General Election 2008

Obama (D) vs. McCain (R)


The 9.25 Democratic Pick-up states

-Summary, Part I-


by Mark Rosenthal


This is a „non-partisan“ end-analysis of the 9.25 states that switched from the Republican party to the Democratic party in the General Election for President on November 4, 2008.


„Non-partisan“ means that I use no arguments for or from either side of the political spectrum to explain what happened or why it happened, but rather, I let the numbers speak for themselves. This is a purely statistical study comparing the results of 2008 with the results of 2004 and in the case of 40 select counties, all the way back to 1960; the results lead to very, very concrete conclusions. I can assure you: this summary contains more detailed and interesting information on the pick-up states than you can find elsewhere.


An exhaustive analysis of the General Election for the entire Union can be found at Google Docs, in 5 parts:


Full analysis Part I, Full analysis Part II, Full analysis Part III, Full analysis Part IV, Full analysis Part V, raw-vote total data, Obama's standing in the national rankings since 1824, Obama's standing in the rankings, per state.



A general analysis of the pick-up states is in Part IV of the above, but not nearly as detailed as what it being presented with this report and also in the following individual analyses. I recommend that you read this summary first and then click on the hyperlinks to the state or states of your choice:


Mid-west:

OHIO - Part I, Part II, Part III , raw data / INDIANA - Part I, Part II and Part III, raw data

IOWA – Part I, Part II, Part III, raw data / NEBRASKA CD-02 – raw data


South:

VIRGINIA – Part I, Part II, Part III, raw data / NORTH CAROLINA – Part I, Part II, Part III, raw data

FLORIDA – Part I, Part II, Part III, raw data


West:

COLORADO – Part I , Part II , Part III , raw data, special 9-county 48-year voting history study

Supplemental to Colorado: DEMOGRAPHIC and ECONOMIC profile of Colorado (p.4, hispanic population)



NEW MEXICO – Part I, Part II, Part III, raw data, special 12 county 48-year voting history study

Supplemental to Nevada: DEMOGRAPHIC and ECONOMIC profile of New Mexico(p.4, hispanic population)


NEVADA - Part I, Part II , Part III, raw data, special 6 county 48-year voting history study

Supplemental to Nevada: DEMOGRAPHIC and ECONOMIC profile of Nevada (p.4, hispanic population)

Quick Census facts on Nevada


A analysis for NE-02 (which is the „.25“ part of „9.25“) will be published when I have received the complete precinct data for Douglas and (part of) Sarpy counties from election officials who are willing to dig up the data over 48 years for me. But a comparison 2008 to 2004 is already possible and here is the raw-data. This is the first time in US history that a presidential candidate has picked-up one elector from Nebraska while the other candidate won the rest of the state.


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



President Barack Obama won the General Election 2008 with 52.87% of the national popular vote. Here a quick review of the national figures, for comparison purposes:


USA DEM GOP IND Total DEM % GOP % IND % Margin Mar % % of NPV growth rate calc electors
2008 69498216 59948140 1996142 131442498 52,87% 45,61% 1,52% 9550076 7,27% 100,00%
538
2004 59028439 62041590 1224499 122294528 48,27% 50,73% 1,00% 3013151 2,46% 100,00%
538
Diff: 10469777 -2093450 771643 9147970 4,61% -5,12% 0,52% 12563227 9,73% 0,00% 7,48%
Obama/Bush 7456626












Through the results of the national popular vote, Obama retained all of the states that John Kerry (D) won in 2004 and then added 9.25 states, worth 113 Electoral Votes, to the Democratic column. Those states are:


Midwest: Ohio, Indiana, Iowa, Nebraska CD-02

South: Virginia, North Carolina, Florida

West: Colorado, New Mexico, Nevada


(The above hyperlinks are to the raw data, county for county, for each state)


Here a quick overview of the pick-ups (alphabetical):


