.

In Iowa, Obama leads by +4% in one poll and trails by -1% in another. Still, there are six Iowa polls taken in the past three weeks—Obama leads in four. Together they put Obama’s probability of taking the state (now) at 98%. Here is the last three months of polling:

Maine has Obama leading by +12% in the new poll. There is also a newly released Critical Insights poll that is a couple weeks old and has Obama up by +16. I didn’t see the poll until after the analyses were running; it’ll be included in subsequent analyses.

Obama hangs on to a small +3.8% in one new Michigan poll. Even so, Obama pulls over 50%. Another huge, but somewhat older, Michigan poll has Obama up by +9. With Obama leading in all eight current polls, by double digits in three of them, Obama is looking unbeatable in this state.

In Nevada, Obama’s lead is just +2% in the current poll. Romney hasn’t won any of the six current polls, and Obama would be expected to win the state now with a 95% probability.

Two new New Hampshire polls have Obama up by +5% and +7%. These are Obama’s strongest showing, now giving him three of the four current polls. Romney’s chances in the state have shrunk to about 10%.

Another North Carolina poll goes Obama’s way, but by only a +2% edge. Obama now has a streak of four consecutive polls in his favor, and he takes four of the five current polls for the state. Here is what the last three months of polling look like graphically:

Two more Ohio polls go to Obama, who leads in all 12 of the polls taken over the past three weeks. In 2008, Obama won Ohio by +4.6%. If the election was held now, we’d expect Obama to win by more than a +6% margin! Here’s the recent trend:

Pennsylvania shows, once again, that it isn’t a swing state. Obama’s +7% and +12% in the new polls fall in line with the rest of the ten current polls.

In Virginia, Obama has a puny +2% lead in two polls and a +6% lead in another. These polls make eleven taken in the past three weeks and Obama leads in every one of them. The analysis suggests Obama would win the hypothetical election now with certainty, even if by a smallish margin. The polling trend makes this apparent:

Last, but not least, we get a new Rasmussen poll in Washington, where Obama leads Romney by a solid +11%, and takes 52% of the vote. All three current polls are double digit leads for Obama…this new poll is the smallest lead.

Now, after a Monte Carlo analysis using 100,000 simulated elections, Obama receives (on average) 347 (+2) 344 (-1) to Romney’s 191 (-2) 194 (+1) electoral votes. Obama won all 100,000 of the simulated elections, suggesting he would certainly win a hypothetical election held now.

We can view the long term trend of this race from a series of elections simulated every seven days using polls from 30 Sep 2011 to 30 Sep 2012, and including polls from the preceding 21 days (FAQ).

Looking at the media electoral vote line (purple), we see that Obama’s position is stronger than at any time in the past year tied with his strongest position over the past year (median of 347 electoral votes).

Here is the distribution of electoral votes [FAQ] from the simulations:
[Read more…]

Obama	Romney
100.0% probability	0.0% probability
345 electoral votes	193 electoral votes

The previous analysis showed President Barack Obama leading Governor Mitt Romney by an average of 341 to 197 electoral votes. The results were strong enough and certain enough that we would expect Obama to be the certain winner in an election held now.

Over the past two days, 14 new polls have been released, covering 12 states.

The net result is that, after 100,000 simulated elections, Obama wins 100,000 times and Romney wins 0 times. Obama receives (on average) 345 to Romney’s 193 electoral votes. That is a gain of +4 EVs based on the new polls and the “aging out” of some older polls. We’d have to say the the evidence is quite strong that Obama would achieve victory in a hypothetical election held today.

Here are the new polls, followed by a discussion of some of the changes over the past couple of days:

		start	end	sample		%	%	%
st	poll	date	date	size	MOE	O	R	diff
AZ	Rasmussen	25-Sep	25-Sep	500	4.5	42	52	R+10
CO	Gravis Marketing	21-Sep	22-Sep	765	3.4	50.2	45.5	O+4.7
CT	PPP	24-Sep	26-Sep	801	3.5	54	41	O+13
FL	Insider Advantage	24-Sep	24-Sep	540	4.1	49	46	O+3
FL	Quinnipiac	18-Sep	24-Sep	1162	3.0	53	44	O+9
IN	Bellweather Research	19-Sep	20-Sep	800	3.5	40	52	R+12
IA	PPP	24-Sep	26-Sep	754	3.6	51	44	O+7
MD	Gonzales Res	17-Sep	23-Sep	813	3.5	55	36	O+19
MA	Rasmussen	24-Sep	24-Sep	500	4.5	55	40	O+15
MO	Chilenski Strategies	20-Sep	20-Sep	817	3.4	44.3	50.2	R+5.9
OH	Quinnipiac	18-Sep	24-Sep	1162	3.0	53	43	O+10
PA	Quinnipiac	18-Sep	24-Sep	1162	3.0	54	42	O+12
PA	Franklin & Marshall	18-Sep	23-Sep	392	4.9	52	43	O+9
WA	Gravis Marketing	21-Sep	22-Sep	625	4.6	56.4	38.6	O+17.8

There are a couple of bright spots for Romney, including a poll from Arizona that has Romney up by a +10%. Romney’s lead in the previous AZ poll was only +3%. The three current polls, taken together bump Romney’s chances of taking the state from 97.5% to 99.3% in an election now.

