Friday, September 23, 2016

The Electoral College Map (9/23/16)



FHQ finally got an update up this week after a two week absence. That return has prompted some questions. Chief among them has been something along the lines of "Why aren't Florida, North Carolina and Ohio red?" Some skipped that and went straight for "you're wrong" while others went for poll "cherrypicking". On the former accusation, perhaps. The simple truth is that North Carolina and Ohio are close calls at the moment. It is exactly those close calls that tend to separate the various models in the end (see Florida, 2012). As for cherrypicking, well, it is hard to cherrypick polls when the whole aim is to include every publicly available poll out there. Like the "wrong" charge, this "cherrypicking" one ultimately comes back to the methodology.

Here are a few thoughts on that Storified from Twitter:


To the day's polls...

New State Polls (9/23/16)
State
Poll
Date
Margin of Error
Sample
Clinton
Trump
Undecided
Poll Margin
FHQ Margin
Arkansas
9/15-9/17
+/- 3.4%
831 likely voters
34
55
4
+21
+14.18
Georgia
9/20-9/22
+/- 4.0%
600 likely voters
38
44
12
+6
+2.04
Nevada
9/10-9/19
+/- 4.9%
400 likely voters
42
42
4
+/-0
+0.27
North Carolina
9/10-9/19
+/- 4.9%
400 likely voters
44
40
2
+4
+1.01
Ohio
9/10-9/19
+/- 4.9%
400 likely voters
39
41
4
+2
+0.69
Pennsylvania
9/10-9/19
+/- 4.9%
400 likely voters
46
38
3
+8
+5.94


Polling Quick Hits:
A slower day to end the work week. Two traditional red states saw new polls as well as some battleground states surveys from GQR.

Arkansas:
Changes (September 23)
StateBeforeAfter
ArkansasLean TrumpStrong Trump
Polling has been light in the Natural state, but the picture has been pretty clear: Trump is ahead, but underperforming Mitt Romney from 2012. While that is still true in the latest Hendrix survey, it is less acute. The former first lady in the state is hovering right around where Obama was in the state four years ago and Trump is within shouting distance now of Romney there. In other words, if this poll is accurate, Arkansas looks normal rather than evidence of any fundamental shift across the whole map.


Georgia:
Compared to the last JMC poll from August in the Peach state, the two major party candidates have swapped positions. Now, it is Trump ahead. The survey provides a little more evidence that Georgia is tighter than 2012, but still a red state. Yes, the margin is still close-ish here, but it is growing as Trump continues his sweep of the September polls.


Nevada:
The first of today's GQR battleground polls in Nevada shows exactly what the firm did there back during its original June wave: a tie. The Silver state has slipped back into the red side of the partisan line, but only just there after a string of pro-Trump polls throughout September. This GQR survey is the first break in that streak, but serves to keep the underlying FHQ average in Nevada close. Close but favoring Trump for now, though.


North Carolina:
GQR also weighed in in North Carolina. The Democratic firm found the former Secretary of State up four, but that was down from the ten point advantage she had in their last poll in the Tar Heel state in June. Where that one may have been an outlier, this one is not, at least not as much. The established range in North Carolina has both Clinton and Trump bouncing around between 40 and 45 percent from poll to poll. The lead may change hands, but more often than not both candidates are within that range. That sort of clustering will produce a close race more often than not.


Ohio:
The same sort of story from North Carolina can extend to Ohio as well. The difference is that the two candidates are not sharing the same range. In the Buckeye state, Trump has found his support in the 40-44 percent range during September while Clinton has lagged in a 37-41 percent window for most polls in that same period. That is right where GQR plots the race right now. Now, obviously, that points to some Trump advantage in the state; an advantage that is not reflected here at FHQ. The Ohio average is slowly but surely tracking toward zero. If Trump's lead holds up in future polls, then that shift across the partisan line will occur. But that has not happened as of yet.


