Wednesday, October 3, 2012

The Links (10/3/12): 2012 Debate Season

A couple of things to set the proper context for tonight's opening to debate season:

1) Tom Holbrook, indispensable as always, on the (limited) impact of debates over time.

1a) John Sides and Jon Bernstein also throw cold water on the idea of debates as game changing events in the campaign.

2) Emma Roller at NPR has another take.

[Eh, I'll side with the political scientist on this one.]

3) Alex Speigel had a nice story on question dodging in debates on Morning Edition this morning.

4) If you have the time before the debate tonight, the PBS documentary, Debating Our Destiny, is always a worthwhile view.

NOTE: FHQ will try to embed a Twitter conversation in a post for the debate this evening. I'll be most active over there, but will try to pull double duty and respond to any comments that may come up over here in response. If you are on Twitter either reply directly to one of my tweets and/or use the #fhqdebate hashtag to insure that your comment/response is included in the feed.


Are you following FHQ on TwitterGoogle+ and Facebook? Click on the links to join in.

Tuesday, October 2, 2012

The Electoral College Map (10/2/12)

On the eve of the first presidential debate between President Obama and Governor Romney, we were treated to new survey data from eight states. And it was a pretty good day for the president with comfortable leads confirmed in several deep blue states, a slight edge in the one Romney toss up state (North Carolina) and a closer than expected Romney lead in Missouri.

New State Polls (10/2/12)
State
Poll
Date
Margin of Error
Sample
Obama
Romney
Undecided
Poll Margin
FHQ Margin
Florida
9/27-9/30
+/- 4.0%
600 likely voters
46
43
8
+3
+1.30
Maryland
9/25-9/27
+/- 3.5%
804 likely voters
57
34
--
+23
+21.05
Missouri
9/25-9/27
+/- 2.9%
1145 likely voters
45
48
6
+3
+7.03
Nevada
9/25-9/27
+/- 3.1%
1078 likely voters
53
42
3
+11
+4.88
New Mexico
9/27
+/- 4.5%
500 likely voters
51
40
2
+11
+10.17
North Carolina
9/29-10/1
+/- 4.2%
573 likely voters
49
47
2
+2
+0.89
Rhode Island
9/26-9/29
+/- 4.38%
501 likely voters
57.3
33.1
7.4
+24.2
+25.30
Virginia
9/19-9/28
+/- 4.0%
589 likely voters
47
39
10
+8
+3.22

Polling Quick Hits:
Florida:
That we get new data out of Florida on a daily basis makes more sense than the rather high frequency of survey work coming out of reliably blue Massachusetts, but the two states do have one thing in common. The individual polling results aggregated are more often than not confirming/calibrating rather than of the sort that fundamentally changes the outlook of the race. The difference is that there is and has been a tight race in Florida all along and a clear Obama advantage in Massachusetts. The new Suffolk poll of the Sunshine state fits that bill. At Obama +3, it is right in line with the range of polling results to have emerged from the state.

Maryland:
FHQ is glad to have the data, but this is merely a confirming polling result. Nothing to see here folks. With or without the new poll from the Baltimore Sun, Maryland is in the Obama column.

Missouri:
We cannot keep coming back to the post-Akin noise argument about Missouri polling. That said, the polling has varied there, but has calmed down since the congressman made his "shut that down" comments in August. Though the polling has been relatively light in the Show Me state, the Romney lead there is in the 3-7 point range with the two of the last three polls indicating something closer to the low end of that range. The FHQ weighted averaging formula currently has the state at the top end of that range, firmly within the Lean Romney category.

Nevada:
My first thought in seeing the Obama +11 result from We Ask America out of the Silver state was that it reminded me of the last minute surge toward Obama in the state in polling and on election day. In the 2012 context, however, that margin feels a little high compared to other recent surveys there, and on top of that, the 2008 shift occurred later in the race than where we are now. October polls can be a bit noisy and this one seemingly settles nicely into that area: noise.