In Colorado, Obama has a +4.7% lead in the latest poll. Obama gets a slight bump from 95% to 96% probability of taking the state (for now) from the poll.

Two more Florida polls go to Obama. He gets a +3% in one and a +9% in the other. Taken together the 15 current polls suggest a 99% chance of Obama taking the state—up from a 93% probability without these two new polls.

Indiana has Romney up by a respectable +12% over Obama. Indiana barely went for Obama in 2008. It seems unlikely the state will repeat that in 2012.

Another Iowa poll goes Obama’s way, this time by +7%. Romney has only led in one of the four current polls taken over the past three weeks. Taken together, Obama gets a slight bump from 95% to 98% of taking the state now.

The other bright spot for Romney is Missouri, where Romney leads in a new poll by +5.9%. Together, the two current polls bump Romney’s chances from 68% to 90% of taking the state now.

Another Ohio poll goes to Obama, this one by double digits. This gives us 12 polls in Ohio over the past three weeks and every one of them goes to Obama. As a result, Obama won all 100,000 simulated elections in that state.

Two new Pennsylvania polls both go to Obama. One by +12% and the other by +9%. The eight current polls all go to Obama. The give Obama, in aggregate, a +9% lead over Romney, and suggest Obama would take the state with certainty in an election right now.

A new poll for Washington state has Obama leading Romney by +17.8%. The five current polls all give Obama double digit leads in the state.

Here is the distribution of electoral votes [FAQ] from the simulations:
[Read more…]

Obama	Romney
100.0% probability of winning	0.0% probability of winning
Mean of 341 electoral votes	Mean of 197 electoral votes

Beginning this analysis, I’ve narrowed the “current polling window.” The old window included all polls taken within the past month. That criterion resulted in some states with many “current” polls, including some pre-conventions polls.

The new criterion is three weeks, so that almost all the polling occurs after the conventions. You can expect the window to shrink to two weeks sometime in October.

There are two effects from shrinking the window. First, the number of individuals polled goes down for some states. With fewer respondents, we have less evidence and, therefore, uncertainty increases (all else being equal).

The other effect is that there is less smoothing of the results. That is, the results become more indicative of trends.

Taken together, we might expect that Governor Mitt Romney’s chances improve through the increase in uncertainty. We can also expect Romney’s chances to decrease as a result of a recent poll surge for Obama. As it happens, the latter has a much stronger effect. The net effect is that Romney takes a beating….

The previous analysis had Obama leading Romney by 331 to 207 electoral votes, and the analysis suggested that Obama would almost certainly win an election held now.

Nineteen new polls covering 13 states have appeared in the past two days. Additionally, I’ve found four older polls (one each in AL, CO, OH, and FL) by comparing my database against Samuel Minter’s database.

Here are the new polls:

		start	end	sample		%	%	%
st	poll	date	date	size	MOE	O	R	diff
AR	Talk Business-Hendrix College	17-Sep	17-Sep	2228	2.0	34.5	55.5	R+21.0
CO	PPP	20-Sep	23-Sep	940	3.2	51	45	O+6
FL	PPP	20-Sep	23-Sep	861	3.3	50	46	O+4
FL	WA Post	19-Sep	23-Sep	769	4.5	51	47	O+4
FL	ARG	20-Sep	22-Sep	600	4.0	50	45	O+5
IA	ARG	20-Sep	23-Sep	600	4.0	51	44	O+7
MI	Rasmussen	20-Sep	20-Sep	500	4.5	54	42	O+12
MN	Mason-Dixon	17-Sep	19-Sep	800	3.5	48	40	O+8
MT	Mason-Dixon	17-Sep	19-Sep	625	4.0	42	51	R+9
NV	ARG	20-Sep	23-Sep	600	4.0	51	44	O+7
NV	NV Retailers	19-Sep	20-Sep	500	4.4	46	46	tie
NV	PPP	18-Sep	20-Sep	501	4.4	52	43	O+9
NJ	Monmouth	19-Sep	23-Sep	613	4.0	52	37	O+15
NC	Civitas	18-Sep	19-Sep	600	4.0	49	45	O+4
OH	Gravis Marketing	21-Sep	22-Sep	594	4.3	45.3	44.3	O+1.0
OH	WA Post	19-Sep	23-Sep	759	4.5	52	44	O+8
PA	Susquehanna	18-Sep	20-Sep	800	3.5	47	45	O+2
PA	Mercyhurst U	12-Sep	20-Sep	522	4.3	48	40	O+8
WI	WeAskAmerica	20-Sep	23-Sep	1238	2.8	52.5	41.0	O+11.5