State DEM GOP IND Total DEM % GOP % IND % Margin Mar % % of NPV
Colorado 1288576 1073589 39196 2401361 53,66% 44,71% 1,63% 214987 8,95% 1,83%
Florida 4282074 4045624 82622 8410320 50,91% 48,10% 0,98% 236450 2,81% 6,40%
Indiana 1374039 1345648 33338 2753025 49,91% 48,88% 1,21% 28391 1,03% 2,09%
Iowa 828940 682379 25804 1537123 53,93% 44,39% 1,68% 146561 9,53% 1,17%
Nebraska CD-02 138809 135439 3561 277809 49,97% 48,75% 1,28% 3370 1,21% 0,21%
Nevada 533736 412827 21285 967848 55,15% 42,65% 2,20% 120909 12,49% 0,74%
New Mexico 472422 346832 10904 830158 56,91% 41,78% 1,31% 125590 15,13% 0,63%
North Carolina 2142651 2128474 39664 4310789 49,70% 49,38% 0,92% 14177 0,33% 3,28%
Ohio 2940044 2677820 103510 5721374 51,39% 46,80% 1,81% 262224 4,58% 4,35%
Virginia 1959532 1725005 38723 3723260 52,63% 46,33% 1,04% 234527 6,30% 2,83%











Total: 15960823 14573637 398607 30933067 51,60% 47,11% 1,29% 1387186 4,48% 23,53%


If we expand the overview to include a comparison to 2004, which allows us to calculate partisan-shift, then the table looks like this:


State DEM GOP IND Total DEM % GOP % IND % Margin Mar % % of NPV growth rate calc electors
Colorado 1288576 1073589 39196 2401361 53,66% 44,71% 1,63% 214987 8,95% 1,83%
9,83
2004 1001732 1101255 27343 2130330 47,02% 51,69% 1,28% 99523 4,67% 1,74%
9,37
Diff: 286844 -27666 9524 268702 6,64% -6,99% 0,35% 314510 13,62% 0,08% 12,61% 0,46













Florida 4282074 4045624 82622 8410320 50,91% 48,10% 0,98% 236450 2,81% 6,40%
34,42
2004 3583544 3964522 61744 7609810 47,09% 52,10% 0,81% 380978 5,01% 6,22%
33,48
Diff: 698530 81102 1302 780934 3,82% -3,99% 0,17% 617428 7,82% 0,18% 10,26% 0,95













Indiana 1374039 1345648 33338 2753025 49,91% 48,88% 1,21% 28391 1,03% 2,09%
11,27
2004 969011 1479438 19553 2468002 39,26% 59,94% 0,79% 510427 20,68% 2,02%
10,86
Diff: 405028 -133790 11745 282983 10,65% -11,07% 0,42% 538818 21,71% 0,08% 11,47% 0,41













Iowa 828940 682379 25804 1537123 53,93% 44,39% 1,68% 146561 9,53% 1,17%
6,29
2004 741898 751957 13053 1506908 49,23% 49,90% 0,87% 10059 0,67% 1,23%
6,63
Diff: 87042 -69578 12849 30313 4,69% -5,51% 0,81% 156620 10,20% -0,06% 2,01% -0,34













Nebraska CD-02 138809 135439 3561 277809 49,97% 48,75% 1,28% 3370 1,21% 0,21%
1,14
2004 97958 153401 3133 254492 38,49% 60,28% 1,23% 55443 21,79% 0,21%
1,12
Diff: 40851 -17962 3571 26460 11,47% -11,52% 0,05% -58813 23,00% 0,00% 10,40% 0,02













Nevada 533736 412827 21285 967848 55,15% 42,65% 2,20% 120909 12,49% 0,74%
3,96
2004 397190 418690 13707 829587 47,88% 50,47% 1,65% 21500 2,59% 0,68%
3,65
Diff: 136546 -5863 7578 138261 7,27% -7,82% 0,55% 142409 15,08% 0,06% 16,67% 0,31













