07 December 2012

Prelim Comparison III: BIG SKY COUNTRY (WA, ID, MT, ND, SD)


Washington State, Idaho, Montana, both Dakotas - all, final results (certified), all non-competitive states from 2012.



Montana is one of the states on the Romney side that was polled pretty extensively, and he exceeded his end polling average, which looked like this:

Nr.MONTANADateSampleMoEObamaRomneyUndMar.

AVERAGEN/AN/AN/A43,4352,106,008,68


















18PPP (D) – FINAL POLL04.11.12836 LV+/-3.4455257
17Lee Newspapers / Mason-Dixon04.11.12625 LV+/-4.04353510
16Rasmussen04.11.12500 LV+/-4.54353510
15Pharos Research30.10.12765 LV+/-3.542,750,46,97,7
14Pharos Research24.10.12828 LV+/-3.441,347,4611,246,16
13PPP (D) / LCV17.10.12806 LV+/-3.54353510
12Rasmussen17.10.12500 LV+/-4.5455358
11Montana State Univ. - Billings10.10.12477 A+/-5.03549514
10PPP (D)10.10.12737 LV+/-3.64152511
9Mellman Group (D)06.10.12600 LV+/-4.0444854
8Mason-Dixon / Lee Newspapers24.09.12625 LV+/-4.0425159
7PPP (D)12.09.12656 LV+/-3.8455055
6Rasmussen20.08.12450 LV+/-4.53855717
5Rasmussen20.06.12450 LV+/-4.5425179
4Rasmussen04.05.12450 LV+/-4.5445157
3PPP (D)02.05.12934 LV+/-3.24348105
2Rasmussen04.04.12500 LV+/-4.54049119
1Rasmussen24.02.12500 LV+/-4.54148117

The yellow shaded areas are the polls that were in the final mix, but I have included polls for the entire year, for they are all very telling.

Actual results?


DEM GOP IND Total DEM % GOP % IND % Margin Mar %
Montana 2012 * 201839 267928 14281 484048 41,70% 55,35% 2,95% 66089 13,65%
Montana 2008 232159 243882 16709 492750 47,11% 49,49% 3,39% 11723 2,38%
Diff: -30320 24046 -2428 -8702 -5,42% 5,86% -0,44% 54366 11,27%

Mitt Romney outperformed his end-polling by almost 6 points, way out of the margin of error. If you go down the list to the Montana State U poll from 10/10, you will see that it was the closest to the actual margin (though the toplines were way, way off), but not in the final mix. In the final mix, both Rasmussen and Mason-Dixon, though they were off by 3.7 to the LEFT, were the closest to the final results (yes, the same Rasmussen and the same Mason-Dixon who put out such terrible, identical to each other results for CT, for instance).

In this case, PPP (D) put out the worst polling, having missed the mark by 6.7 points to the left. That is not a good result. Pharos was the second worst, showing Romney +7.7.

But look at the entire table and notice how often single digit margins between 5 and 9 appear. In other words, for most of the year, even the most conservative of pollsters were seeing single digit margins in this once unbreakable bastion of Conservative strength and John McCain barely won by around 2.4 points and a minority win in 2008.

Montana was one of a number of "red" states where Mitt Romney vastly improved upon John McCain's statistic from 2008 (he added 11.27% to the margin), but got to only about 1/2 of the Bush margins from 2004 and 2000, respectively.

------------------------------------------------------------------------------------------

Washington State, unsurprisingly, went for President Obama by a landslide, but the end-polling was indeed interesting:

Nr. WASHINGTON Date Sample MoE Obama Romney Und. Mar.

AVERAGE N/A N/A N/A 54,02 40,28 5,70 13,74


















28 YouGov 04.11.12 837 LV +/-3.3 54 40 6 14
27 PPP (D) – FINAL POLL 03.11.12 932 LV +/-3.2 53 46 1 7
26 Univ. of Washington / KCTS9 01.11.02 632 LV +/-3.9 57,1 36,4 6,5 20,7
25 SUSA 01.11.12 555 LV +/-4.2 54 40 6 14
24 Strategies 360 (D) 22.10.12 500 LV +/-4.4 52 39 9 13
23 Univ. of Washington / KCTS9 18.10.12 644 LV +/-3.9 51,9 42,9 5,2 9
22 PPP (D) / LCV 17.10.12 574 LV +/-3.0 50 45 5 5
21 Rasmussen 17.10.12 500 LV +/-4.5 55 42 3 13
20 SUSA 16.10.12 543 LV +/-4.3 54 40 6 14
19 YouGov 16.10.12 747 LV +/-4.3 56 39 5 17
18 SUSA 03.10.12 540 LV +/-4.3 56 36 8 20

Again, the states marked in yellow were in the final mix.

