Poll Analysis: Romney takes back a couple more

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).

Here are the new polls included:

		start	end	sample		%	%	%
st	poll	date	date	size	MOE	O	R	diff
AZ	Arizona	01-Oct	03-Oct	594	4.0	44	53	R+9
CO	WeAskAmerica	25-Sep	27-Sep	1273	2.8	49.3	45.5	O+3.8
FL	Marist	30-Sep	01-Oct	890	3.3	47	46	O+1
FL	Gravis Marketing	29-Sep	30-Sep	914	3.4	49.0	48.4	O+0.6
FL	Suffolk	27-Sep	30-Sep	600	4.0	45.8	43.2	O+2.7
IA	WeAskAmerica	25-Sep	27-Sep	1273	2.8	47.5	43.7	O+3.8
LA	SMOR	11-Sep	20-Sep	600	4.0	39	45	R+6
MD	Baltimore Sun	25-Sep	27-Sep	804	3.5	57	34	O+23
MA	WBUR	26-Sep	28-Sep	504	4.4	60	32	O+28
MI	WeAskAmerica	25-Sep	27-Sep	1064	3.1	52.0	39.9	O+12.1
MO	PPP	01-Oct	03-Oct	700	3.7	45	51	R+6
MO	WeAskAmerica	25-Sep	27-Sep	1145	2.9	44.5	47.7	R+3.2
MO	Gravis Marketing	16-Sep	17-Sep	1959	2.3	39.3	47.7	R+8.4
NV	WeAskAmerica	25-Sep	27-Sep	1078	3.1	52.5	42.0	O+10.5
NH	U NH	27-Sep	30-Sep	600	4.0	54	39	O+15
NH	PPP	24-Sep	25-Sep	862	3.3	51.0	43.5	O+7.5
NJ	Eagleton-Rutgers	27-Sep	30-Sep	645	4.0	56	37	O+19
NM	Rasmussen	27-Sep	27-Sep	500	4.5	51	40	O+11
NM	WeAskAmerica	25-Sep	27-Sep	1258	2.9	50.9	40.6	O+10.3
NC	Rasmussen	02-Oct	02-Oct	500	4.5	47	51	R+4
NC	SurveyUSA	29-Sep	01-Oct	573	4.2	49	47	O+2
NC	ARG	28-Sep	30-Sep	600	4.0	46	50	R+4
NC	PPP	27-Sep	30-Sep	981	3.1	48	48	tie
OH	Marist	30-Sep	01-Oct	931	3.2	51	43	O+8
OH	PPP	27-Sep	30-Sep	897	3.3	49	45	O+4
RI	WPRI	26-Sep	29-Sep	501	4.4	57	33	O+24
TX	Texas Lyceum	10-Sep	26-Sep	1175	2.8	39	58	R+19
VA	Marist	30-Sep	01-Oct	969	3.1	48	46	O+2
VA	Roanoke College	19-Sep	28-Sep	589	4.0	48	40	O+8
WI	Marquette	27-Sep	30-Sep	894	3.3	52.8	42.3	O+10.5

Romney has a solid lead in Arizona. He leads Obama by +9% in the new poll and also leads in the three other current polls.

Obama has a six-poll streak in Colorado, and he takes seven of the eight current polls.

Obama pulls a hat trick in Florida today. This brings Obama’s streak up to ten consecutive Florida polls. Romney only takes a couple of the earliest of the 14 current polls.

The new Iowa polls goes to Obama by +3.8%. Romney has taken two of the seven current polls and the analysis gives Obama a 99% probability of taking the state (right now).

I was a little surprised that Louisiana has Romney up by only +6% over Obama. The only other poll in the state was from a year ago and had Romney up by +16%.

A double digit lead for Obama in Michigan. Obama takes all eight current polls.

In Missouri, Romney leads Obama in all three new polls. The four current polls have Romney with a quite solid lead.

Nevada gives Obama a double digit lead over Romney. Aside from one tie, Obama leads in all seven current polls. His chances are, like, 99.5% of taking the state right now.

New Hampshire isn’t a swing state, if you believe the new poll’s +15% for Obama over Romney. That is bolstered by a +7% in another poll. The other five current polls show a narrower spread, and Romney even led in one.

In New Mexico, Obama leads Romney by +11% in one new poll and +10.3% in another. With the other +9% current poll, Obama’s chances are now at 100%.

The best news for Romney is the new pack of North Carolina polls. Two have Romney up by +4% over Obama. Obama leads in another by +2%, and the last one is a tie. Combined with the six other current polls, Obama ekes out a 50.1% to 49.9% lead, and an overall 56% probability of taking the state in an election right now.

Ohio has Obama up by +8% over Romney, but with 51%, in one poll and +4% in another. With all nine of the current polls in his favor, Obama has established a solid +6% lead that offers him 100% of the simulated elections.

Two new Virginia polls go to Obama by +2% and +8%. Obama now leads in all twelve of the current polls, giving him 100% chances with about a 6% lead expected.