New Mexico:
Another day and another low double digit polling margin in the Land of Enchantment. This time the new information is from Rasmussen. New Mexico, close in 2000 and 2004, just has not materialized as a toss up (or even lean) state in 2012. In that way, it has been more like 2008 than either of those previous elections. It has swung back to the GOP some, but not nearly enough to bring the state into a more competitive light.

North Carolina:
Any (October) day that has a poll showing Obama ahead in North Carolina is -- zero-sum -- a win from the Obama perspective and a loss on the Romney side. That isn't to say that the president will win the 15 North Carolina electoral votes. In fact, Romney has consistently held the advantage in the FHQ averages. Instead, the importance of North Carolina is figuring into the strategic calculus of the race. Romney resources expended there are resources that are not being spent in states that Romney has to turn red over the last handful of weeks of the campaign.

Rhode Island:
See Maryland, but with the caveat that this is the first survey conducted in Rhode Island gauging respondents' opinions on the presidential horse race. The resulting data, however, does little to alter the thinking here at FHQ regarding the state of the race in the Ocean state. It is blue and is way out on the Obama end of the Electoral College Spectrum. ...still.

Virginia:
Further south in Virginia, Roanoke College has posted a poll finding Obama up eight. The key here in the "why +8" is that while Obama's share of support in the survey is consistent with other recent polling in the Old Dominion (not to mention the FHQ weighted average level of support), the Romney share of response is well below his established weighted average level of support. Five points lower. For context, no poll has shown Romney below 40% in Virginia since -- and this was the only other time -- Marist/NBC found the governor 17 points behind the president in early March. The eight point margin from Roanoke is perhaps a bit high compared to other recent polling in the state. ...and it is almost completely a function of Romney's position in it.


Methodological changes and resultant map/spectrum shifts aside, the same picture of the race that existed last week seems to be reiterated this week as the race heads into debate season. There were some minor changes at the margins of the three FHQ categories of states when the cutpoints were lowered, but the same story continues to be told. Obama has the lead in all of the toss up states save North Carolina. Those are not insurmountable advantages, but they have been consistent. And that must be the most troubling thing to Romney campaign. Heading into the debates, Romney should gain simply by being on the same stage with the president, but that may not necessarily be the type of, sigh, "game change" the governor will need to alter the course of the race.

Needless to say, the map remains unchanged today and the only shift on the Electoral College Spectrum involves Maryland and Massachusetts changing positions for the second consecutive day. The remaining seven states in which new information was introduced held the line relative to a day ago.

The Electoral College Spectrum1
VT-3
(6)2
WA-12
(158)
NV-6
(257)
MO-10
(166)
ND-3
(55)
RI-4
(10)
NJ-14
(172)
OH-183
(275/281)
MT-3
(156)
KY-8
(52)
NY-29
(39)
CT-7
(179)
VA-13
(288/263)
IN-11
(153)
AL-9
(44)
HI-4
(43)
NM-5
(184)
IA-6
(294/250)
GA-16
(142)
KS-6
(35)
MD-10
(53)
MN-10
(194)
CO-9
(303/244)
SC-9
(126)
AR-6
(29)
MA-11
(64)
OR-7
(201)
FL-29
(332/235)
TX-38
(117)
AK-3
(23)
IL-20
(84)
PA-20
(221)
NC-15
(206)
LA-8
(79)
OK-7
(20)
CA-55
(139)
MI-16
(237)
SD-3
(191)
NE-5
(71)
ID-4
(13)
ME-4
(143)
WI-10
(247)
AZ-11
(188)
WV-5
(66)
WY-3
(9)
DE-3
(146)
NH-4
(251)
TN-11
(177)
MS-6
(61)
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 281 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 same is true for the Watch List with one exception. New Hampshire rounded up to a 5 point margin yesterday and was still technically eligible for the list. Today, however, the gap was just a shade above 5 points and the Granite state goes off the list as soon as it came back on. On the list or not, New Hampshire is still among that quartet of states (Michigan, Nevada and Wisconsin are the others.) that Romney would like to bring into play rather than fighting it out in a constrained group of toss ups where there is little or no margin for error.