Actual results?


DEM GOP IND Total DEM % GOP % IND % Margin Mar %
Washington 2012 * 1755396 1290670 79450 3125516 56,16% 41,29% 2,54% 464726 14,87%
Washington 2008 1750848 1229216 73152 3053216 57,34% 40,26% 2,40% 521632 17,08%
Diff: 4548 61454 6298 72300 -1,18% 1,03% 0,15% -56906 -2,22%



Fascinating results: once again, just as in 2008, SUSA pretty much nailed it, with +14 for the President. So did YouGov.

Univ. of Washington/KCTS9 was off by 6 to the LEFT, but PPP was off by 8 to the RIGHT. Once again, people who want to complain that PPP is only LEFT oriented need to remember a state like Washington State. Rasmussen was not in the final mix.

In historical context, yes, Obama's win in Washington State is about 2.2 points under his 2008 statistic but still bigger than any other previous margin for any candidate going back to 1972 (Nixon +18.28%), making Obama's margin from 2012 the fourth highest of any election of the 13 cycles since 1964, after LBJ 1946, Nixon 1972 and Obama 2008. In terms of margin, compared to the "red" side of the spectrum, Obama's win in 2012 is between Romney's wins in Alaska and Texas - both rock solid states that no one expected would switch sides in this year.

------------------------------------------------------------------------------------------

In poor, forgotten Idaho, there was only one poll all year long:

Nr. IDAHO Date Sample MoE Obama Romney Mar. Mar.

AVERAGE N/A N/A N/A





















1 Mason-Dixon / Idaho Statesman 15.10.12 625 LV +/-4.0 27 63 10 36

Actual result?


DEM GOP IND Total DEM % GOP % IND % Margin Mar %
Idaho 2012 * 212787 420911 18576 652274 32,62% 64,53% 2,85% 208124 31,91%
Idaho 2008 236440 403012 19002 658454 35,91% 61,21% 2,89% 166572 25,30%
Diff: -23653 17899 -426 -6180 -3,29% 3,32% -0,04% 41552 6,61%

The Mason-Dixon was 4 points to the RIGHT in estimating the margin in Idaho, but it is really not fair to compare only one poll for an entire year to the results. The same applies to DC or HI on the Democratic side. You really do need a healthy gene-pool of polling results to get a more accurate "pulse" of a state. Nuff said.

------------------------------------------------------------------------------------------

In North Dakota, here the end-polling:

Nr. North Dakota Date Sample MoE Obama Romney Und Mar.

AVERAGE N/A N/A N/A 37,43 54,93 7,65 17,50


















10 Mason-Dixon 30.10.12 625 A +/-4.0 40 54 6 14
9 Pharos Research 30.10.12 752 LV +/-3.6 37,7 54,7 7,6 17
8 Pharos Research 24.10.12 827 LV +/-3.44 39,38 49,44 11,18 10,06
7 Essman Research 23.10.12 500 LV +/-4.4 32 57 11 25
6 Rasmussen 22.10.12 600 LV +/-4.0 40 54 6 14
5 Mason-Dixon 08.10.12 625 A +/-4.0 40 54 6 14
4 DFM Research (D) 28.09.12 600 LV +/-4.0 39 51 10 12
3 DFM Research (D) 27.07.12 400 RV +/-4.9 35 54 11 19
2 Rasmussen 13.07.12 500 LV +/-4.5 36 51 13 15
1 Mason-Dixon 11.06.12 625 LV +/-4.0 39 52 9 13

Actual results?


DEM GOP IND Total DEM % GOP % IND % Margin Mar %
North Dakota 2012 * 124966 188320 9646 322932 38,70% 58,32% 2,99% 63354 19,62%
North Dakota 2008 141403 168887 7448 317738 44,50% 53,15% 2,34% 27484 8,65%
Diff: -16437 19433 2198 5194 -5,81% 5,16% 0,64% 35870 10,97%

Mitt Romney outperformed his polling in North Dakota by about 2 points. Pharos came the closest, with Romney +17, but that is still 2.6 points under.

