Monday, August 13, 2012

The Electoral College Map (8/13/12)

Reports of Missouri's general level of competitiveness in the presidential race are greatly exaggerated. But Survey USA's first polling foray in the Show Me state this year certainly makes Missouri appear more swingy. The only problem is that this runs counter to most of the recent -- and by recent I mean arbitrarily polling over the last couple of months -- polls that have shown the race in the former bellwether breaking toward Romney.

Only Public Policy Polling -- in a May survey -- has shown Missouri to be as close as one point between the two nominees. And that type of proximity between the two candidates in Missouri, especially after a near tie there in 2008 and with an overall swing back toward the Republican Party in the time since, probably breaks with the conventional wisdom that a tight national race does not include a tight Missouri (presidential) race.

Of course, by that same logic, North Carolina would not be close in 2012 either. Alas, it is still pre-convention August.

New State Polls (8/13/12)
State
Poll
Date
Margin of Error
Sample
Obama
Romney
Undecided
Poll Margin
FHQ Margin
Missouri
8/9-8/12
+/- 4.1%
585 likely voters
44
45
5
+1
+5.50

Polling Quick Hits:
Missouri:
The emphasis from this poll should be placed on its position within the existing pool of polls in Missouri. Obama's share in the Survey USA sample is comparable to the share the president has enjoyed -- if one can enjoy being behind -- in recent polling there, but the Romney share lags well behind the 49-51% portion of responses the former Massachusetts governor has garnered in those same recent surveys.

This happens to be the first state-level poll to be in the field during and after the newly-selected Republican vice presidential nominee was tapped. However, this poll certainly doesn't say much about the impact of the Ryan rollout. First of all, survey respondents were in no way prompted about Ryan within the poll (not that respondent have to be to show some impact). More importantly, we are likely going to need all two weeks worth of polling prior to the conventions to begin measuring the impact of Romney's selection of Ryan. We may require more data than the two weeks have to offer, but we will have the convention effects to deal with after the August 27-30 week in Tampa.

As always, don't read anything into that connection unless it is confirmed in the coming days in other polls.


As was the case with the lone Iowa poll last Friday, the Missouri survey does little to change how things looked coming into this new week. Sure, the result is that FHQ weighted average in Missouri contracts about 0.8 percent, leapfrogging the state over Tennessee in the Electoral College Spectrum. But the Show Me state is still comfortably red; one of those states that if Mitt Romney were to lose it, the Republican ticket would lose the overall race anyway. Missouri feels like a Lean Romney state in 2012 and it will take a lot of polling data to convince FHQ otherwise.

The Electoral College Spectrum1
RI-4
(7)2
NJ-14
(160)
NH-4
(257)
AZ-11
(167)
MS-6
(55)
HI-4
(11)
WA-12
(172)
OH-183
(275/281)
MT-3
(156)
ND-3
(49)
NY-29
(40)
MN-10
(182)
VA-13
(288/263)
GA-16
(153)
AL-9
(46)
VT-3
(43)
NM-5
(187)
CO-9
(297/250)
WV-5
(137)
KY-8
(37)
MD-10
(53)
CT-7
(194)
IA-6
(303/241)
IN-11
(132)
KS-6
(29)
CA-55
(108)
OR-7
(201)
FL-29
(332/235)
SC-9
(121)
AK-3
(23)
MA-11
(119)
PA-20
(221)
NC-15
(206)
LA-8
(112)
OK-7
(20)
IL-20
(139)
WI-10
(231)
MO-10
(191)
NE-5
(104)
ID-4
(13)
DE-3
(142)
NV-6
(237)
TN-11
(181)
AR-6
(99)
WY-3
(9)
ME-4
(146)
MI-16
(253)
SD-3
(170)
TX-38
(93)
UT-6
(6)
1 Follow the link for a detailed explanation on how to read the Electoral College Spectrum.

2 The numbers in the parentheses refer to the number of electoral votes a candidate would have if he won all the states ranked prior to that state. If, for example, Romney won all the states up to and including Ohio (all Obama's toss up states plus Ohio), he would have 272 electoral votes. Romney's numbers are only totaled through the states he would need in order to get to 270. In those cases, Obama's number is on the left and Romney's is on the right in italics.