Pennsylvania:
Yes, yes, this GQR survey is one from a Democratic-leaning firm. Yet, there is it in black and white that Clinton's lead in the state -- across most polls -- has proven more durable than in other states throughout this September swoon of hers. Trump does not have to win Pennsylvania, but he will have to reach into the Lean Clinton area to pick off a state or two to get to 270. Pennsylvania, with its 20 electoral votes, is the quickest route to that goal, but also does not look reachable at this time.


--
The addition of these polls did little to shake things up here at FHQ. North Carolina barely inched off the Watch List but remains quite close (though still favoring Clinton). Arkansas' is off the list too. Meanwhile, Arkansas and Pennsylvania shuffled around on the Spectrum. Arkansas is still very much a red state but looks a little redder now, and Pennsylvania slid up a couple of cells deeper into blue territory on the Spectrum. The latter is part of a tightly knot quartet including Colorado, Maine and Virginia.





The Electoral College Spectrum1
HI-42
(7)
NJ-14
(175)
NH-43
(269 | 273)
TX-38
(155)
TN-11
(56)
MD-10
(17)
DE-3
(178)
RI-43
(273 | 269)
MS-6
(116)
AR-6
(45)
MA-11
(28)
NM-5
(183)
FL-29
(302 | 265)
SC-9
(110)
SD-3
(39)
VT-3
(31)
MN-10
(193)
NC-15
(317 | 236)
AK-3
(101)
ND-3
(36)
CA-55
(86)
WI-10
(203)
OH-18
(335 | 221)
UT-6
(98)
ID-4
(33)
NY-29
(115)
MI-16
(219)
NV-6
(203)
KS-6
(92)
NE-5
(29)
IL-20
(135)
VA-13
(232)
IA-6
(197)
IN-11
(86)
OK-7
(24)
WA-12
(147)
PA-20
(252)
AZ-11
(191)
MT-3
(75)
WV-5
(17)
CT-17
(154)
ME-4
(256)
GA-16
(180)
KY-8
(72)
AL-9
(12)
OR-7
(161)
CO-9
(265)
MO-10
(164)
LA-8
(64)
WY-3
(3)
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 or she won all the states ranked prior to that state. If, for example, Trump won all the states up to and including New Hampshire (all Clinton's toss up states plus New Hampshire), he would have 273 electoral votes. Trump's numbers are only totaled through the states he would need in order to get to 270. In those cases, Clinton's number is on the left and Trumps's is on the right in bold italics.


To keep the figure to 50 cells, Washington, DC and its three electoral votes are included in the beginning total on the Democratic side of the spectrum. The District has historically been the most Democratic state in the Electoral College.

3 New Hampshire and Rhode Island are collectively the states where Clinton crosses the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line. If those two states are separated with Clinton winning New Hampshire and Trump, Rhode Island, then there would be a tie in the Electoral College.



NOTE: Distinctions are made between states based on how much they favor one candidate or another. States with a margin greater than 10 percent between Clinton and Trump are "Strong" states. Those with a margin of 5 to 10 percent "Lean" toward one of the two (presumptive) nominees. Finally, states with a spread in the graduated weighted averages of both the candidates' shares of polling support less than 5 percent are "Toss Up" states. The darker a state is shaded in any of the figures here, the more strongly it is aligned with one of the candidates. Not all states along or near the boundaries between categories are close to pushing over into a neighboring group. Those most likely to switch -- those within a percentage point of the various lines of demarcation -- are included on the Watch List below.