With this new poll, seven of eight current Colorado polls go to Obama suggesting he would win an election now with about a 95% probability. One can sense from the last three months of polling that Obama is gaining slowly in the state:

Florida has Obama up by +4% in two new polls and +5 in another. Obama has now led in five consecutive polls in the state. That trend we can almost discern in Colorado is even more apparent in Florida:

Romney took the previous Iowa poll, but this new one goes +7% for Obama. The three current polls, take together, suggest Obama has a 95% chance of taking the state right now.

Another solid Michigan poll for Obama leaves him with a certain win in the state (at least, for now).

Three new Nevada polls have Obama up by +7%, +9%, and +0% (i.e. tied). It looks like Obama is regaining the lead he held there six months ago:

The Civitas Institute is a right wing think tank in North Carolina, but they release all their polls. This one has Obama leading by +4%. Obama leads in four of the six current polls and the last three in a row. Still, the weight of evidence goes very slightly to Romney. The trend seems to be in Obama’s favor:

Two new Ohio polls favor Obama, one by a whisper, the other by +8%. Looking at the three month polling trend in the state, the past few weeks have not been kind to Romney:

Obama gets his best and his worst poll in recent months in today’s Pennsylvania collection. Still, it is hard to argue that Obama’s lead isn’t solid:

Finally, a double-digit lead for Obama in Wisconsin caps off a very favorable collection of recent polls in the state for the President. Whatever momentum Romney had in the state in August seems to have almost entirely vanished:

Now after 100,000 simulated elections, Obama wins 100,000 times and Romney wins 0 times. Obama receives (on average) 341 (+10) to Romney’s 197 (-10) electoral votes. In a hypothetical election held now, Obama would have a 100.0% probability of winning.

[Read more…]

Obama	Romney
100.0% probability of winning	0.0% probability of winning
Mean of 331 electoral votes	Mean of 207 electoral votes

I’ve finally succumbed to the pressure of “doing something about South Carolina.” The last straw was Goldy harassing me over the phone for the umpteenth time this past week. So, I caved and changed the inclusion criteria for states lacking “current polls.” Note that I did not simply drop the vexing S.C. poll, but the new inclusion criteria means that when there are no current polls, multiple old polls will be pooled.

Here is what I have been doing: If a state does not have a “current poll”, I use the single most recent poll available. Right now, the definition of “current poll” is any poll taken over the past month (this “window” will shrink as the election approaches and the pace of polling increases).

That simple rule worked pretty well in 2008. But a single, large poll in South Carolina has plagued these analyses since late last year. The poll is perfectly valid, and might even be correct in giving Obama an 85% probability of winning the state. But it is quite old, and there are other slightly older polls in the state that contradict it. What’s an analyst to do?

I modified a suggestion from Richard Pope and used an inclusion window for older polls, the length of which depends on how old the most recent poll is. Here his the new rule set:

1. Use only polls taken in the “current polling window” (one month, right now)
2. If the most recent poll is outside the current polling window and less than 3 months old, use that poll and any others taken over the prior two weeks
3. if the most recent poll is 3-6 months old, use that polls and any others from the prior month
4. if the most recent poll is 6-9 months old, use that poll and any other from the prior three months
5. if the most recent poll is 9-12 months old, use that poll and any other from the prior six months
6. if the most recent poll is more than a year old, use that poll and any other from the prior year

There are several disadvantages of this new rule set: First, it is complicated. As you know, I strive to minimize arbitrary assumptions in the methods, but now I’ve gone and added a complex, arbitrary rule. Yuck. Another disadvantage is that the new rules will tend to overestimate the winning certainty for states with multiple old polls included in an analysis. On the other hand, a state that hasn’t been polled in a long time probably hasn’t been polled much, so most of the time the single most recent poll will still be used. South Carolina is an exception. There were a bunch of polls taken late last year. I believe this happened because there was “piggyback” polling of the general election by pollsters covering the contentious G.O.P. primary in that state.

Okay…so let’s get to it. Just three days ago the analysis had President Barack Obama leading Gov. Mitt Romney by 336 to 202 electoral votes. The results suggested that, if an election was held now, Obama would almost certainly win.