The Watch List1
State
Switch
Indiana
from Strong Romney
to Lean Romney
Iowa
from Toss Up Obama
to Lean Obama
Minnesota
from Strong Obama
to Lean Obama
Montana
from Strong Romney
to Lean Romney
Nevada
from Lean Obama
to Toss Up Obama
North Carolina
from Toss Up Romney
to Toss Up Obama
Ohio
from Toss Up Obama
to Lean Obama
Virginia
from Toss Up Obama
to Lean Obama
1 The Watch list shows those states in the FHQ Weighted Average within a fraction of a point of changing categories. The List is not a trend analysis. It indicates which states are straddling the line between categories and which states are most likely to shift given the introduction of new polling data. Nevada, for example, is close to being a Toss Up Obama state, but the trajectory of the polling there has been moving the state away from that toss up distinction.

Please see:


Are you following FHQ on TwitterGoogle+ and Facebook? Click on the links to join in.

Monday, October 1, 2012

The Electoral College Map (10/1/12)

With just 36 days until election day, ten new polls were made public from nine states. The picture looked largely the same with Obama stretching his leads across the majority of these polls in blue states. The lone holdout was North Carolina, which has remained in the Romney column albeit in a somewhat tenuous fashion. In other words, there was no Obama lead to strengthen, but Romney's hold on the Tarheel state continues to be rather small.

New State Polls (10/1/12)
State
Poll
Date
Margin of Error
Sample
Obama
Romney
Undecided
Poll Margin
FHQ Margin
Colorado
9/25-9/27
+/- 2.8%
1273 likely voters
49
46
4
+3
+2.77
Florida
9/29-9/30
+/- 3.4%
914 likely voters
49
48
3
+1
+1.25
Iowa
9/25-9/27
+/- 2.8%
1273 likely voters
48
44
7
+4
+3.04
Massachusetts
9/26-9/28
+/- 4.4%
504 likely voters
60
32
6
+28
+20.51
Michigan
9/25-9/27
+/- 3.1%
1064 likely voters
52
40
7
+12
+6.04
New Hampshire
9/27-9/30
+/- 4.0%
588 likely voters
54
39
4
+15
+5.00
New Mexico
9/25-9/27
+/- 2.85%
1258 likely voters
51
41
5
+11
+10.06
North Carolina
9/27-9/30
+/- 3.1%
981 likely voters
48
48
4
0
+0.87
North Carolina
9/28-9/30
+/- 4.0%
600 likely voters
46
50
3
+4
--
Ohio
9/27-9/30
+/- 3.3%
897 likely voters
49
45
7
+4
+3.83

Polling Quick Hits:
Colorado:
Rare are the Colorado polls that have a wide lead for either candidate and/or find the president below about the 48% mark post-convention. Prior to convention season that was also the case in terms of the former. The lead just hasn't changed that much. If anything it has ever so slightly contracted over that time. As for Obama's share of support, it has grown in the time since the convention, but again, only slightly; from the 46-47% range to 48% now. Meanwhile, Mitt Romney has also seen some growth in the Centennial state overall as well. His share of support in post-convention polling has consistently hovered in the 44-47% area. That doesn't differ substantially from the pre-convention polling, but some of the lower 40s numbers that were there before September now appear to be gone as undecideds weigh in.

Florida:
Changes (October 1)1
StateBeforeAfter
IndianaLean RomneyStrong Romney
MinnesotaLean ObamaStrong Obama
MontanaLean RomneyStrong Romney
NevadaToss Up ObamaLean Obama
New HampshireToss Up ObamaLean Obama
1 These changes and those to the map and Electoral College Spectrum below reflect not a poll-based change, but a change in the thresholds defining the three FHQ categories of states. To incorporate the fact that leads are harder to overcome for the trailing candidate in a given state the closer to election day the campaign gets, the cutpoints have been lowered. A toss up state is now a state with an FHQ weighed average below 4%, a lean state is one with and average of 4-9% and any state above the 9% barrier is a strong state for that candidate.
There really isn't that much to say about yet another one point lead -- no matter in which direction it goes -- in Florida. The Sunshine state continues to tip toward the president, but by the smallest of margins. The one silver lining from the Romney perspective in Florida is that a one point lead -- even for the president -- is a good number heading into the debate later this week. That gap is smaller than most recent polls have indicated the race is in Florida. However, it does represent a modest positive shift toward the president in the time since the last Gravis survey there.

