Sunday, November 4, 2012

Do We Believe the State or National Polls? (Final Popular Vote Projection)

Here is how the past few election cycles voted based on exit polls in terms of political ideology:

Year

Democrat %

Republican %

Independent %

Notes

2000

39

35

27

Gore Wins by 0.1% Popular Vote

2002

38

40

22

Republicans pick up a few House seats

2004

37

37

26

Bush by 3%

2006

38

36

26

Democrats make big gains in the House

2008

39

32

29

Obama by over 7%

2010

35

35

29

Republicans make big gains in the House

Average

37.6666667

35.83333333

26.5

Democrats by 1.88%

The difficult thing to measure is what is going to be the breakdown in political ideology in the 2012 Election. The numbers are crunched at the end of this post.

No matter how you crunch the national numbers, the state poll numbers continue to look challenging for Romney. Despite moving to the lead or tied in the national poll averages (My poll averages go back a month so Obama has a slight lead but the trend is to Romney), Obama still has a big lead in the Electoral College. Romney has to win Florida, Virginia, and Ohio and those state polls are a statistical tie – and Ohio still seems to favor Obama. In fact, other than North Carolina (Romney 4 points), Florida (Romney 1 point), and Virginia (0.3 points) Romney is down in all swing states including – Colorado, Nevada, New Hampshire, Iowa, Wisconsin, and Michigan. And if Romney loses Virginia and or Ohio, he is going to have to win some combination of CO, NV, NH, MI, WI, and IA to win the election. Right now the electoral count is 290 to 248. Good news for Romney is that polls are tightening in Pennsylvania and Minnesota, but Obama still has a 4 to 5 point lead. This may give Romney more options to get to 270 but time is running out as the Romney camp tries to expand their electoral map chances. My internal polling methods have Romney ahead in Florida by 1.3% and Obama ahead in Ohio by 1.2%. If Romney does not win Ohio, his path to the Presidency is going to be hard. I project the final Electoral College count at this time to be 277 to 261 in favor of Obama (More on that tomorrow).

There is still reason for hope. Republicans have become more excited about Romney as the conservative choice for President and polls indicate they are more energized than Democrats because they do not want another 4 years of Obama (look at Gallup Poll demographics and which groups of people are definitely going to vote http://www.gallup.com/poll/154559/US-Presidential-Election-Center.aspx - this says a lot about who is excited about the election). The generic congressional ballot between Democrats and Republicans is even and that is good news for Republicans because they historically trail in this poll. Also, for the first time in the campaign Romney’s favorability ratings are higher than Obama’s. In other good news, polls indicate that Republicans are leading Democrats in early voting, this is an area Democrats usually dominate. Even biased polls show Romney ahead by double digits among Independent voters, a group of voters Obama won by over 7% in 2008. And it would be negligent to point out that millions of Americans are suffering in the aftermath of Sandy, yet the President is campaigning. Finally, my models of state polls indicate that Obama will win the election by 2.9% nationally. However, national polls have the race tied (Obama up by 0.4%). This is a 2.5% divergence between state and national polls! The divergence is even greater in battleground states (nearly 5%). Which polls are right? If the national polls are right, Romney has a chance. If the state polls are right then Obama will win. It is all about turnout and maybe Romney can pull off a few state upsets. My guess is that state polls are wrong because they are widely oversampling Democrat turnout. Is the data listed in the national polls below correct? The average of the last polls by 17 firms has Obama ahead by 0.1% points (47 – 47). However, many of these polls still seem to be oversampling Democrats when compared to historical trends outlined above. My best guess at an ideological breakdown in 2012 election is 38-D/34-R/29-I (the polls listed below have an average breakdown of 38-D/33-R/27-I). If my guess is right, when this election is modeled using these numbers we get the following:

Obama – 47% / Romney 49% (Romney by 1.7%)

Also, using the below data in the polls when broken down by party, gender, and race (not all polls give this data) then my models show the following (I am expecting 53% / 47% margin for women and a 76.5% / 23.5% margin for white versus non-white voters in 2012).

By Party: Obama 48% / Romney 48% (0.3% advantage for Obama)

By Gender: Obama 48% / Romney 47% (0.4% advantage for Obama)

By Ethnicity: Obama 47% / Romney 46% (0.4% average for Obama)

There you have it. I have modeled the polls 5 ways and Obama comes out on top in 4. If I average all 5 methods then we are in a dead heat: Obama 47%, Romney 47% (actual spread favors Romney by.1%). If Democrats only have a 3% advantage on Election Day, well Romney can win the popular vote hands down. If voter turnout favors Republicans as both Gallup and Rasmussen polls suggest then Romney will win the popular vote easily. But the Electoral College is another story (state Polls favor Obama and do not mesh with national polls) – but if voter turnout favors Republicans Romney should win the Electoral College. My final projection for the popular vote is 48.9% to 48.8% (In favor of Romney) with 2.3% voting for another candidate.