New Mexico 472422 346832 10904 830158 56,91% 41,78% 1,31% 125590 15,13% 0,63%
3,4
2004 370942 376930 8432 756304 49,05% 49,84% 1,11% 5988 0,79% 0,62%
3,33
Diff: 101480 -30098 2471 73853 7,86% -8,06% 0,20% 131578 15,92% 0,01% 9,76% 0,07













North Carolina 2142651 2128474 39664 4310789 49,70% 49,38% 0,92% 14177 0,33% 3,28%
17,64
2004 1525849 1961166 13992 3501007 43,58% 56,02% 0,40% 435317 12,43% 2,86%
15,4
Diff: 616802 167308 25672 809782 6,12% -6,64% 0,52% 449494 12,76% 0,42% 23,13% 2,24













Ohio 2940044 2677820 103510 5721374 51,39% 46,80% 1,81% 262224 4,58% 4,35%
23,42
2004 2741167 2859768 26973 5627908 48,71% 50,81% 0,48% 118601 2,11% 4,60%
24,76
Diff: 198877 -181948 59736 76665 2,68% -4,01% 1,33% 380825 6,69% -0,25% 1,36% -1,34













Virginia 1959532 1725005 38723 3723260 52,63% 46,33% 1,04% 234527 6,30% 2,83%
15,24
2004 1454742 1716959 26666 3198367 45,48% 53,68% 0,83% 262217 8,20% 2,62%
14,07
Diff: 504790 8046 12057 524893 7,15% -7,35% 0,21% 496744 14,50% 0,22% 16,41% 1,17


The partisan shift is found in the „Mar %“ column in each „Diff:“ row, i.e. 14.50% in the case of Virginia.


Here the geography of the pick-ups:







Here is how the pick-ups fit geographically into the map alongside the Democratic retentions:



We see that 6 of the Democratic pick-up states are geographically attached to the Democratic territory from 2004.



Here is how the Democratic states fit geographically into the map with the five narrowest Republican wins from 2008, 4 of which were considered battleground states by one side or both:


 




Again, we see a real geographic proximity of at least 4 of these narrow GOP wins to Democratic territory.


How does this compare on a historical level?


We have never had a General Election where absolutely no state or states changed hands at all. Pick-ups (and correspondingly, losses) happen in every cycle. Is a pick-up of 9.25 states large, average or small? And how do we wish to measure this?


First, here is a table of the pick-ups for each cycle, going back to 1952:


Year

Dem pick-ups

GOP pick-ups

Ind Pick-ups

Total pick-ups

Total pick-up EV

Notes

2008 (O)

9.25

0

0

9.25

113

.25 = 1 elector from NE-02. Obama elected with an electoral landslide and a solid PV margin. Obama's win transforms the electoral map, with gains in the south, the midwest and the west. So-called "re-alignment election".

2004

1

2

0

3

4 / 12 (+8) (+15)

Bush re-elected narrowly.The election came down to Ohio. Bush however gains 3% over 2000 in the PV.

2000 (SO)

0

10

0

10

112

Razor-thin election & "electoral misfire". Closest electoral win since 1876. Narrowest popular vote win since 1960 and the third narrowest in history, after 1880 and 1960. One Gore Elector abstained (DC).

1996

2

3

0

5

33 / 24 (+9)

Clinton re-elected (3 way) in an electoral landslide; minority PV win.

1992

22

0

0

22

269

Clinton elected (3 way) in an electoral landslide; minority PV win.

1988

9

0

0

9

98

Bush Sr. wins with an electoral landslide and a solid PV margin; Democrats regain a small amount of territory over losses from 1984.

1984

0

4

0

4

36

Reagan re-elected in a massive landslide: a 49 state sweep, including Massachusetts. Highest number of electors won by a presidential candidate in US history. Most evenly spread sweep in US history.

1980

0

18

0

18

249

Reagan elected in landslide (3 way race), but at the same time, with a bare majority PV win. So-called "re-alignment" election and beginning of the "Reagan revolution".