Iowa:
Though perhaps a shade smaller than some of the other polls of Iowa following the conventions, this We Ask America survey seems to nail the state of the race in the Hawkeye state. The raw average of the eight post-convention polls conducted in Iowa find a margin of 3.88 points (compared to a four point margin in the survey). Additionally, the FHQ weighted averages of the two candidates' shares of support are 47.6% (Obama) and 44.5% (Romney). That compares to the 48% share Obama held and the 44% of responses Governor Romney garnered. If Iowa has settled in post-convention and pre-debate, then Obama is ahead by about three to four points.

Massachusetts:
FHQ won't spend much time here. In our parlance of late, this WBUR survey is a confirming poll; confirming that the president's lead in the Bay state is safe.

Michigan:
In the Great Lakes state, the latest We Ask America survey is the ninth post-convention poll. Of those nine, seven have found Obama north of the 50% threshold. The best Romney has been able to muster in any of those polls is 46% and that is/was four points above where the FHQ weighted average of the governor's support in Michigan is. The overall 6% Obama lead in the state may be a smidgen high, but that is a lot of ground for Romney to make up over the last month of the campaign. Yes, there are three face-to-face debates ahead, but rarely have they proven to be transformative enough to shift six points in the aggregate.

New Hampshire:
James Pindell of WMUR said of this Granite State Poll that "either it is wildly off or we will look back five weeks from now on Election Day and point to this date as the moment we knew Mitt Romney lost New Hampshire." Perhaps, but in the context of the extant polling in the Granite state, this one is an outlier. That does not mean that the wildly off description is apt however. The president has pushed above 50% in a handful of recent New Hampshire polls, but a 54% share is almost as inflated as Romney's 39% share understates where the governor has been in most New Hampshire polling. The FHQ weighted average smooths out some of that volatility and has the gap at about 5% in Obama's favor heading into the debates.

New Mexico:
Even with former Governor Gary Johnson included in the new We Ask America survey of New Mexico, the berth between Governor Romney and the president is still right at the double digit mark. New Mexico has swung back toward the Republicans since 2008, but that it has swung back to just +10 Obama is indicative of just how much ground Romney and the Republicans had to make up in the Land of Enchantment. To this point, New Mexico has been resilient and that does not look to be changing any time soon.

North Carolina:
For the most part, the post-convention polling in North Carolina -- while shifting some toward the president -- has continued to show Mitt Romney ahead in the state, albeit narrowly. Using the same principle in North Carolina as was (will be) used (in future analyses) with the WMUR poll of New Hampshire above, we can throw out the one Civitas/Survey USA poll that showed Romney ahead by ten points. FHQ does that to illustrate that Romney has consistently polled in the 44-50% range while Obama has been in a slightly tighter 44-49% area. Post-convention, then, the candidates are oscillating in the same territory, but with Romney maintaining a very small edge. If one looks at Obama peaking during the time following the conventions, even though he found several individual polling leads, he was unable to surpass Romney in the averages. If that's Obama's best, then North Carolina is likely to be on the wrong side of the partisan line in the electoral college count come election day.

Ohio:
The four point edge the president has in the latest Public Policy Polling survey of Ohio is narrower than a number of other polls recently conducted in the Buckeye state, but as was the case with the Iowa poll discussed above, this one is right on. The unadjusted, post-convention polling margin average in Ohio is about 4.6%, and the FHQ weighted average margin has now risen to 3.8%. The average margin and this poll are closer to one another than the weighted average share of support and the responses in the PPP survey. Compared to the averages, PPP is about two points above where FHQ has both candidates.