The Watch List1
State
Switch
Colorado
from Lean Clinton
to Toss Up Clinton
Delaware
from Strong Clinton
to Lean Clinton
Indiana
from Strong Trump
to Lean Trump
Iowa
from Toss Up Trump
to Toss Up Clinton
Maine
from Lean Clinton
to Toss Up Clinton
Missouri
from Lean Trump
to Toss Up Trump
Nevada
from Toss Up Trump
to Toss Up Clinton
New Hampshire
from Lean Clinton
to Toss Up Clinton
New Jersey
from Strong Clinton
to Lean Clinton
Ohio
from Toss Up Clinton
to Toss Up Trump
Oregon
from Strong Clinton
to Lean Clinton
Pennsylvania
from Lean Clinton
to Toss Up Clinton
Rhode Island
from Lean Clinton
to Toss Up Clinton
Virginia
from Lean Clinton
to Toss Up Clinton
1 Graduated weighted average margin within a fraction of a point of changing categories.


Thursday, September 22, 2016

The Electoral College Map (9/22/16)



New State Polls (9/22/16)
State
Poll
Date
Margin of Error
Sample
Clinton
Trump
Undecided
Poll Margin
FHQ Margin
Arizona
9/12-9/14
+/- 4.5%
484 likely voters
33
37
19
+2
+1.55
California
9/12-9/14
+/- 4.3%
515 likely voters
57
30
9
+27
--
California
9/9-9/18
+/- 4.5%
1055 likely voters
47
31
4
+16
+20.50
Colorado
9/14-9/18
+/- 6.3%
350 likely voters
41
34
10
+7
--
Colorado
9/13-9/21
+/- 3.9%
644 likely voters
44
42
2
+2
+5.84
Florida
9/19-9/21
+/- 4.4%
500 likely voters
44
45
7
+2
+2.16
Georgia
9/13-9/21
+/- 3.9%
638 likely voters
40
47
4
+7
+1.76
Illinois
9/19-9/21
+/- 3.6%
700 likely voters
44.5
38.7
7.3
+5.8
+14.68
Iowa
9/13-9/21
+/- 4.0%
612 likely voters
37
44
5
+7
+0.51
Louisiana
9/15-9/17
+/- 4.4%
500 likely voters
32.6
48.9
10.8
+16.3
+13.23
Maryland
9/17-9/20
+/- 4.3%
514 likely voters
58
25
8
+33
+31.15
Nevada
9/12-9/14
+/- 4.9%
398 likely voters
40
42
10
+2
+0.29
North Carolina
9/16-9/19
+/- 3.6%
782 likely voters
41
41
5
+/-0
+0.88
Virginia
9/11-9/20
+/- 3.4%
841 likely voters
44
37
9
+7
--
Virginia
9/13-9/21
+/- 3.8%
659 likely voters
45
39
6
+6
+5.94
Washington
9/12-9/14
+/- 4.4%
505 likely voters
38
28
14
+10
+12.62
Wisconsin
9/19-9/20
+/- 3.6%
700 likely voters
44.5
38.4
4.1
+6.1
+7.03


Polling Quick Hits:
There is a lot to look at today with just 47 days until Election Day: 17 new surveys from 14 states.

Arizona:
Other than the Cronkite survey of Arizona at the end of August, this Insights West poll is the lowest share of support either candidate has garnered in the Grand Canyon state all year. Both polls share a high number of undecideds, but the bottom line, perhaps, is that this is another survey with Trump ahead but in the toss up area. His lead there has stabilized as September has not seen Trump trail Clinton in Arizona.


California:
Clinton's 47 percent in that PPIC poll jumps off the screen at first glance except that it is actually an improvement over her share of support in last PPIC poll from July. She may be lagging behind Obama in 2012 in the Golden state, but Trump continues to trail not only Clinton but Romney's pace four years ago as well. California is still blue.


Colorado:
The two polls out of the Centennial state are pretty representative of the polls over there: Clinton at or just above 40 percent and Trump moving around from the mid-30s to around 40 percent. Colorado could be close, but it will take Trump's support firming up some there to pull it off.


Florida:
Yesterday was a good day for Clinton in Florida, but the polling there has been back and forth throughout September. And there was a snap back toward Trump today in the latest Suffolk survey. Of course, while the polling has been volatile, the average margin in the Sunshine state here at FHQ has not. Clinton's advantage has been stuck in that two to three point range for a while now.