In the three days since my previous analysis, there have been some 19 new polls (plus, I am including an old poll, previously overlooked). Here are the polls:

		start	end	sample		%	%	%
st	poll	date	date	size	MOE	O	R	diff
AZ	Purple Poll	15-Sep	19-Sep	600	4.0	45	48	R+3
CA	PPIC	09-Sep	16-Sep	995	4.4	53	39	O+14
CO	Purple Poll	15-Sep	19-Sep	600	4.0	48	45	O+3
CO	Marist	16-Sep	18-Sep	971	3.1	50	45	O+5
FL	Mason-Dixon	17-Sep	19-Sep	800	3.5	48	47	O+1
FL	Purple Poll	15-Sep	19-Sep	600	4.0	47	48	R+1
GA	InsiderAdvangage	18-Sep	18-Sep	483	4.5	35	56	R+21
IA	Marist	16-Sep	18-Sep	898	3.3	50	42	O+8
NE	Wiese Res	17-Sep	20-Sep	800	3.5	40	51	R+11
NE2	Wiese Res	17-Sep	20-Sep	400	4.9	44	44	tie
NC	Purple Poll	15-Sep	19-Sep	600	4.0	48	46	O+2
NC	High Point	08-Sep	18-Sep	448	4.7	48	44	O+4
OH	Purple Poll	15-Sep	19-Sep	600	4.0	48	44	O+4
OH	FOX News	16-Sep	18-Sep	1009	3.0	49	42	O+7
OH	Ohio Poll	13-Sep	18-Sep	861	3.3	51	45	O+6
OH	Gravis Marketing	07-Sep	08-Sep	1548	2.7	47.3	43.2	O+4.1
PA	Rasmussen	19-Sep	19-Sep	500	4.5	51	39	O+12
SD	Neilson Brothers	29-Aug	06-Sep	512	4.3	38.7	53.9	R+15.2
VA	Purple Poll	15-Sep	19-Sep	600	4.0	46	43	O+3
WI	Marist	16-Sep	18-Sep	968	3.2	50	45	O+5

The race seems to tighten in Arizona where Romney just squeaks by with a +3%. Also, this is the first time since early June that Romney has been under 50%.

On the other hand, California tightens as well. Obama gets +14% where he used to pull in the low +20%.

Another pair of Colorado polls go to Obama, who leads in seven of the eight current polls, and seems to have a 91% chance of taking the state (in an election held now).

Florida gives each candidate a poll, by +1%. From twelve polls pooled over the past month, the ~11,000 responses go to Obama 50.8% of the time and to Romney 49.2% of the time. That translates into an 88% probability that Obama would win the state now.

The good news for Romney is that Georgia shows the strongest result for him this year.

Iowa has a strong +8% result for Obama. This poll was taken slightly before the Rasmussen poll I included last analysis that had Romney up by +3%. Even so, with Obama taking two of the three current polls, Obama would seem to have an 89% chance of taking the state right now.

We finally have some polling in Nebraska, where Romney bests Obama by +11%. That the good news for Romney. Recall that Nebraska assigns one elector to the winner of each CD, and the overall state winner gets the other two electors. The bad news for Romney is that Nebraska’s second congressional district polls at a 44% tie. Obama won NE-2 in 2008 and may do it again! (The polling report mentions that Romney has a solid lead in the other two CD’s, but the numbers are not given; instead, an older poll is used for those two CDs. Romney leads in both.)

Two new North Carolina polls go to Obama by small margins. Even so, Romney has led in five of nine current polls (and there was a tie). Overall, the pooled polls favor Romney with an 83% probability of winning an election held now.

With the three new Ohio polls, Obama has a streak of eight consecutive poll leads in the state. Obama won 99% of the simulated elections in the state.

A double-digit lead for Obama in the newest Pennsylvania poll. Obama has won all four of the state’s current polls.

South Dakota gives Romney his first double digit SC lead of the year.

Like Ohio, the new Virginia poll gives Obama a streak of eight consecutive poll leads in the state. But Obama only won 95% of the simulated elections in Virginia.

The new Wisconsin poll means Obama leads in all five current polls, and gives Obama a very high probability of taking the state (in a hypothetical election held now).

After a Monte Carlo analysis with 100,000 simulated elections, Obama wins 100,000 times and Romney wins 0 times. Obama receives (on average) 331 to Romney’s 207 electoral votes. Obama has a 100.0% probability of winning and Romney has a 0.0% probability of winning. And that is with South Carolina going to Romney 95% of the time.

Using the new “old poll inclusion criteria,” here is the picture of the race over time:

Hmmm…it looks pretty much like the old one that used the simple “single most recent” rule.

[Read more…]

Obama	Romney
97.9% probability of winning	2.1% probability of winning
Mean of 306 electoral votes	Mean of 232 electoral votes

Now that both conventions have come and gone and the post-convention polls are coming out, it’s a good time to assess the “score” in the 2012 Presidential Election.

