Thursday, October 29, 2020

The Electoral College Map (10/29/20)

Update for October 29.

Well, it is too bad there were not any new polls on the Thursday before the election concludes next week. Oh wait.

With five days left in the general election campaign there was actually a flood of new state-level polling data. Here at FHQ 34 surveys from 18 states spanning all six categories were released since the last update to the projection yesterday. In two-thirds of those states, the additional polls moved the FHQ average margins in Joe Biden's direction. And that, too, encompassed states in all six categories. But President Trump was not without a silver lining in the face of all that. Margins in both Arizona and Minnesota pushed toward him on the day. However, despite another day of subtle movements, little changed in the meta-analysis. It was another date marked off the calendar in which Trump failed to meaningfully cut into Biden's advantage across the country.

On to the polls...

Polling Quick Hits:
(Trump 58, Biden 39 via AU-Montgomery )
[Current FHQ margin: Trump +19.88] 
Trump 57, Biden 37 in September poll
If this was a day that mostly maintained the status quo in the race for the White House, then the latest update from Auburn University in Montgomery is a solid lead off example. There not only was not much movement from September to now in the series, but the new poll closely tracks with the 58-38 (rounded) edge Trump has in the Yellowhammer state. Overall, that is enough to make Alabama safe for the president next week, but it is in line to shift toward the Democrats by about eight points since 2016.

(Trump 52, Biden 43 via Gravis Marketing)
[Current FHQ margin: Trump +6.34] 
No previous Gravis poll [Current FHQ averages in Alaska: Trump 50, Biden 44 (rounded)]
Recall that Alaska was a state where Trump barely topped 50 percent in 2016. That has been true in most polls in 2020, including this new Gravis survey of the Last Frontier. All that means that Trump has not exactly built on his win there four years ago, but he also has not lost much. Biden, on the other hand, more than seven points ahead of Clinton's 2016 share of support on election day in 2020 polling. That is enough to make Alaska more competitive, but not enough to get in the likely range of what the former vice president should expect on a really good night next Tuesday.

(Trump 49, Biden 45 via Pulse Opinion Research | Biden 47, Trump 47 via Ipsos)
[Current FHQ margin: Biden +2.90] 
Pulse Opinion Research: Biden 48, Trump 46 in last poll
Ipsos: Biden 50, Trump 46 in last poll
If there was one state today where that broke with the status quo, it was Arizona. There was movement in the Grand Canyon state in Trump's direction in both updates released in the state today. The Pulse update to a recent poll was reminiscent of the InsiderAdvantage survey released out of Pennsylvania earlier this week. Both saw a fairly dramatic shift, poll-to-poll, from Biden to the president. And neither really matches the movement -- or lack thereof -- in October polling. Although the Pulse survey is astray of the 48-45 (rounded) advantage Biden has in Arizona at FHQ, it has Trump at the top of his range and Biden at the bottom of his. 

(Biden 50, Trump 46 via Monmouth | Biden 51, Trump 47 via Marist | Biden 45, Trump 42 via Quinnipiac | Biden 48, Trump 47 via Ipsos)
[Current FHQ margin: Biden +3.14] 
Monmouth: Biden 49, Trump 46 in September poll
Marist: Biden 48, Trump 48 in September poll
Quinnipiac: Biden 51, Trump 40 in early October poll
Ipsos: Biden 51, Trump 46 in poll last week
[NOTE: Using data from the high turnout Monmouth model -- a 51-45 Biden advantage -- would have increased the average FHQ margin to Biden +3.16 in Florida.]
With the exception of the Quinnipiac survey, which updated an outlier from a couple of weeks ago, Biden benefited from the updates in the other three polls that were released in Florida today. There was not movement toward Biden in all of them, but his advantages stretched in all three. More importantly, the Democratic nominee hit 50 percent in that trio of Sunshine state surveys as well. Like the discussion about Arizona polling a day ago, there may be evidence of some narrowing in Florida as well. But it is offset by how frequently Biden has been at or above 50 percent in recent Florida polling. The former vice president has been at or beyond the majority mark in 41 of 108 surveys in the Sunshine state this year. And nearly half of that 41 -- 19 surveys -- have found Biden hitting or exceeding 50 percent in just October alone. 