1976 (SO)

22

0

0

22

280

Carter wins narrowly; Ohio and Mississippi put him over the top at 02:30 am. Bare majority win.

1972

0

17

0

17

219

5 of 17 GOP pick-ups are from independent (Wallace) states  from 1968. Nixon wins in 3nd largest landslide in history, percentually and in number of electors captured, but his sweep is not as evenly spread as Reagan's from 1984. Nixon's raw vote margin is currently still the largest in history. His win transformed the south for the Republican party and paved the way for the Reagan revolution.

1968 (O)

0

27

5

32

46 / 249

1 TN elector voted IND. Extremely divisive election. Most historians contend that Wallace (I) spoiled the election for Humphrey, regardless of the Democratic party's unpopularity over the Vietnam war.

1964

22

6

0

28

180 + DC = 183 / 52

DC added to electoral college; 1 GOP pick-up was from an undeclared state in 1960. Johnson wins with largest PV landslide percentage in US history and the second largest winning margin, after Warren Harding (1920). The southern states that defected to Goldwater (R) did so over the Civil Rights Act of 1964. Obama's win in 2008 can be statistically compared in many ways to 1964.

1960

16

0

1

17

244 / 14 (+230)

Kennedy elected by 0.17% PV margin, the second smallest in history, after 1880. Historians are not convinced that 1960 was a re-alignment election and Kennedy did not live long enough to complete the platform upon which he campaigned.

1956

1

3

0

4

13 / 28 (+15)

Immensely popular Eisenhower re-elected in a landslide. Virginia becomes cemented into the Republican column.

1952 (O)

4

22

0

26

38 / 253 (+253)

The 4 Dem pick-ups are from an independent from 1948, in the south. Eisenhower wins in a landslide. So-called "re-alignment election" and end of 24 years of Democratic rule.

1948 (SO)

5

9

4

18

56 / 146 / 39

(-129)

1 TN elector voted for Thurmond. This election was the warning shot in the coming debate and action over civil rights. And was the most stunning upset of the 20th century: most were convinced that Dewey (R) would win.



From the table above we see that George W. Bush, Jr. picked up 112 electors in 2000, almost identical to Obama's pick-up total from 2008.


However, in 1992 (Dem win, GOP incumbent unseated), 1980 (GOP win, DEM incumbent unseated) and 1976 (DEM win, GOP non-elected incumbent unseated), the elector pick-up was between 249 and 280 electors, making for roughly 2.2-2.5 times as many switched electors. So, the table suggests that as the nation becomes more and more ideologically divided, the more difficult it is to pick-up over 120 EV in a given cycle. The largest elector swing in an open election (as 2008 was) since 1952 was indeed 1952: Eisenhower +253 EV. The largest elector swing in a re-election was 1972, but it is difficult to compare a 2 way win in a re-election to a 3-way win in a first term.


By Bush in 2004 there are two EV totals in parenthesis, one in italic. Bush picked up 12 electors over 2000 (New Mexico, Iowa), but lost 4 electors (New Hampshire). Nonetheless, he gained 15 electors total over 2000 due to re-apportionment after the 2000 census. This means that 6 electors were reapportioned from „blue“ states in 2000 to „red“ states in 2004.


Notice that the number of electors gained by the winner in a re-election has steadlily DECREASED since 1984: Reagan picked up 36 EV in 1984. Clinton picked-up 33 EV but lost 24 EV (total +9). Bush 43 picked up 12 EV but lost 4 (with reapportionment, +15).


So, the answer to the above question is: Obama's pick-up total is on par with the last open election. It is less than Bill Clinton's pick-ups in 1992, but then again, the Democratic party had a much longer way to go just to get to 270 EV in 1992.


On to the county level:



These 9.25 pick-up states encompass 696 out of 3,141 „counties“ in the USA.

These 696 counties account (numerically) for 22.16% of all counties in the USA.