My previous Monte Carlo analysis of the race was posted ten days ago (between the two conventions) and showed President Barack Obama leading Governor Mitt Romney by a mean of 305 to 233 electoral votes. An election held then would be expected to end in Obama’s favor with a 97.2% probability. Romney had a 2.8% probability of winning.

Thirty two new state head-to-head polls covering 18 states have been released since then:

		start	end	sample		%	%	%
st	poll	date	date	size	MOE	O	R	diff
AZ	PPP	07-Sep	09-Sep	993	3.1	44	53	R+9
CA	SurveyUSA	09-Sep	11-Sep	524	4.3	57.4	34.9	O+22.5
CO	Keating	10-Sep	11-Sep	503	4.4	49	44	O+5
CO	PPP	31-Aug	02-Sep	1001	3.1	49	46	O+3
FL	Rasmussen	12-Sep	12-Sep	500	4.5	48	46	O+2
FL	Marist	09-Sep	11-Sep	980	3.1	49	44	O+5
FL	Mclaughlin & Assoc	09-Sep	10-Sep	600	4.0	47	50	R+3
FL	SurveyUSA	07-Sep	09-Sep	596	4.1	48.2	44.5	O+3.7
FL	Gravis Marketing	02-Sep	02-Sep	1288	2.7	46.7	48.0	R+1.3
IL	WeAskAmerica	05-Sep	05-Sep	1382	2.8	54	37	O+17
MI	EPIC/MRA	08-Sep	11-Sep	600	4.5	47	37	O+10
MI	PPP	31-Aug	02-Sep	1001	3.4	51	44	O+7
MN	PPP	10-Sep	11-Sep	824	3.4	51	44	O+7
MN	SurveyUSA	06-Sep	09-Sep	551	4.3	49.6	40.4	O+9.3
MO	Rasmussen	11-Sep	11-Sep	500	4.5	45	48	R+3
MT	PPP	10-Sep	11-Sep	656	3.8	45	50	R+5
NH	U NH	04-Sep	10-Sep	592	4.0	45	40	O+5
NJ	Quinnipiac	27-Aug	02-Sep	1471	2.6	51	44	O+7
NM	PPP	07-Sep	09-Sep	1122	2.9	53	42	O+11
NM	Research & Polling	03-Sep	06-Sep	667	3.8	45	40	O+5
NY	Quinnipiac	04-Sep	09-Sep	1486	2.5	62	34	O+28
NC	PPP	07-Sep	09-Sep	1087	3.0	49	48	O+1
NC	SurveyUSA	04-Sep	06-Sep	500	4.5	43	53	R+10
OH	Rasmussen	12-Sep	12-Sep	500	4.5	47	46	O+1
OH	PPP	07-Sep	09-Sep	1072	3.0	50	45	O+5
OH	Gravis Marketing	07-Sep	08-Sep	1548	2.7	47.3	43.2	O+4.1
OH	Gravis Marketing	02-Sep	02-Sep	1381	2.9	43.7	46.8	R+3.1
TX	WPA	09-Sep	11-Sep	1000	3.1	40	55	R+15
VA	Marist	09-Sep	11-Sep	996	3.1	49	44	O+5
VA	Gravis Marketing	08-Sep	09-Sep	2238	2.2	44.0	49.4	R+5.4
WA	PPP	07-Sep	09-Sep	563	—	53	42	O+11
WA	SurveyUSA	07-Sep	09-Sep	524	4.4	54.4	37.6	O+16.8

I’ll only discuss a few states.

In Colorado, Obama takes both new polls, but by modest margins. Colorado seems to have settled for Obama by a small margin:

Five new polls from Florida ought to be enough to tell us who’s ahead. But, no: Obama takes three and Romney takes two. This gives us a total of 12 “current polls” (i.e. those taken in the past month), and in aggregate, Romney comes up with an advantage. In an election now, Romney would take Florida with about an 80% probability.

Obama takes both new Michigan polls by fairly nice margins. There are a total of seven “current polls” for the state—all but two go to Obama—and Obama won 94% of the simulated elections in the state.

Obama seems to have Minnesota pretty well locked up, with the new polls.

The new Missouri poll give Romney a small +3% lead over Obama. We have four other “current” Missouri polls from August. Romney takes all but one. Unless there is some kind of post convention second thoughts coming from the state’s voters, Romney looks likely to win the state.

Romney takes the latest Montana poll by just +5%. The previous poll, from mid-August had Romney up by +17%.

Obama turns in a weak +5% performance In New Hampshire, suggesting he only has an 81% chance of taking the state now.

In New Mexico, Obama is as strong as anywhere. The two new poll have him at +11% and +5%. The older current poll from mid-August is a +14%.