3 Ohio is the state where Obama crosses the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line.

Missouri does re-enter the Watch List with the new polling data, but FHQ is less concerned with the red states appearing there than the blue ones with Toss Up next to them.

The Watch List1
State
Switch
Connecticut
from Strong Obama
to Lean Obama
Georgia
from Strong Romney
to Lean Romney
Michigan
from Lean Obama
to Toss Up Obama
Missouri
from Lean Romney
to Toss Up Romney
Nevada
from Lean Obama
to Toss Up Obama
New Hampshire
from Toss Up Obama
to Lean Obama
New Mexico
from Strong Obama
to Lean Obama
North Carolina
from Toss Up Romney
to Toss Up Obama
Ohio
from Toss Up Obama
to Lean Obama
West Virginia
from Strong Romney
to Lean Romney
Wisconsin
from Lean Obama
to Toss Up Obama
1 Weighted Average within a fraction of a point of changing categories.

Please see:



Are you following FHQ on Twitter, Google+ and Facebook? Click on the links to join in.
Posted by Josh Putnam at 5:15 PM 0 comments
Labels: , , , ,
Share |

Saturday, August 11, 2012

Paul Ryan, GOP Vice Presidential Nominee: Some Thoughts

FHQ will not go all long form here (or will try not to anyway). Plenty of others have more than adequately gotten the ball rolling in reaction to the Mitt Romney campaign's decision to tap Wisconsin representative, Paul Ryan, as the governor's running mate.

First thing's first: Let's check the speculation at the door here, shall we? I think we can all agree that the Ryan selection will have an impact. Subtract him and add, say, Tim Pawlenty and you get a potentially different outcome long term. That's the fun part, right? Treating this like the NBA Trade Machine. You can trade and trade and trade parts until you subjectively/hypothetically improve your team's chances of success. But other than gaining some insight into what the campaign is thinking, I don't know that the risk/reward calculus is all that fruitful an exercise. Hindsight is always 20/20 on these things. The one rule to always remember on VP selection is exactly what George W. Bush said about presidential debates, "I don't think you can ever win them, but you darn sure can lose them." In other words, these things are always "do no harm".

Instead of the calculus many are talking about then, let's talk about what we know. What data do we have?

1. Who is Paul Ryan?
A majority of Americans don't know. Those that do know Ryan are slightly more likely to view him unfavorably. That means that the race is on over the next two weeks -- before the Republican Convention kicks off in Tampa (and perhaps afterward) -- to define Ryan in much the same way that the campaigns' efforts have been about defining Romney this summer. Republicans will want to paint the Wisconsin representative as the visionary their standard bearer made the then-unknown pick to be in a recent NBC interview. Oppositely, Democrats will both want to tie Ryan to the frame they have attempted to construct around Romney (well off and out of touch with ordinary Americans) and to link Romney to the House Budget Committee chairman's controversial budget plan (which they view as a liability for the Republicans).

Just know that this attention on Ryan will be short-lived.  He is a number two and unless the Romney campaign roll out is gaffe-prone and rife with problems, the attention will ultimately shift back to the person at the top of the ticket and the race will, in Holbrook's terms, revert to its Romney equilibrium.

2. VP selection bounce
From FHQ's perspective, the best thing about the Ryan announcement occurring when it did is that the race will have a full two weeks before the Republican Convention in which to look at polls. That's a plus for a couple of reasons. First, we should see an uptick in polling starting now. Formerly risk-averse polling outlets will have a reason to go into the field: to test Ryan's impact. Secondly and unlike the two 2008 selections, we will have some time between the VP announcement and the conventions in which to -- admittedly only partially -- tease out the impact of each event. Both the Biden and Palin selections happened on the eves of their respective conventions. And in the Palin case that made it very difficult to discern whether or how much of McCain's early September jump in the polls was convention effect or VP effect.