In the pick-up states, the Democratic party retained 146 of 148 Democratic counties from 2004 and then picked-up an additional 89 counties, for a total of 235 counties (33.76%). The Republican party lost 89 counties from 2004, retaining 459 counties and then picked-up 2 counties, for a total of 461 (66.24%). 235 + 461 = 696 counties. More detail on all of this after the tables below.



Here the totals for all the pick-ups combined:


County DEM GOP IND Total DEM % GOP % IND % Margin Mar % County – main City % of PV Growth rate % of 9 pick-ups
9.25 pick-up states












2008 15960823 14573637 398607 30933067 51,60% 47,11% 1,29% 1387186 4,48% 2008 23,53% 10,94% 100,00%
2004 12884033 14784086 214596 27882715 46,21% 53,02% 0,77% 1900053 6,81% 2004 22,80%
100,00%
Diff: 3076790 -210449 184011 3050352 5,39% -5,91% 0,52% 3287239 11,30% Diff: 0,73%
0,00%
Obama/Bush 1176737













In the table above we see that the overall winning margin for all 9.25 pick-up states combined (+4.48%) was smaller than the national winning margin (+7.27%) and comes closest to Obama's winning margin in the state of OHIO (+4.85%). However, the partisan shift (or swing) that was caused by winning these states (+11.30%) is LARGER than the national partisan shift (+9.73%). The average growth rate of the 9.25 pick-up states combined was 10.94%, well above the national growth rate of 7.48%.


The last column, called „calc electors“, requires some explanation: I did a quick calculation to see what the electoral college would look like if the distribution of electors were to be decided AFTER all the votes in the GE were tallied and those electors were to be assigned to each state on a purely percentual basis. In other words, since the total votes from the 9.25 pick-up states accounted for 23.53% of the national popular vote, 23.53% of 538 electors = 122.61 electors. In reality, those states combined have 113 electors. This is an experiment for an analysis at a later date, but the advantage of this system would obviously be increased voter turnout: people would no longer think that their vote doesn't count, for the higher the percentage of the national popular vote from your state, the higher the total of electors your state would be awarded, and still according to the WTA (winner takes all) system.


Here an overview of the 696 counties, in descending order of the number of counties per state.


State

EV

Counties

DEM counties

DEM %

GOP counties

GOP %

Virginia

13

134*

49

36.57%

85

63.43%

North Carolina

15

100

33

33.00%

67

67.00%

Iowa

7

99

53

53.53%

46

46.47%

Indiana

11

92

15

16.30%

77

83.70%

Ohio

20

88

22

25.00%

66

75.00%

Florida

27

67

15

22.39%

52

77.61%

Colorado

9

64

26

40.63%

38

59.37%

New Mexico

5

33

18

54.54%

15

45.45%

Nevada

5

17*

3

17.65%

14

82.35%

NE-02

1

2

1

50.00%

1

50.00%

TOTAL:

113

696

235 of 696

33.76%

461 of 696

66.24%


*Virginia has both counties and "independent cities". Those independent cities are treated as counties for census, administrative and electoral purposes. Nevada has 16 counties and 1 "incorporated city".


Here a similar overview, with the total number of counties per party broken down into retentions and pick-ups:


State

Counties

DEM retentions

DEM pick-ups

DEM %

GOP retentions

GOP pickups

GOP %

Virginia

134

31

18

36.57%

83

2

63.43%

North Carolina

100

20

13

33.00%

67

0

67.00%

Iowa

99

32

21

53.53%

46

0

46.47%

Indiana

92

4

11

16.30%

77

0

83.70%

Ohio

88

16

6

25.00%

66 of 88

0

75.00%

Florida

67

11

4

22.39%

52

0

77.61%

Colorado

64

19

7

40.63%

38

0

59.37%

New Mexico

33

12

6

54.54%

15

0

45.45%

Nevada

17

1

2

17.65%

14

0

82.35%

NE-02

2

0

1

50.00%

1

0

50.00%

TOTAL:

696

146

89

33.76%

459

2

66.24%



Numerically and percentually, Obama picked up the most counties in IOWA: 21 (21.21% of state total).