Two new North Carolina polls offer dissimilar pictures. The PPP poll offers Obama a +1% edge over Romney, but the Survey USA poll done for the Civitas Institute (a very right-wing state “think tank”) has Romney up by +10%. The Survey USA poll was taken during the convention and the PPP poll was taken after, so one could argue that there is a post-convention bounce for Obama. But the evidence to support that is vanishingly thin. The polling for the past month gives one poll to Obama, two ties, and four to Romney. Overall, the evidence still supports a good Romney advantage in the state. Here is all the polling in a nicepicture:

Four new Ohio polls split three to one for Obama. The three Obama polls are all post-convention. Romney’s poll is pre-convention. The nine polls taken over the past month go five for Obama, two for Romney and two ties. The evidence gives Obama a slightly stronger advantage in Ohio than Romney has in North Carolina.

The Virginia split, giving Obama a +5% in one and Romney a +5.4% in an even larger poll. The other three current polls go one for Obama, one for Romney and one tie. The state is still really, really, close, although Romney wins about 68% of the simulated elections.

Two Washington polls offer no surprises: Obama dominates with a double-digit lead. End of story.

With all those new polls, a Monte Carlo analysis of 100,000 simulated elections, gives Obama 97,854 wins and Romney 2,146 wins (including the 239 ties). Obama receives (on average) 306 (+1) to Romney’s 232 (-1) electoral votes. These results suggest that, for an election held now, Obama would have a 97.9% (+0.5%) probability of winning and Romney, a 2.1% (-0.5%) probability of winning.

In other words, little has changed in the past ten days. Obama still maintains a solid lead in the electoral college vote. It is interesting to note that the popular vote (as assessed by national polls) seems to be swinging in Obama’s direction now as well.

The long term trends in this race can be seen from a series of elections simulated every seven days using polls from 13 Sep 2011 to 13 Sep 2012, and including a one month polling window (FAQ).

Compared to 2008, this is not a very dynamic race.

Here is the distribution of electoral votes [FAQ] from the simulations:
[Read more…]

Obama	Romney
97.2% probability of winning	2.8% probability of winning
Mean of 305 electoral votes	Mean of 233 electoral votes

The previous analysis showed President Barack Obama leading Mitt Romney by 302 to 236 electoral votes. If the election was held then, Obama would be expected to win with a 96.9% probability, and Romney with a 3.1% probability.

Only seven new polls covering five states have been released since then. But, on the heels of the Republican convention and the start of the Democratic convention, this seems like a good place to take stock of the race. First, the new polls:

		start	end	sample		%	%	%
st	poll	date	date	size	MOE	O	R	diff
FL	PPP	31-Aug	02-Sep	1548	2.5	48	47	O+1
MI	EPIC/MRA	28-Aug	28-Aug	1200	2.6	49	46	O+3
MO	PPP	28-Aug	29-Aug	621	3.9	41	53	R+12
NC	PPP	31-Aug	02-Sep	1012	3.1	48	48	tie
NC	SurveyUSA	26-Aug	30-Aug	543	4.3	43	46	R+3
NC	Elon U	25-Aug	30-Aug	1089	3.0	43	47	R+4
WV	R.L. Repass	22-Aug	25-Aug	401	4.9	38	52	R+14

Close as ever, the new Florida poll has Obama leading Romney by +1%. This poll has been taken as evidence for a lack of a strong convention (or Ryan) bump. But, who knows in Florida, the next poll may go strongly for Romney.

Michigan puts Obama over Romney by a rather weak +3%. The longer term trend suggests a much tighter race than earlier in the year. But Obama still seems to have the advantage:

A surprise poll comes out of Missouri, where Romney leads Obama by +12%. This double digit lead is a wider gap than we have seen recently. But the real surprise is the increasing variability seen among pollsters. Just two polls ago, Rasmussen found Obama leading Romney by +1%:

Three North Carolina polls mostly favor Romney. The two tie in the most recent poll. The two other polls give Romney a +3% and +4% advantage. With six “current” polls in the state, the weight of evidence is that Romney would win the state (now) with a 69% probability:

In West Virginia, Romney leads Obama by a +14%. Believe it or not, this race has tightened up since the last couple of polls.

Now, after 100,000 simulated elections, Obama wins 97,197 times and Romney wins 2,803 times (including the 330 ties). Obama receives (on average) 305 (+3) to Romney’s 233 (-3) electoral votes. Obama has a 97.2% (+0.3%) probability of winning and Romney has a 2.8% (-0.3%) probability of winning an election held now. The difference from the previous analysis is really too small to “make” anything over.

Now, let’s consider the longer term trends in this race. I’ve done as series of Monte Carlo analyses, conducted every seven days using all the polls from the last year. For each simulation, I follow the same rules of including only polls from the past month (and the most recent poll before then if there are none) (FAQ). The following graph shows how the election is “scored” over time. The middle (magenta) line is the median number of electoral votes for Obama. When that line dips below the dashed line, Romney wins. Above it, Obama wins.