As Jonathan Bernstein rightly pointed out, this VP bounce is a built-in part of the process. The structural impact of a vice presidential selection is approximately two points nationally. On the state level, however, that may be felt slightly differently. It isn't necessarily uniform across states. This is particularly true of the vice presidential nominee's home state. The Palin selection, for example, turned scantily-polled Alaska from a surprisingly competitive state to a rock-red state (a position the Last Frontier would likely have ended up in on election day anyway). Ryan's Wisconsin has been a comfortably blue state (Lean Obama throughout), but that consistency may bely the fact that the state hovers close to the breaking point in the FHQ metric between Lean and Toss Up (Obama) state. This could also potentially aid the Republican ticket in other midwest/blue-collar states.

But again, let's wait and see how the polls react to that instead of speculating.

Regardless, Romney will enjoy something of a bump out of the Ryan rollout, but we'll have to wait and see how exactly that manifests itself and where (in terms of location and groups of voters).



Are you following FHQ on Twitter, Google+ and Facebook? Click on the links to join in.
Posted by Josh Putnam at 3:30 PM 0 comments
Labels: , , , , ,
Share |

Friday, August 10, 2012

The Electoral College Map (8/10/12)

Iowa is underpolled.

No other state has simultaneously proven as consistently competitive yet has been so sporadically surveyed in 2012. Other than a double digit Obama lead in a Public Policy Polling survey in May and another from the firm in mid-July (+5 Obama), the Hawkeye state has shown only a slight overall margin in either direction, but never anything more three points. In fact, it could be argued that what is driving the Obama lead in the current FHQ averages is the presence of those two PPP surveys. Without them, Obama's advantage -- 1.61 points -- disappears and becomes a small Romney lead -- 0.71 points.

There are a couple of take home messages from this:
1) Iowa is very much in the toss up category and deserving of the attention of both campaigns.

2) Iowa is also deserving of much more polling than it has received given how close the Hawkeye state appears to be. FHQ suspects this will change moving forward and especially after the two conventions are concluded.

It isn't fair to remove the PPP surveys altogether -- It is data after all. -- but they do appear to be out of step with the scant other polling that has been conducted in Iowa in 2012. Context, context, context. [Recall also, that those past polls are discounted in the averages by FHQ's weighting scheme.]

New State Polls (8/10/12)
State
Poll
Date
Margin of Error
Sample
Obama
Romney
Undecided
Poll Margin
FHQ Margin
Iowa
8/8
+/- 4.5%
500 likely voters
44
46
6
+2
+1.61

Polling Quick Hits:
Iowa:
This new Rasmussen poll of Iowa looks a lot like the last one the firm released: Romney ahead, but within the margin of error (something we have seen of Obama's numbers outside of the PPP polls of the state, FHQ should add). The bottom line is that this new one is consistent with the previous polling in Iowa. Nothing more, nothing less.


Strategically, the Obama lead in the electoral college (based on the FHQ weighted averages) has been tenuous at best. But the fact that most of the toss up states broke -- however slightly -- in the president's direction jibed well with our understanding of states quadrennially slotted into those positions. They tend to break toward one candidate or the other relative to the baseline set by the previous election. But in a closer election, it more often becomes a matter of trying to determine where to draw the line differentiating between the toss ups. While Obama has had his lead in the national polls balloon this week, polling updates in Colorado and Iowa have simultaneously pushed those states closer to Florida and the blue/red line between the two candidates' groups of states.

And really, that isn't that much of a change in the outlook. That Ohio-through-Florida group of states in the Electoral College Spectrum below is the same group of states -- give or take a handful of others -- that will be talked about between now and election day in November. Each of those are within or right at 4 points right now. In other words, that blue/red line can be drawn at various points throughout that group of states, but unless all of them break for Romney, Obama will be reelected. [That assumes that the order of those states remains the same. Romney could also win with all those states minus Iowa or all of those states minus Colorado. Oddly enough, those are the two states that have drawn closer this week with the introduction of new polling data.]