(Biden 48, Trump 46 via Public Policy Polling)
[Current FHQ margin: Biden +0.40] 
Biden 47, Trump 46 in early October poll
At first glance, little change from the early October Georgia poll from PPP, but then, that is par for the course in this series in 2020. Other than the June poll in which Biden led by four, the remaining surveys have all 1) had him ahead and 2) by margins in the one to two point range. This poll not only fits that trend, but is consistent with where most of the recent polling has been in the Peach state. 

(Trump 47, Biden 46 via Quinnipiac)
[Current FHQ margin: Trump +0.64] 
Biden, 50, Trump 45 in early October poll 
FHQ does not know if Quinnipiac was spooked by that Florida outlier cited above, but today's batch from the university pollster seems to be a correction of some form. That may or may not be the case, but the percentage of undecideds grew from the last poll to now in three of the four polls including this Iowa survey. Only in Pennsylvania did the share of undecideds decrease. Regardless, that dynamic, whether a correction of some sort or natural movement (not really seen in other state-level polling) has closed the gap in Iowa from a poll just a couple of week ago. And it is also true that the last round of Q-polls were simple outliers and this is a regression to the mean. This Iowa poll is consistent with the 47-46 lead Trump currently holds in the FHQ averages. But that undecided shift is a curious footnote.

(Biden 53, Trump 39 via Survey USA)
[Current FHQ margin: Biden +13.54] 

Maine CD1
(Biden 59, Trump 40 via Survey USA)
[Current FHQ margin: Biden +23.36] 

Maine CD2
(Biden 51, Trump 49 via Survey USA)
[Current FHQ margin: Biden +1.77] 
No previous Survey USA poll [Current FHQ averages in ME CD2: Biden 46, Trump 45 (rounded)]
FHQ will not dwell on this new Survey USA survey that encompassed all three jurisdictions with electoral votes at stake in the Pine Tree state. The second congressional district will be the most competitive of the bunch and the ranked choice voting-simulated version of that poll -- one that took second, third and fourth preferences -- gave the former vice president a two point edge that closely resembles the projected advantage he currently holds for that one electoral vote. There has not been a ton of polling in these two Maine districts, but what has been in the field has tended to favor Biden more often than not. 

(Biden 52, Trump 42 via Mitchell Research)
[Current FHQ margin: Biden +7.21] 
Biden 51, Trump 41 in mid-October poll
The Mitchell poll a week ago was on the upper end of the margins Biden has enjoyed in recent polling of the Great Lakes state, and the update this week ahead of the election does not look that much different. It still has Biden on the upper end of his range in the state and Trump near the bottom of his. This poll does expand the Democratic nominee's advantage in Michigan, but not by much. 

(Biden 47, Trump 42 via Survey USA)
[Current FHQ margin: Biden +8.39] 
Biden 48, Trump 42 in last week's poll
The new Survey USA poll of Minnesota is not that different from the Michigan update above. But it is, perhaps, on the opposite end of the spectrum. There was not much change from one week to the next through a Survey USA lens that has Biden at the bottom of his October range and Trump consistent with his FHQ average share of support in the Land of 10,000 Lakes. Survey USA has also conducted three of the nine October surveys in Minnesota and has provided three of the four instances in which Biden has failed to reach 50 percent in the state. 

(Trump 52, Biden 45 via MSU-Billings)
[Current FHQ margin: Trump +7.76] 
No previous MSU-Billings poll [Current FHQ averages in Montana: Trump 51, Biden 44 (rounded)]
First of all, this MSU-Billings survey of the Treasure state is consistent with the established FHQ average. But it also adds even more evidence to the massive potential shift there from 2016 to now. The existing Montana polling points to a more than 13 point swing toward the Democrats. That is enough to pull Montana into the Lean Trump category, but that is probably it. Still, that potential shift highlight just how uncompetitive Montana ended up being four years ago. 

New Hampshire
(Biden 58, Trump 39 via American Research Group | Biden 53, Trump 43 via UMass-Lowell)
[Current FHQ margin: Biden +11.13] 
American Research Group: Biden 53, Trump 44 in September poll
UMass-Lowell: Biden 52, Trump 44 in September poll
There was little movement in the UMass-Lowell series in the Granite state from September to now, but that was not the case in the ARG series. As surprising as it may seem, the Alabama poll above and this ARG survey of New Hampshire mirror each other. And yes, Alabama is closer than is typical of the Yellowhammer state in presidential elections, but an equivalent 19 point gap in New Hampshire grabs the attention. That is doubly true for a state that was tight in 2016 and that the president's reelection effort was targeting earlier in this cycle. [The president was just there over the weekend too.] Given a uniform swing from four years ago, New Hampshire would not be competitive necessarily, but it also would not likely by tipped nearly 20 points in Biden's direction either.