The pair of tan lines show the interval of electoral college totals within which 75% of the results for Obama were found. And the outer green lines show the interval of electoral college totals within which 95% of the results for Obama were found.

Romney was at his best in the 4th quarter of 2011. Even then, his probability of winning the election was never greater than about 30%. Beginning in late January, Romney slipped into zero percent territory, and has been there until recently. He has now recovered to same place he was in late January.

The general election has just started, so we are at a point with much potential for big change in the standings. In 2008, we saw McCain begin to do, only to have Obama run away with it in the end.

[Read more…]

Obama	Romney
96.9% probability of winning	3.1% probability of winning
Mean of 302 electoral votes	Mean of 236 electoral votes

The previous analysis showed G.O.P. presidential candidate Mitt Romney edging up to almost a 1% probability of winning an election held now. Romney lagged in expected electoral votes to President Barack Obama by 230 to 308.

Since then, 15 new polls covering 12 states have been released.

		start	end	sample		%	%	%
st	poll	date	date	size	MOE	O	R	diff
CO	Keating	21-Aug	22-Aug	500	4.4	48	44	O+4
CT	Quinnipiac	22-Aug	26-Aug	1472	2.6	52	45	O+7
CT	PPP	22-Aug	23-Aug	881	—	53	40	O+13
FL	CNN/TIME	22-Aug	26-Aug	895	3.5	50	46	O+4
IA	PPP	23-Aug	26-Aug	1244	2.8	47	45	O+2
MI	Mitchell	23-Aug	23-Aug	1277	2.7	46.6	46.7	R+0.1
MO	Mason-Dixon	22-Aug	23-Aug	625	4.0	43	50	R+7
MO	Rasmussen	22-Aug	22-Aug	500	4.5	47	46	O+1
NV	PPP	23-Aug	26-Aug	831	3.4	50	47	O+3
NJ	Eagleton-Rutgers	23-Aug	25-Aug	710	3.5	51	37	O+14
NC	CNN/Time	22-Aug	26-Aug	766	3.5	47	48	R+1
NC	SurveyUSA	18-Aug	23-Aug	540	4.3	43	43	tie
OH	Columbus Dispatch	15-Aug	25-Aug	1758	2.1	45	45	tie
PA	Philadelphia Inquirer	21-Aug	23-Aug	601	4.0	51	42	O+9
VA	Rasmussen	23-Aug	23-Aug	500	4.5	47	47	tie

Obama takes the latest Colorado poll by +4% over Romney, and he leads in four of the five current polls for the state.

Two polls in Connecticut both go to Obama. The Quinnipiac poll has Obama up by a modest +7%. A slightly older PPP poll has Obama up by +13%.

The latest Florida poll has Obama leading Romney by +4%. The current Florida polls go 4 to 2 for Romney, and Romney is given a 97% chance of winning the state right now.

Obama is up by +2% in the new Iowa poll. The candidates split the two current polls, but the poll that has Obama up is the much larger of the two:

In Michigan, Romney leads Obama by a weak +0.1%. The candidates split the four current polls, but the weight of the evidence has Obama up slightly with a 57% probability of winning an election held now.

The candidates split the two Missouri polls, with Romney up by +7% in one and Obama up by +1% in another. Obama has only led in this one poll out of the 6 current Missouri polls:

Nevada has Obama up by a slender +3% over Romney, slightly beating the +2% he had in the other current poll:

Little surprise that New Jersey has Obama up by a double-digit lead (+14%) over Romney.

In North Carolina, Romney has a +1% lead over Obama in one poll and the candidates are tied in another. The weight of evidence in the five current polls has Romney up by the slightest margin and a 56% probability of taking the state in an election held now:

Ohio is a tie at 45% each in the new poll. But Obama has led in three of the current six polls (with two ties), so the weight of evidence gives him an 87% probability of taking the state right now:

Another Pennsylvania poll gives Obama a +9% lead over Romney. Obama take all three of the current polls and would be expected to win an election now with a 99% probability.

Virginia is all tied up at 47% in the newest poll. Overall, Obama takes 4 of the six current polls, so he ends up with a 90% probability of winning an election held now.

Now, after 100,000 simulated elections, Obama wins 96,863 times and Romney wins 3,137 times (including the 449 ties). Obama receives (on average) 302 (-6) to Romney’s 236 (+6) electoral votes. In an election held now, we would expect Obama to win with a 96.9% (-2.3%) probability and Romney with a 3.1% (+2.3%) probability.