The Electoral College Spectrum1
RI-4
(7)2
NJ-14
(160)
NH-4
(257)
AZ-11
(167)
MS-6
(55)
HI-4
(11)
WA-12
(172)
OH-183
(275/281)
MT-3
(156)
ND-3
(49)
NY-29
(40)
MN-10
(182)
VA-13
(288/263)
GA-16
(153)
AL-9
(46)
VT-3
(43)
NM-5
(187)
CO-9
(297/250)
WV-5
(137)
KY-8
(37)
MD-10
(53)
CT-7
(194)
IA-6
(303/241)
IN-11
(132)
KS-6
(29)
CA-55
(108)
OR-7
(201)
FL-29
(332/235)
SC-9
(121)
AK-3
(23)
MA-11
(119)
PA-20
(221)
NC-15
(206)
LA-8
(112)
OK-7
(20)
IL-20
(139)
WI-10
(231)
TN-11
(191)
NE-5
(104)
ID-4
(13)
DE-3
(142)
NV-6
(237)
MO-10
(180)
AR-6
(99)
WY-3
(9)
ME-4
(146)
MI-16
(253)
SD-3
(170)
TX-38
(93)
UT-6
(6)
1 Follow the link for a detailed explanation on how to read the Electoral College Spectrum.

2 The numbers in the parentheses refer to the number of electoral votes a candidate would have if he won all the states ranked prior to that state. If, for example, Romney won all the states up to and including Ohio (all Obama's toss up states plus Ohio), he would have 272 electoral votes. Romney's numbers are only totaled through the states he would need in order to get to 270. In those cases, Obama's number is on the left and Romney's is on the right in italics.

3 Ohio is the state where Obama crosses the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line.

With just one new poll out before this post went live the odds were low that any state would be added or stricken from the Watch List; especially when it was one that was not already on it. That was certain the case with Iowa. The average margin contracted some, but not enough to add it to the Watch List. As of now there are no states on the list that are in range of a move into another candidate's tally of states other than North Carolina (and if anything the Tarheel state is trending away from that distinction).

The Watch List1
State
Switch
Connecticut
from Strong Obama
to Lean Obama
Georgia
from Strong Romney
to Lean Romney
Michigan
from Lean Obama
to Toss Up Obama
Nevada
from Lean Obama
to Toss Up Obama
New Hampshire
from Toss Up Obama
to Lean Obama
New Mexico
from Strong Obama
to Lean Obama
North Carolina
from Toss Up Romney
to Toss Up Obama
Ohio
from Toss Up Obama
to Lean Obama
West Virginia
from Strong Romney
to Lean Romney
Wisconsin
from Lean Obama
to Toss Up Obama
1 Weighted Average within a fraction of a point of changing categories.

Please see:



Are you following FHQ on Twitter, Google+ and Facebook? Click on the links to join in.
Posted by Josh Putnam at 2:00 PM 0 comments
Labels: , , , ,
Share |

Wednesday, August 8, 2012

The Electoral College Map (8/8/12)

New today were five polls from three states. While none of them really shook up the existing FHQ weighted averages, the continued two-day progression across surveys of Colorado drew the most reaction.

New State Polls (8/8/12)
State
Poll
Date
Margin of Error
Sample
Obama
Romney
Undecided
Poll Margin
FHQ Margin
Colorado
7/31-8/6
+/- 2.6%
1463 likely voters
45
50
4
+5
+2.76
Virginia
8/7
+/- 4.5%
500 likely voters
48
46
3
+2
+3.01
Virginia
7/31-8/6
+/- 2.6%
1412 likely voters
49
45
5
+4
--
Wisconsin
8/2-8/5
+/- 2.9%
1188 likely voters
50
45
5
+5
+5.92
Wisconsin
7/31-8/6
+/- 2.6%
1428 likely voters
51
45
4
+6
--

Polling Quick Hits:
Colorado:
I'm entirely sympathetic to the charge that the Quinnipiac/NYT/CBS poll released out of the Centennial state today is an outlier. Look, Sesame Street -- as I recall (my kids are well past enjoying those sorts of shows) -- devotes quite a bit of time to teaching kids to choose the item or items out of a group that doesn't/don't match the others. It is a basic skill. And yes, this poll does appear to be an outlier on the surface. None of the surveys that have been in the field in Colorado in 2012 have shown Romney ahead; tied with Obama a few times, but not ahead. The president's share of support in the poll (45%) is in line with others there, but the Romney share (50%) is definitely a high water mark for the governor thus far.

...by three percentage points.

But that is just polling variability -- broader than we have witnessed in Colorado for Romney at this point, but polling variability nonetheless. Again, the true test is to see whether the same sort of result repeats itself in subsequent polling. Of course, that polling is likely to show even more variability across the board with a vice presidential announcement and a couple of conventions.