The worst pollsters were Essman Research (R), which was 5.2 points to the RIGHT and, once again, Mason-Dixon and Rasmussen, which both showed Romney +14 and were 5.6 points to the LEFT.

Interestingly enough, a pollster who was NOT in the final mix, DFM Research,  a Democratic firm from the state, nailed the results, but the poll from July was cold coffee by election day.

------------------------------------------------------------------------------------------

In South Dakota, the end-polling:

Nr. SOUTH DAKOTA Date Sample MoE Obama Romney Und Mar.

AVERAGE N/A N/A N/A





















6 Nielson Brothers 05.11.12 633 LV +/-3.96 41 53 6 12
5 Nielson Brothers 02.11.12 634 LV +/-3.89 42 50 8 8
4 Nielson Brothers 12.10.12 762 LV +/-3.55 41,1 51,6 7,3 10,5
3 Nielson Brothers 06.09.12 512 LV +/-4.33 38,7 53,9 7,4 15,2
2 Nielson Brothers 03.08.12 546 RV +/-4.19 43 49 8 6
1 Nielson Brothers 27.02.12 559 RV +/-4.19 39 48 13 9

There is no end polling for South Dakota as Nielson Brothers was the only firm to poll the state. Their final result, Romney +12, was to the LEFT by 6 points:


DEM GOP IND Total DEM % GOP % IND % Margin Mar %
South Dakota 2012 * 145039 210610 8166 363815 39,87% 57,89% 2,24% 65571 18,02%
South Dakota 2008 170924 203054 7997 381975 44,75% 53,16% 2,09% 32130 8,41%
Diff: -25885 7556 169 -18160 -4,88% 4,73% 0,15% 33441 9,61%

As was the case with Idaho, it is hard to make a comparison of just one poll, or in this case, just one pollster, to the final results. But there are two common threads between the Dakotas that are absolutely in line with history:

-the polling in North Dakota showed a larger Republican margin than in South Dakota.
-the two states are very close to each other in margin, usually right next to each other in the partisan rankings, with North Dakota showing a slightly larger Republican margin.

With very few exceptions, it has been this way in these two states since around 1920. And so it was in 2012:

Rank '12 State EV 2012 Margin '12 Swing: '12 / '08 2012 FINAL PA Diff: '12/'12 PA
24 NC 15 2,04 2,37 2,40 -0,36
23 GA 16 7,82 2,62 8,25 -0,43
22 AZ 11 9,06 0,58 7,67 1,39
21 MO 10 9,38 9,25 10,12 -0,74
20 IN 11 10,20 11,23 11,20 -1,00
19 SC 9 10,47 1,49 6,00 4,47
18 MS 6 11,50 -1,67 15,00 -3,50
17 MT 3 13,65 11,27 8,68 4,97
16 AK 3 13,99 -7,55 22,00 -8,01
15 TX 38 15,78 4,02 17,50 -1,72
14 LA 8 17,21 -1,42 14,07 3,14
13 SD 3 18,02 9,61 12,00 6,02
12 ND 3 19,62 10,97 17,50 2,12
11 TN 11 20,39 5,33 18,00 2,39
10 KS 6 21,72 6,80 19,80 1,92
9 NE 5 21,87 6,94 17,25 4,62
8 AL 9 22,19 0,61 15,00 7,19
7 KY 8 22,69 6,47 14,00 8,69
6 AR 6 23,69 3,84 27,00 -3,31
5 WV 5 26,73 13,64 14,00 12,73
4 ID 4 31,91 6,61 36,00 -4,09
3 OK 7 33,54 2,25 25,90 7,64
2 WY 3 40,82 8,58 40,00 0,82
1 UT 6 48,04 20,02 46,33 1,71



---------------------------------------------------------------------------------

It is interesting to note that in the BIG SKY STATES, the pollsters pretty much switched roles: the Conservative leaning pollsters, when they were off, were generally off to the LEFT, and the Liberal leaning pollsters, when they were off, were off to the RIGHT.

No comments:

Post a Comment

Constructive comments and critique are always welcome. Please keep it polite and respectful.