Georgia:
In the Peach state, Trump has found a consistent footing in the mid-40s; something confirmed in Quinnipiac's first survey in the state. On the other hand, Clinton looks a lot like Trump does in a number of the states he will likely need to get to and over 270. In other words, she is stuck around 40 percent and playing catch up.


Illinois:
So far in 2016, the Clinton-Trump race in Illinois has been only sporadically surveyed. It is a shame, then, that the two most recent polls -- two surveys showing a Democrat dipping into the 40s -- come from landline-only Emerson and Loras (whose only other poll was a big outlier in Iowa earlier in the summer). Look, if the other poll from Emerson today -- a Clinton +6 in Wisconsin -- is right on the mark, then it would be highly unusual for Illinois to closer. Or put it this way: if Clinton is only ahead  by around six in Illinois, then the expectation would be that Trump is leading in Wisconsin. None of the polling there has borne that out to this point.


Iowa:
Like Arizona and Georgia, Iowa has seen a string of Trump leads in the polls during September. That streak continues in the latest Quinnipiac survey of the Hawkeye state. What is more is that the near ties in August have been displaced by results that would fall in the Lean Trump category. Those are the types of polls, when strung together, that can build a durable lead. What is Trump's 40 percent problem in a number of the Lean Clinton states is Clinton's problem in the Arizona, Georgia and Iowa trio.


Louisiana:
One could focus on how Trump continues to fall short of Mitt Romney's share of the 2012 vote in states like Louisiana, but that tends to miss the point. Clinton is running well enough behind Trump that the Pelican state is still safely red; off to the far right hand column on the Electoral College Spectrum below.


Maryland:
Like Wyoming yesterday, both candidate are behind the marks set by the 2012 candidates in Maryland. But that matters less in view of the fact that the Old Line state is so far off to the Democratic side of the Spectrum.


Nevada:
Nevada is another state, like a handful above, that has been good to Trump in in September. With only one exception, the New York businessman has led every poll in that stretch but the very first one of the month. The narrow Clinton lead that developed following the conventions disappeared and has  been replaced by a narrow Trump advantage in the state. That edge can only grow so much as long as the poll margins remain as close as they are in polls like the Insights West survey.



North Carolina:
The Tar Heel state has seen no such polling streak for either candidate over the last month. Clinton's overall advantage in the state here at FHQ continues to shrink. But the range of results it a pretty tight band around something close to zero. It is fitting, then, that the new survey of the state from Siena shows just that: a tie. North Carolina, along with Iowa, Nevada and Ohio, are all within a fraction of a point of shifting across the partisan line. Unlike the other three, though, there is no clear trend toward Trump. Instead, North Carolina resembles a closer version of volatile Florida.


Virginia:
As FHQ mentioned in yesterday's update, Virginia is one of those "sticky" Lean Clinton states where the polling (and resulting average) has consistently had Clinton out front by a margin beyond the Lean/Toss Up line (in the Lean category). The two new polls of the Old Dominion from Roanoke and Quinnipiac do nothing to shake up that particular conclusion.


Washington:
The margin in the new Insights West survey looks about right in Washington if past elections are any guide. However, both candidates are collectively only pulling in the support of about two-thirds of the respondents in the poll. That has both Clinton and Trump well behind where Obama and Romney were in the Evergreen state four years ago. Washington is underpolled, but the picture gleaned from what data there is still finds the state in a comfortable area for the Democrats.


Wisconsin:
In the Badger state, just wait on other polling. The Emerson poll there may be in line with expectations, but standing alongside the other survey from the firm today, it looks off. Or something about the two surveys is mismatched. The clear picture in Wisconsin is that Clinton's cushion has dissipated, but that she has a durable, though small, advantage there.


--
There was some subtle shuffling on the Electoral College Spectrum today, but nothing that broke with the patterns established throughout this cycle. No states changed categories and the map and Spectrum remain largely the same. Colorado's addition to the Watch List and into a tight cluster of states (Maine, Pennsylvania and Virginia) along that Lean/Toss Up line.





The Electoral College Spectrum1
HI-42
(7)
NJ-14
(175)
NH-43
(269 | 273)
TX-38
(155)
LA-8
(58)
MD-10
(17)
DE-3
(178)
RI-43
(273 | 269)
MS-6
(116)
TN-11
(50)
MA-11
(28)
NM-5
(183)
FL-29
(302 | 265)
SC-9
(110)
SD-3
(39)
VT-3
(31)
MN-10
(193)
NC-15
(317 | 236)
AK-3
(101)
ND-3
(36)
CA-55
(86)
WI-10
(203)
OH-18
(335 | 221)
UT-6
(98)
ID-4
(33)
NY-29
(115)
MI-16
(219)
NV-6
(203)
KS-6
(92)
NE-5
(29)
IL-20
(135)
VA-13
(232)
IA-6
(197)
AR-6
(86)
OK-7
(24)
WA-12
(147)
ME-4
(236)
AZ-11
(191)
IN-11
(80)
WV-5
(17)
CT-17
(154)
CO-9
(245)
GA-16
(180)
MT-3
(69)
AL-9
(12)
OR-7
(161)
PA-20
(265)
MO-10
(164)
KY-8
(66)
WY-3
(3)
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 or she won all the states ranked prior to that state. If, for example, Trump won all the states up to and including New Hampshire (all Clinton's toss up states plus New Hampshire), he would have 273 electoral votes. Trump's numbers are only totaled through the states he would need in order to get to 270. In those cases, Clinton's number is on the left and Trumps's is on the right in bold italics.


To keep the figure to 50 cells, Washington, DC and its three electoral votes are included in the beginning total on the Democratic side of the spectrum. The District has historically been the most Democratic state in the Electoral College.

3 New Hampshire and Rhode Island are collectively the states where Clinton crosses the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line. If those two states are separated with Clinton winning New Hampshire and Trump, Rhode Island, then there would be a tie in the Electoral College.



NOTE: Distinctions are made between states based on how much they favor one candidate or another. States with a margin greater than 10 percent between Clinton and Trump are "Strong" states. Those with a margin of 5 to 10 percent "Lean" toward one of the two (presumptive) nominees. Finally, states with a spread in the graduated weighted averages of both the candidates' shares of polling support less than 5 percent are "Toss Up" states. The darker a state is shaded in any of the figures here, the more strongly it is aligned with one of the candidates. Not all states along or near the boundaries between categories are close to pushing over into a neighboring group. Those most likely to switch -- those within a percentage point of the various lines of demarcation -- are included on the Watch List below.


The Watch List1
State
Switch
Arkansas
from Lean Trump
to Strong Trump
Colorado
from Lean Clinton
to Toss Up Clinton
Delaware
from Strong Clinton
to Lean Clinton
Indiana
from Strong Trump
to Lean Trump
Iowa
from Toss Up Trump
to Toss Up Clinton
Maine
from Lean Clinton
to Toss Up Clinton
Missouri
from Lean Trump
to Toss Up Trump
Nevada
from Toss Up Trump
to Toss Up Clinton
New Hampshire
from Lean Clinton
to Toss Up Clinton
New Jersey
from Strong Clinton
to Lean Clinton
North Carolina
from Toss Up Clinton
to Toss Up Trump
Ohio
from Toss Up Clinton
to Toss Up Trump
Oregon
from Strong Clinton
to Lean Clinton
Pennsylvania
from Lean Clinton
to Toss Up Clinton
Rhode Island
from Lean Clinton
to Toss Up Clinton
Virginia
from Lean Clinton
to Toss Up Clinton
1 Graduated weighted average margin within a fraction of a point of changing categories.