Virginia:
The surveys out of Virginia today were more of the same. The president leads there, but it is a marginal/competitive edge. The new Rasmussen poll was an almost exact replica of the firm's July poll in the Old Dominion. The new Q-poll, while being within range of other recent polls, departs from the last poll conducted in the commonwealth which showed a tie. Romney's share is largely the same, but Obama's grew by five points over a poll that was in the field just a couple of weeks ago. FHQ won't make much of this because, again, one can fine quirks in just about any poll, and this is just polling noise. In the aggregate the noise favored Romney in Colorado and Obama in Virginia.

Wisconsin:
The two new Wisconsin polls, well, what's really to say? Both were mirror images of each other and were similar to other recent surveys of the Badger state. Call Colorado too hot, Virginia too cold and Wisconsin just right for Quinnipiac. All three states overall, stayed pretty much where they started in the FHQ weighted averages.


That means no change to the map and only a small shift on the Electoral College Spectrum. Colorado and Virginia switch places without fundamentally changing the calculus of how each candidate gets to 270 electoral votes. Obama could do without either state as long as he holds Ohio and New Hampshire,  and Romney needs both barring an uncharacteristic win -- assuming he doesn't swing Virginia and Colorado --  in one of the Lean Obama states. That isn't to suggest that it won't happen, but rather that it is unlikely that one (Lean Obama state flipping) happens without the other (Romney winning in most of or sweeping the Toss Up states).

The Electoral College Spectrum1
RI-4
(7)2
NJ-14
(160)
NH-4
(257)
AZ-11
(167)
MS-6
(55)
HI-4
(11)
WA-12
(172)
OH-183
(275/281)
MT-3
(156)
ND-3
(49)
NY-29
(40)
MN-10
(182)
VA-13
(288/263)
GA-16
(153)
AL-9
(46)
VT-3
(43)
NM-5
(187)
CO-9
(297/250)
WV-5
(137)
KY-8
(37)
MD-10
(53)
CT-7
(194)
IA-6
(303/241)
IN-11
(132)
KS-6
(29)
CA-55
(108)
OR-7
(201)
FL-29
(332/235)
SC-9
(121)
AK-3
(23)
MA-11
(119)
PA-20
(221)
NC-15
(206)
LA-8
(112)
OK-7
(20)
IL-20
(139)
WI-10
(231)
TN-11
(191)
NE-5
(104)
ID-4
(13)
DE-3
(142)
NV-6
(237)
MO-10
(180)
AR-6
(99)
WY-3
(9)
ME-4
(146)
MI-16
(253)
SD-3
(170)
TX-38
(93)
UT-6
(6)
1 Follow the link for a detailed explanation on how to read the Electoral College Spectrum.

2 The numbers in the parentheses refer to the number of electoral votes a candidate would have if he won all the states ranked prior to that state. If, for example, Romney won all the states up to and including Ohio (all Obama's toss up states plus Ohio), he would have 272 electoral votes. Romney's numbers are only totaled through the states he would need in order to get to 270. In those cases, Obama's number is on the left and Romney's is on the right in italics.

3 Ohio is the state where Obama crosses the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line.

The Watch List remained unchanged by Wednesday's polls.

The Watch List1
State
Switch
Connecticut
from Strong Obama
to Lean Obama
Georgia
from Strong Romney
to Lean Romney
Michigan
from Lean Obama
to Toss Up Obama
Nevada
from Lean Obama
to Toss Up Obama
New Hampshire
from Toss Up Obama
to Lean Obama
New Mexico
from Strong Obama
to Lean Obama
North Carolina
from Toss Up Romney
to Toss Up Obama
Ohio
from Toss Up Obama
to Lean Obama
West Virginia
from Strong Romney
to Lean Romney
Wisconsin
from Lean Obama
to Toss Up Obama
1 Weighted Average within a fraction of a point of changing categories.

Please see:



Are you following FHQ on Twitter, Google+ and Facebook? Click on the links to join in.
Posted by Josh Putnam at 5:30 PM 0 comments
Labels: , , , , , ,
Share |
Subscribe to: Posts (Atom)