New Jersey
(Biden 61, Trump 37 via Rutgers/Eagleton)
[Current FHQ margin: Biden +19.74] 
Biden 56, Trump 33 (among registered voters) in May poll
That Biden has hit 60 percent in the update to May's Rutgers/Eagleton poll of New Jersey is noteworthy in and of itself. He has only reached those heights in a couple of the Survey Monkey polls of the Garden state this year. But it also serves to highlight just how little the former vice president has improved on Clinton's showing in New Jersey four years ago. It stands out as the exception rather than the rule. The president, on the other hand, has been consistently in the upper 30s there, a little more than four points off his 2016 pace there.

North Carolina
(Biden 48, Trump 45 via Siena/NYT Upshot | Biden 48, Trump 48 via UMass-Lowell)
[Current FHQ margin: Biden +1.85] 
Siena: Biden 46, Trump 42 in earlier October poll
UMass-Lowell: Biden 47, Trump 47 in September poll
The Citizen Data survey (described as part of the collective wave below) did all the heavy lifting in pushing the FHQ average margin in North Carolina closer to Biden +2, but the other two polls of the Tar Heel state released today maintained the status quo. Neither the Siena nor the UMass-Lowell surveys saw any substantial shifts from the last polls in those respective series and neither update differs much from the range of polls out of North Carolina in October (or the rest of the year for that matter). Biden continues to maintain a 48-46 (rounded) advantage at FHQ.

(Trump 49, Biden 47 via Gravis Marketing | Biden 48, Trump 43 via Quinnipiac)
[Current FHQ margin: Trump +0.67] 
No previous Gravis poll
Quinnipiac: Biden 48, Trump 47 in mid-October poll
While the first Gravis poll in the Buckeye state is in line with the small lead that Trump has had there throughout much of 2020, the Quinnipiac survey is not. But unlike the Florida and Iowa Q-polls, the one in Ohio swung toward Biden and not the president. It did, however, see the share of undecideds grow from the previous poll in the series. That drew in the margin, pushing Ohio closer to the partisan line, but the Buckeye state remains a state that is close but slightly tilted toward the president. The wiggle room in Ohio polling is that neither candidate has really been north of 50 percent much all year. Of the 42 polls that have been in the field there in 2020, Trump has hit or surpassed the majority mark only eight times and Biden has only done so in three surveys. Trump is the only one to get there in October, but even then, only once.

(Biden 51, Trump 45 via RMG Research | Biden 51, Trump 44 via Quinnipiac | Biden 50, Trump 44 via Franklin and Marshall)
[Current FHQ margin: Biden +5.41] 
RMG Research: Biden 49, Trump 43 in earlier October poll
Quinnipiac: Biden 51, Trump 43 in earlier October poll
Franklin and Marshall: Biden 48, Trump 42 in September poll
If there is any new Pennsylvania poll out, then odds are it will have Biden around 50 percent and Trump somewhere in the mid-40s. All three polls fit that mold today, and while both the RMG and the Franklin and Marshall polls showed positive movement for both candidates, both new surveys maintained the margins from the previous polls in each series. This string of polls nudged Biden that much closer to 50 percent in the averages at FHQ as well.

(Trump 48, Biden 47 via UMass-Lowell)
[Current FHQ margin: Trump +1.40] 
UMass-Lowell: Trump 49, Biden 46 in September poll 
Minimal movement from one UMass-Lowell survey of Texas to the next did little to change the outlook in the Lone Star state. Trump still has the advantage in a state that continues to look like the North Carolina of the Trump side of the partisan line: close, but consistently tipped toward the president. And like North Carolina, it was the Citizen Data survey that triggered the changes in the Texas average margin.

(Biden 62, Trump 32 via co/efficient)
[Current FHQ margin: Biden +36.71] 
No previous co/efficient poll [Current FHQ averages in Vermont: Biden 66, Trump 30 (rounded)]
A rare Vermont poll failed to offer any real suspense. The Green Mountain state is a heavy Biden lean off on the far left end of the order on the Electoral College Spectrum, and other than Biden ending up closer to 60 percent rather than 70 percent, this poll -- an internal Republican survey for the lieutenant governor's race in Vermont -- did not stray too far from what one might expect. Still, Vermont is one of those states where the 2020 polling relative to where the candidates ended up in 2016 is a tad askew. While Trump hovers near where he was in Vermont -- lagging less than a point behind his 2016 pace -- Biden is more than nine points out in front of Clinton from four years ago. 