[Read more…]

Obama	Romney
99.2% probability of winning	0.8% probability of winning
Mean of 308 electoral votes	Mean of 230 electoral votes

The previous analysis showed President Barack Obama leading Mitt Romney by a mean of 325 to 213 electoral votes. In an election held then, we would have expected Obama to win with a 99.8% and Romney with a 0.2% probability.

In the last two days there have been a plethora of polls released—16 covering 13 states. Here’s what I’ve found:

		start	end	sample		%	%	%
st	poll	date	date	size	MOE	O	R	diff
CT	Rasmussen	21-Aug	21-Aug	500	4.5	51	43	O+8
FL	Quinnipiac	15-Aug	21-Aug	1241	3.0	49	46	O+3
FL	Foster McCollum White	17-Aug	17-Aug	1503	2.5	39.1	53.3	R+14.3
GA	20/20 Insight	15-Aug	18-Aug	1158	2.9	44	47	R+3
MA	PPP	16-Aug	19-Aug	1115	2.9	55	39	O+16
MI	Glengariff Group	18-Aug	20-Aug	600	4.0	47.5	42.0	O+5.5
MO	PPP	20-Aug	20-Aug	500	4.4	42	52	R+10
MT	Rasmussen	20-Aug	20-Aug	500	4.5	38	55	R+17
NV	SurveyUSA	16-Aug	21-Aug	869	3.4	47	45	O+2
NM	Rasmussen	21-Aug	21-Aug	500	4.5	52	38	O+14
OH	Ohio Poll	16-Aug	21-Aug	847	3.4	49	46	O+3
OH	Quinnipiac	15-Aug	21-Aug	1253	3.0	50	44	O+6
PA	Muhlenberg	20-Aug	22-Aug	422	5.0	49	40	O+9
VT	Castleton Poll	11-Aug	21-Aug	477	—	62	25	O+37
WI	Quinnipiac	15-Aug	21-Aug	1190	3.0	49	47	O+2
WI	Marquette	16-Aug	19-Aug	576	4.2	49.8	44.3	O+5.5

The unsurprising Obama polls are from Vermont, Massachusetts, and Connecticut. Likewise, Romney makes a strong showing in Montana and Missouri.

Romney makes a surprisingly weak +3% showing in Georgia, where he has been typically been doing high single digit or double digit leads.

There is a big surprise in Florida. A Quinnipiac poll gives Obama a +3% edge over Romney. That’s not the surprise. A second poll by Foster McCollum White has Romney up by an amazing (unbelievable?) +14%. This poll requires some discussion.

This poll has triggered a lot of chatter among poll collectors. Is it real? Is it honest? Is it an “outlier.” One of the thing I noticed is that the raw tallies in the poll add up to more than 100%—102.13%, to be exact!

I went to the source; I had a chat with the firm’s pollster Eric Foster. Eric was friendly and most helpful. The poll is not sponsored by anyone. Rather, they firm did the poll on their own, as a way to get their feet wet in Florida (so to speak). It is an independent poll. I asked about the percentages summing to greater than 100%, he suggested this is a result of their weighting. I proposed renormalizing the numbers to proportionately reduce them to 100%, and he agreed this was a good strategy.

We then had a discussion about their weighting method. They went to each county to get voter turn-out information by age group and race/ethnicity and used that to weigh the poll (some of the details were lost in our lousy cell phone to cell phone connection). He stands by the numbers.

In other words, a polling firm that is new to Florida used an elaborate and unorthodox turnout model to weight the poll, and ended up with some kind of error that resulted in percentages totaling greater than 100%. I cannot say that I am convinced that what they have done was done correctly. Even so, it meets my poll criteria and it could be correct. (They’ll look like geniuses if Florida goes double digits for Romney!) So I include it here, and recognize that Florida will be reddish for the next month. But Georgia is not quite red enough and we still need a new poll in South Carolina! That’s the way the polls roll.

Earlier this week two polls in Michigan split between the candidates. The new poll swings back in Obama’s favor (+5.5%).

Nevada has Obama up by a fragile +2%, but Obama has maintained the edge since the start of the year:

New Mexico loves Obama, giving him a +14% lead over Romney this poll.

Two new Ohio polls both go for Obama by +3% and +6%. Likewise, Pennsylvania seems to prefer Obama over Romney by +9%.

Finally, Ryan may have had a small effect on Wisconsin, where Obama scored a +2% in one poll and +5.5% in another. The race has certainly tightened but a lasting effect is not obvious based on these newest polls:

Today, after 100,000 simulated elections, Obama wins 99,173 times and Romney wins 827 times (including the 175 ties). Obama receives (on average) 308 (-17) to Romney’s 230 (+17) electoral votes. In an election held now, Obama would be expected to win with a 99.2% (-0.6%) probability; Romney would have about a 0.8% (+0.6%) probability of winning.

[Read more…]