As the footnote to the "Changes" table above indicates, FHQ has lowered the thresholds dividing each of the three categories of states. This is not a big shift overall, but it does have some important implications. I'll move past the lean-to-strong shifts and focus on the two former Toss Up Obama states that moved into the Lean Obama category. FHQ has come back to the quartet of states on the Toss Up/Lean line time and again. What the averages look like in Michigan, Wisconsin, Nevada and New Hampshire has a direct bearing on the race to 270 electoral votes. The extent to which Romney can force Obama to play in any of those states siphons off resources that the president could use in defending larger bargaining chips like Florida, Ohio and Virginia. On the other hand, if, as is the case on October 1, all four of those states are (and continue to be) tipped in the president's direction by something on or above 4 points, then the calculus changes, becoming a much heavier lift for Romney. If those four states are (hypothetically) off the board, then the governor has to run the table in the remaining toss up states. That isn't impossible, but it is fairly unprecedented in modern presidential elections.

The Electoral College Spectrum1
VT-3
(6)2
WA-12
(158)
NV-6
(257)
MO-10
(166)
ND-3
(55)
RI-4
(10)
NJ-14
(172)
OH-183
(275/281)
MT-3
(156)
KY-8
(52)
NY-29
(39)
CT-7
(179)
VA-13
(288/263)
IN-11
(153)
AL-9
(44)
HI-4
(43)
NM-5
(184)
IA-6
(294/250)
GA-16
(142)
KS-6
(35)
MA-11
(54)
MN-10
(194)
CO-9
(303/244)
SC-9
(126)
AR-6
(29)
MD-10
(64)
OR-7
(201)
FL-29
(332/235)
TX-38
(117)
AK-3
(23)
IL-20
(84)
PA-20
(221)
NC-15
(206)
LA-8
(79)
OK-7
(20)
CA-55
(139)
MI-16
(237)
SD-3
(191)
NE-5
(71)
ID-4
(13)
ME-4
(143)
WI-10
(247)
AZ-11
(188)
WV-5
(66)
WY-3
(9)
DE-3
(146)
NH-4
(251)
TN-11
(177)
MS-6
(61)
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 281 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.

Shifting focus to the Watch List, the cutpoint changes also alter the landscape of states on the verge of changing categories. Off go Georgia, Michigan, New Mexico and Wisconsin and on come Iowa, Ohio and Virginia. If this is the discussion we are going to have between now and election day -- those new additions threatening to jump in to the Lean Obama category -- then perhaps we should move on to talking about which party will control the Senate. However, if could also be that we are entering a point in the polling of this race -- pre-debate -- where Obama support has crested, post-convention, and we will begin to see some narrowing as a result. The Iowa and Ohio polls above, for example, might have been in line with where FHQ has the races in those respective states, but in both cases were smaller leads for the president than the bulk of recent polling conducted in each state.

The Watch List1
State
Switch
Indiana
from Strong Romney
to Lean Romney
Iowa
from Toss Up Obama
to Lean Obama
Minnesota
from Strong Obama
to Lean Obama
Montana
from Strong Romney
to Lean Romney
Nevada
from Lean Obama
to Toss Up Obama
New Hampshire
from Lean Obama
to Toss Up Obama
North Carolina
from Toss Up Romney
to Toss Up Obama
Ohio
from Toss Up Obama
to Lean Obama
Virginia
from Toss Up Obama
to Lean Obama
1 The Watch list shows those states in the FHQ Weighted Average within a fraction of a point of changing categories. The List is not a trend analysis. It indicates which states are straddling the line between categories and which states are most likely to shift given the introduction of new polling data. Nevada, for example, is close to being a Toss Up Obama state, but the trajectory of the polling there has been moving the state away from that toss up distinction.

Please see:


Are you following FHQ on TwitterGoogle+ and Facebook? Click on the links to join in.