(Biden 51, Trump 39 via Virginia Commonwealth)
[Current FHQ margin: Biden +11.84] 
Biden 53, Trump 39 in September poll 
In the Old Dominion, Biden lost a couple of points in the VCU poll since its last survey in September. Both candidates fall a couple of points shy of their FHQ average shares of support in this latest update, but the margin is right on target. That further cements Virginia as something other than  the battleground it has been in recent cycles. 

Citizen Data (October battleground polls -- initial wave of releases):
Ohio: Trump 44, Biden 43
Texas: Biden 49, Trump 45
Georgia: Biden 48, Trump 44
Florida: Biden 50, Trump 45
Pennsylvania: Biden 44, Trump 39
North Carolina: Biden 50, Trump 44
Michigan: Biden 50, Trump 41

This strange set of polls from Citizen Data was made all the more strange by the correction it had to make to a Texas poll that initially (and mistakenly) indicated that Biden was ahead by ten points. A four point lead for the vice president in the Lone Star state is already off to the extreme Biden end of of the range in Texas polling. If that was not enough, the order of states in this wave is not at all consistent with the established order at FHQ. North Carolina being to the Biden side of both Florida and Pennsylvania stands out, but then again the undecideds/others-inflated totals in that survey of the Keystone state do as well. 


The Electoral College Spectrum1
(273 | 285)
(279 | 265)
SC -9
(308 | 259)
NE CD1-1
(319 | 230)
ME CD2-1
(335 | 219)
(351 | 203)
ME CD1-1
NE CD2-1
NE CD3-1
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 (Biden's toss up states plus the Pennsylvania), he would have 285 electoral votes. Trump's numbers are only totaled through the states he would need in order to get to 270. In those cases, Biden's number is on the left and Trump's is on the right in bold italics.

3 Pennsylvania
 is the state where Biden crosses the 270 electoral vote threshold to win the presidential election, the tipping point state. The tipping point cell is shaded in yellow to denote that and the font color is adjusted to attempt to reflect the category in which the state is.

On the weight of two new polls -- especially that ARG poll -- New Hampshire shifts off the Watch List. It is now outside of a point of moving into the Lean Biden category and is a more comfortable Strong Biden state. 

Similarly, a wide margin in the latest New Jersey poll pushed the Garden state past Oregon in the order and one cell away from the partisan line. New Jersey is a safe Strong Biden state as well.

But again, this was another day of subtle changes here at FHQ and not the sort of big changes that the president needs at this late date. The electoral vote tally remains at Biden, 351-187 despite the changes further out in the order on the Spectrum and on the Watch List. 

5 days to go.

Where things stood at FHQ on October 29 (or close to it) in...

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 Biden 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
Potential Switch
from Toss Up Biden
to Toss Up Trump
from Toss Up Trump
to Toss Up Biden
from Lean Trump
to Strong Trump
from Toss Up Biden
to Lean Biden
New Mexico
from Strong Biden
to Lean Biden
from Toss Up Trump
to Toss Up Biden
from Lean Biden
to Toss Up Biden
1 Graduated weighted average margin within a fraction of a point of changing categories.

Methodological Note: In past years, FHQ has tried some different ways of dealing with states with no polls or just one poll in the early rounds of these projections. It does help that the least polled states are often the least competitive. The only shortcoming is that those states may be a little off in the order in the Spectrum. In earlier cycles, a simple average of the state's three previous cycles has been used. But in 2016, FHQ strayed from that and constructed an average swing from 2012 to 2016 that was applied to states. That method, however, did little to prevent anomalies like the Kansas poll that had Clinton ahead from biasing the averages. In 2016, the early average swing in the aggregate was  too small to make much difference anyway. For 2020, FHQ has utilized an average swing among states that were around a little polled state in the rank ordering on election day in 2016. If there is just one poll in Delaware in 2020, for example, then maybe it is reasonable to account for what the comparatively greater amount of polling tells us about the changes in Connecticut, New Jersey and New Mexico. Or perhaps the polling in Iowa, Mississippi and South Carolina so far tells us a bit about what may be happening in Alaska where no public polling has been released. That will hopefully work a bit better than the overall average that may end up a bit more muted.

Related posts:

Follow FHQ on TwitterInstagram and Facebook or subscribe by Email.

No comments: