Wednesday, August 24, 2016

The Electoral College Map (8/24/16)

Latest Update



New State Polls (8/24/16)
State
Poll
Date
Margin of Error
Sample
Clinton
Trump
Undecided
Poll Margin
FHQ Margin
Florida
8/19-8/22
+/- 2.7%
1200 likely voters
41
43
5
+2
+3.14
New Mexico
8/19-8/21
+/- _._%
1103 registered voters
40
31
9
+9
+8.64
North Carolina
8/20-8/23
+/- 4.9%
401 likely voters
44
42
6
+2
+2.21
South Carolina
8/18-8/21
+/- 4.0%
600 likely voters
39
39
16
+/-0
+1.98


Polling Quick Hits:
Rare are the weeks in which polls are simultaneously released in North and South Carolina. This is the second such day this week bringing new polls from each along with a new survey from Florida and an update in New Mexico.


Florida:
FHQ likely does a disservice by talking about average margins in these daily write-ups. The weighted average is taken on the candidates' shares of support and the polls and the margin calculated from the difference. That came to mind as I was typing up my thoughts on the Saint Leo poll yesterday. It comes up again today with another -- though less egregious -- outlier in Florida. It is not that a Trump +2 is bad or anything. Rather, that sort of outcome has become more sporadic in the post-conventions landscape.

But it may be more instructive to examine the candidates' shares of support in these two surveys released over the last couple of days instead of the margins. What makes today's FAU poll "better" is that Trump 43 - Clinton 41 offers levels of support that are well within the range of results for both across all Florida polls. The Saint Leo poll by comparison only got halfway there. Trump's 38 percent in that poll is at least in his Florida range. Clinton's 52 percent is not. Again, it is not yet. Things could change.

The straight average share across these two outliers is a bit more plausible -- Clinton 46.5 - Trump 40.5 -- but still a wider gap than the more modest difference in the FHQ methodology.


New Mexico:
There is a lot to say about Florida, less so in New Mexico. The tale in the Land of Enchantment is a brief one based on just two surveys now. Both are from the same firm (PPP) and both show basically the same thing despite three months separating them: Clinton is ahead in the upper single digits in a state Obama carried by just more than ten points in 2012. That is the end of the story for the time being.


North Carolina:
Nothing against the Marist polling in North Carolina, but few other polls -- even in the post-convention uptick for Clinton -- have had the Tar Heel state above anything more than about +4 for either candidate. But the balance of the polling through the FHQ graduated weighted average tilts in Clinton's direction by a couple of points. Needless to say, adding in a Clinton +2 from Monmouth does little to alter the course there. North Carolina is close just as it was in 2012. The difference in 2016 is that it has consistently fallen on the Democratic side of the partisan line rather than the Republican end.


South Carolina:
One could undoubtedly see a tie in South Carolina in this Feldman survey and get carried away, and then see the resulting average and repeat that process. Let's take a step back from both for a moment. The data, limited though it may be, is pointing toward a closer than usual race in the Palmetto state. However, it is still a state that tips toward Trump by what might seem like a decreasing amount over time. Perhaps, but shunt that average to the side for a moment. South Carolina did not budge on the Spectrum below despite that change. It still occupies a space that is distinct from closer Trump toss ups like Arizona and Georgia. And it is probably closer to Missouri though the averages do not quite reflect that at the moment. Very simply, we need more data in South Carolina. But what we have has consistently shown a close race that advantages Trump.


The new polls today did little to change either the Electoral College Spectrum or the Watch List. Both remain unchanged from a day ago.


NOTE: A description of the methodology behind the graduated weighted average of 2016 state-level polling that FHQ uses for these projections can be found here.


The Electoral College Spectrum1
HI-42
(7)
NJ-14
(175)
PA-203
(269 | 289)
MO-10
(155)
TN-11
(58)
MD-10
(17)
DE-3
(178)
NH-43
(273 | 269)
AK-3
(145)
LA-8
(47)
RI-4
(21)
WI-10
(188)
FL-29
(302 | 265)
KS-6
(142)
SD-3
(39)
MA-11
(32)
ME-4
(192)
OH-18
(320 | 236)
UT-6
(136)
ND-3
(36)
VT-3
(35)
NM-5
(197)
NC-15
(335 | 218)
TX-38
(130)
ID-4
(33)
CA-55
(90)
MI-16
(213)
IA-6
(341 | 203)
IN-11
(92)
NE-5
(29)
NY-29
(119)
OR-7
(220)
NV-6
(347 | 197)
MS-6
(81)
AL-9
(24)
IL-20
(139)
CT-7
(227)
AZ-11
(191)
AR-6
(75)
OK-7
(15)
WA-12
(151)
CO-9
(236)
GA-16
(180)
MT-3
(69)
WV-5
(8)
MN-10
(161)
VA-13
(249)
SC-9
(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 Pennsylvania (all Clinton's toss up states plus Pennsylvania), he would have 289 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 Pennsylvania 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 Pennsylvania and Trump, New Hampshire, 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
Alaska
from Lean Trump
to Toss Up Trump
Arizona
from Toss Up Trump
to Toss Up Clinton
Arkansas
from Strong Trump
to Lean Trump
Delaware
from Strong Clinton
to Lean Clinton
Georgia
from Toss Up Trump
to Toss Up Clinton
Indiana
from Lean Trump
to Strong Trump
Mississippi
from Strong Trump
to Lean Trump
Nevada
from Toss Up Clinton
to Toss Up Trump
New Hampshire
from Lean Clinton
to Toss Up Clinton
New Jersey
from Strong Clinton
to Lean Clinton
Wisconsin
from Lean Clinton
to Strong Clinton
1 Graduated weighted average margin within a fraction of a point of changing categories.



at

2 comments:

  1. Don DresserAugust 25, 2016 at 1:03 PM

    Just as a minor presentation point - when the "Yellow" victory line coincides with the break point in your color-coded categories, it makes it hard or impossible to tell whether PA and NH (in today's spectrum) are "Lean" or "Toss-up" states (the only way I see to do this is to note that NH is identified in the Watch List as a Lean Clinton state - albeit on the edge of shifting into Toss Up Clinton - and if PA ranks above NH on the list, it must also be Lean. Do your tools give you an easy way to mark the victory line with a cell border or come other indicator that would allow the color coding for the states to show through?

    Thanks for all your work on this - I have appreciated the electoral spectrum display format for the last couple of election cycles - and now see that others are hopping on (with 538's "snake" graph being perhaps the most visible attempt to present similar information.

    ReplyDelete
  2. Josh PutnamAugust 25, 2016 at 2:10 PM

    Thanks for the kind words about the site, Don.

    The cell border idea is one I had not considered. It's worth looking into. I have made an attempt at changing the text color in these situations. There was a divide a couple of weeks ago between the categories New Hampshire and Virginia -- then, the two states straddling 270 -- where it was more obvious. One had a lighter blue shade than the other.

    Both have since crossed over into the Lean category and when that occurred a couple of weeks ago I did ponder ways to better highlight the "proper" category for those yellow states, but one without adding yet another footnote to the table. I gave it some thought but then I looked up a bit and saw the map right above there and saw where they were. That stopped me from making any changes at the time and mostly satisfies me now. But there is a lack of uniformity between the two that is not lost on me.

    I'll resume brainstorming.

    ReplyDelete