A new double digit lead for Obama in Wisconsin, and all the current polls go Obama’s way.

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

What an interesting distribution! There is slightly greater than 50:50 chance that one of four outcomes would result from an election now: Obama winning by 347, 348, 333, or 332 electoral votes. The results come as pairs one electoral vote apart because of the tie in Nebraska-2.

Ten most probable electoral vote outcomes for Obama:

• 347 electoral votes with a 14.74% probability
• 348 electoral votes with a 14.23% probability
• 333 electoral votes with a 12.37% probability
• 332 electoral votes with a 11.43% probability
• 341 electoral votes with a 3.31% probability
• 356 electoral votes with a 3.29% probability
• 340 electoral votes with a 2.71% probability
• 355 electoral votes with a 2.61% probability
• 349 electoral votes with a 2.59% probability
• 346 electoral votes with a 2.56% probability

After 100,000 simulations:

• Obama wins 100.0%, Romney wins 0.0%.
• Average (SE) EC votes for Obama: 342.1 (10.1)
• Average (SE) EC votes for Romney: 195.9 (10.1)
• Median (95% CI) EC votes for Obama: 346 (323, 359)
• Median (95% CI) EC votes for Romney: 192 (179, 215)

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	205
Strong Obama	126	331
Leans Obama	1	1	332
Weak Obama	15	15	15	347
Weak Romney	1	1	1	191
Leans Romney	9	9	190
Strong Romney	102	181
Safe Romney	79

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*	404	39.6	60.4	0.2	99.8
AK			3	0*				(0)	(100)
AZ			11	4	2044	46.5	53.5	1.1	98.9
AR			6	1	2006	38.3	61.7	0.0	100.0
CA			55	1	915	57.6	42.4	100.0	0.0
CO			9	7	6205	51.7	48.3	97.4	2.6
CT			7	2	1193	58.8	41.2	100.0	0.0
DE			3	0				(100)	(0)
DC			3	1*	94	88.3	11.7	100.0	0.0
FL			29	14	11413	51.5	48.5	98.7	1.3
GA			16	1	439	38.5	61.5	0.1	99.9
HI			4	1*	517	64.8	35.2	100.0	0.0
ID			4	0*				(0)	(100)
IL			20	2*	2277	58.8	41.2	100.0	0.0
IN			11	1	736	43.5	56.5	0.6	99.4
IA			6	7	4804	52.3	47.7	98.8	1.2
KS			6	2*	1143	39.4	60.6	0.0	100.0
KY			8	1	557	42.4	57.6	0.4	99.6
LA			8	1	504	46.4	53.6	12.9	87.1
ME			2	4	2536	58.4	41.6	100.0	0.0
ME1			1	1	412	63.3	36.7	100.0	0.0
ME2			1	1	364	53.8	46.2	85.9	14.1
MD			10	2	1471	61.5	38.5	100.0	0.0
MA			11	7	3708	63.0	37.0	100.0	0.0
MI			16	8	7283	54.3	45.7	100.0	0.0
MN			10	1	704	54.5	45.5	96.1	3.9
MS			6	1*	717	40.0	60.0	0.0	100.0
MO			10	4	4204	46.6	53.4	0.0	100.0
MT			3	1	581	45.1	54.9	4.5	95.5
NE			2	1	728	44.0	56.0	1.1	98.9
NE1			1	1*	389	45.5	54.5	10.8	89.2
NE2			1	1	352	50.0	50.0	49.2	50.8
NE3			1	1*	284	35.9	64.1	0.0	100.0
NV			6	7	4633	52.7	47.3	99.5	0.5
NH			4	6	3809	53.2	46.8	99.7	0.3
NJ			14	2	1146	59.3	40.7	100.0	0.0
NM			5	3	4562	55.1	44.9	100.0	0.0
NY			29	1*	1426	64.6	35.4	100.0	0.0
NC			15	9	5551	50.1	49.9	56.0	44.0
ND			3	1*	348	41.4	58.6	0.9	99.1
OH			18	12	11674	53.3	46.7	100.0	0.0
OK			7	1*	431	33.4	66.6	0.0	100.0
OR			7	1*	499	54.9	45.1	93.8	6.2
PA			20	9	7087	54.9	45.1	100.0	0.0
RI			4	1	451	63.4	36.6	100.0	0.0
SC			9	3*	4199	48.2	51.8	4.9	95.1
SD			3	1*	474	41.8	58.2	0.5	99.5
TN			11	1*	654	46.0	54.0	7.5	92.5
TX			38	1	1140	40.2	59.8	0.0	100.0
UT			6	1*	1149	27.7	72.3	0.0	100.0
VT			3	1*	415	71.3	28.7	100.0	0.0
VA			13	12	11432	52.4	47.6	100.0	0.0
WA			12	2	1059	57.9	42.1	100.0	0.0
WV			5	1*	361	42.1	57.9	1.9	98.1
WI			10	7	6210	54.3	45.7	100.0	0.0
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.