Numerically, he picked up the least counties in NV: 2

Percentually, he picked up the least counties in FL: 4 (5.97% of state total)


The GOP picked up 2 counties, both in VA.


How to measure change? One way is to measure the raw vote gain or loss over 2004 in each county:



State

Counties

DEM num loss

DEM num gain

GOP num loss

GOP num gain

IND num loss

IND num gain

VA

134

8

126

44

90

20

114

NC

100

1

99

93

7

2

98

IA

99

11

88

98

1

2

97

IN

92

0

92

85

7

1

91

OH

88

17

71

82

6

0

88

FL

67

12

55

56

11

6

61

CO

64

0

64

51

13

4

60

NM

33

0

33

32

1

8

25

NV

17

0

17

14

3

0

17

NE-02

1

0

2

2

0

0

2

TOT:

696

49

647

557

139

43

653


Here, the Democratic party added votes over 2004 in 647 counties and the independent vote improved also numerically – in 653 counties, while the Republican party lost raw votes over 2004 in 557 counties.


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


Another method is to measure % loss or gain per county:


State

Counties

DEM% loss

DEM% gain

GOP % loss

GOP % gain

IND % loss

IND % gain

VA

134

6

128

126

8

25

109

NC

100

11

89

94

6

3

97

IA

99

2

97

99

0

1

98

IN

92

0

92

92

0

1

91

OH

88

16

72

83

5

0

88

FL

67

25

42

44

23

7

60

CO

64

0

64

64

0

5

59

NM

33

0

33

33

0

2

31

NV

17

0

17

17

0

0

17

NE-02

2

0

2

2

0

0

2

TOT:

696

60

636

654

42

44

652



From the table above, the Democratic party gained on percentage in 636 counties, while the Republican party lost on percentage in 654 counties. The independent vote improved percentually in 652 counties. But a gain in percentage does not guarantee that the state trends in one way or the other. It is possible for both major parties to slightly lose or gain in percentage, depending on the 3rd party vote change.


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


But the method that measures the „swing“ or partisan shift of a county is the measurement of margin gain or loss over 2004 is the method that is used the most:



State

Counties

DEM margin loss

DEM margin gain

GOP margin loss

GOP margin gain

VA

134

9

125

125

9

NC

100

8

92

92

8

IA

99

1

98

98

1

IN

92

0

92

92

0

OH

88

11

77

77

11

FL

67

25

42

42

25

CO

64

0

64

64

0

NM

33

0

33

33

0

NV

17

0

17

17

0

NE-02

2

0

2

2

0

TOT:

696

54

642

642

54


Nationally, all 9.25 states trended Democratic as Obama won them and their electors according to the WTA (winner-takes-all) system, but when we look at the inner details, the picture is much clearer: 642 of 696 counties in the pick-up states (92.24%) swung Democratic. The remaining 54 counties (7.76%) trended Republican. This indicates a statistical grand sweep for the Democratic party in the pick-up states.


In 4 states, the ENTIRE state trended Democratic: Colorado, New Mexico, Nevada (all three western pick-ups) and Indiana.The pick-up in Indiana is historic not only because this is the first time since 1964 that a Democrat has won the state, but it also because Indiana had the largest cross-partisan shift of all 50 states in the GE 2008: +21.71%


The state with the largest contra-trend (Republican) against the national trend: Florida.


From the state analyses, most every county in OH, VA and NC (28 total) that trended more Republican than in 2004 was a border county to the Appalachian region of the country. Example: OHIO


 



And Virginia:


 


And North Carolina:


 




This point will be addressed again in a future study of the "Appalachian" states and their performance in 2008. But there is statistical evidence that the small pull toward the GOP in the election was a specifically regionally based pull, mostly in the Appalachian region.


Here to Part II