All Numbers are in %

Poll

D

R

I

O

RR

MA

Date

IPOS/Reuters

46

38

19

46

46

D 8

4-Nov

Gallup

38

33

29

46

51

D 5

Daily

Rasmussen

37

35

34

49

49

D 2

Daily

NBC

32

26

40

48

47

D 6

4-Nov

FOX

42

37

18

46

46

D 5

31-Oct

PEW

37

34

29

50

47

D 3

4-Nov

National Journal

37

29

29

50

45

D 8

31-Oct

NPR

37

31

31

47

48

D 6

30-Oct

CNN

40

32

28

47

48

D 8

25-Oct

CBS

36

31

33

48

47

D 5

30-Oct

TIPP

40

32

28

45

44

D 8

Daily

Battleground

43

41

15

49

49

D 2

4-Nov

Survey USA

36

33

31

45

48

D 3

22-Oct

AP/GfK

31

25

27

45

47

D 6

25-Oct

WA Times

37

35

28

45

44

D 2

4-Nov

YouGov

45

36

20

48

48

D 9

1-Nov

ABC

35

31

33

48

48

D 4

Daily

PPP

35

29

29

48

49

D 6

22-Oct

UCONN

44

38

18

48

45

D 6

19-Oct

Poll Average

38.32

32.947

27

47

47

5.37

0.105

2012 Model

38

34

28

1

1

1.03

Poll Scale

47

49

-1.8

Poll

DO

RO

IO

DR

RR

IR

MO

FO

MR

FR

WO

NOW

WR

NWR

M

F

W

NW

IPOS/Reuters

87

4

42

8

92

54

Gallup

91

6

43

6

92

43

42

48

52

45

38

78

56

17

48

52

76

24

Rasmussen

90

6

47

8

92

50

NBC

90

7

40

9

92

50

43

51

53

43

36

80

56

17

48

52

74

26

FOX

89

2

39

5

95

46

42

50

51

42

37

78

56

14

47

53

77

23

PEW

94

7

42

5

91

46

42

53

50

40

39

81

54

17

46

54

74

26

National Journal

91

5

43

6

92

49

42

53

52

44

39

81

55

19

46

54

73

27

NPR

94

6

40

6

94

52

45

49

50

46

37

80

58

17

46

54

74

26

CNN

93

4

41

6

96

49

47

53

50

44

41

73

56

22

47

53

74

26

CBS

93

6

39

5

91

51

44

52

51

44

39

83

56

14

46

54

79

21

TIPP

87

7

38

7

85

44

46

50

45

40

39

83

51

8

47

53

78

22

Battleground

90

5

44

7

92

43

44

53

52

45

40

84

55

14

77

23

Survey USA

91

6

33

6

90

52

40

49

51

45

38

78

57

17

47

53

78

22

AP/GfK

93

7

43

7

93

57

47

50

51

48

37

71

58

21

50

50

65

28

WA Times

93

8

42

7

92

49

49

47

47

50

42

80

58

17

47

53

74

26

YouGov

48

52

73

27

ABC

91

3

46

8

97

46

43

52

51

44

40

77

55

22

47

53

76

24

PPP

88

5

40

9

94

55

40

49

54

45

39

78

58

18

46

54

72

28

UCONN

88

6

37

9

90

38

41

55

53

38

36

83

56

14

48

52

74

26

Averages

90.72

5.5556

41

7

92

49

44

50.9

51

43.94

39

79.3

56

17

47

53

75

25

2008 Averages

39

32

29

20

27

24

23.287

29.1

18.6

42

3.94

46

54

76

24

2008 Averages

93

7

51

7

93

49

50

57

50

43

45

86

55

14

46

54

76

24

2010 Averages

35

35

29

47

53

78

22

2012 Model

38

34

28

38

34

28

47

53

76

24

34.474

1.88889

11

2.6

31

14

1

1

1

0.9

2012 Model Results

47.86

48

0

48

47

0.395

47

46

0.4

O

RR

Poll Ave

47.26

47.158

Poll Scaled

46.87

48.665

Poll Affiliation

47.86

47.569

Poll Gender

47.5

47.107

Poll Ethnicity

46.62

46.217

Average

47.22

47.343

Key

All Numbers are in %

D - Democrat

R - Republican

I - Independent

O - Obama

RR - Romney

MA - Margin

DO = Democrats for Obama

RO - Republicans for Obama

IO - Independents for Obama

DR - Democrats for Romney

RR - Republicans for Romney

IR - Independents for Romney

MO - Males for Obama

FO - Females for Obama

MR - Males for Romney

FR - Females for Romney

WO - Whites for Obama

NWO - Non Whites for Obama

WR - Whites for Romney

NWR - Non Whites for Romney

F - Female

M - Male

W - Whites

NW - Non Whites

No comments:

Post a Comment