Sunday, October 4, 2020

The Electoral College Map (10/4/20)

Update for October 4.


Changes (October 4)
StateBeforeAfter
Alaska
Toss Up Trump
Lean Trump
Georgia
Toss Up Biden
Toss Up Trump
Louisiana
Lean Trump
Strong Trump
New Hampshire
Lean Biden
Strong Biden
New Mexico
Strong Biden
Lean Biden
No, Sunday did not bring a flood of new poll releases. There were just a couple from YouGov out of the Rust Belt. But FHQ did incorporate the four waves -- June, July, August and September -- of surveys across the country from Survey Monkey. The addition of more than 200 polls into the dataset was not without consequence, but the changes were limited in the face of that flood of newly introduced information.

The most consequential change was Georgia hopping back over the partisan line into the Trump group of states after having just shifted to Biden by the slimmest of margins a day before. But again, as FHQ cautioned yesterday, the shade that the Peach state takes on is probably less important than the fact that it is as close as it is. Georgia may have pushed back onto Trump turf, but there are just 0.12 points separating the candidates there.

The other changes brought about by the Survey Monkey polls were more about calibrating the states' positioning in the rank ordering (see Electoral College Spectrum below) than they were about any fundamental shake up to the overall alignment. Alaska and Louisiana moving deeper into Trump territory makes some sense. The Last Frontier moving out of the Toss Up Trump category and into the lower end of the Lean Trump range "feels" right even if the changes brought about by the four new surveys brings Trump basically in line with his showing there on election day in 2016. In other words, the lack of swing now in Alaska is not consistent with the more than two points Trump has lost on average across the country in the time.

On the other side of the partisan line, the shifts in New Hampshire and New Mexico are noteworthy but a function of a general lack of polling overall in each. New Hampshire was at least trending toward the Strong/Lean line on the Biden side of the ledger, but the Survey Monkey polls were all overly Biden-friendly and had small sample sizes. Things in the Land of Enchantment are much more clearly about four new polls being added to a small existing pool of surveys there. Its push into the upper end of the Lean Biden category would have raised more red flags had it not ended up on the Watch List below. New Mexico, then, remains in range of the Strong/Lean line.

All that aside from Survey Monkey, here is a look at the other polls to close out the weekend.


Polling Quick Hits:
Ohio
(Trump 47, Biden 47)
[Current FHQ margin: Trump +0.31]
The narrow one point Trump lead that YouGov showed in the Buckeye state when the firm was last in the field there in July has dissipated in the time since. Now, the race is even more a dead heat in the latest survey of Ohio from YouGov. Honestly, this does not represent any real change in the state over the last couple of months. Ohio has been close at FHQ. Period. Sometimes that has meant a Trump lead of more than a point, but more often than not since FHQ began these updates in mid-June, the Buckeye state has been on the Watch List. [And yes, it did start out as a Toss Up Biden state before switching over.]


Pennsylvania
(Biden 51, Trump 44)
[Current FHQ margin: Biden +5.39]
If Ohio has shifted eight points in the Democrats' direction since 2016, as the above poll would indicate, then the new YouGov survey of Pennsylvania is consistent with that. More or less a tie in 2016, the Keystone state is, according to this poll in any event, about seven points more Democratic than it was at the end of the last cycle. In truth, the average margin at FHQ in the commonwealth is a bit more modest than this poll, but this YouGov survey is not all that divergent from other recent Pennsylvania polls. Furthermore, it is consistent with the last poll the firm conducted there in August (49-43, Biden). In other words, it maintained the status quo there.


NOTE: A description of the methodology behind the graduated weighted average of 2020 state-level polling that FHQ uses for these projections can be found here.


The Electoral College Spectrum1
DC-3
VT-3
(6)2
IL-20
(162)
WI-10
(253)
SC-9
(125)
TN-11
(60)
MA-11
(17)
NJ-14
(176)
PA-203
(273 | 285)
AK-3
(116)
KY-8
(49)
MD-10
(27)
OR-7
(183)
NV-6
(279 | 265)
MO-10
(113)
AL-9
(41)
CA-55
(82)
ME-2
(185)
FL-29
(308 | 259)
KS-6
(103)
SD-3
(32)
NY-29
(111)
CO-9
(194)
AZ-11
ME CD2-1
(320 | 230)
NE CD1-1
MT-3
(97)
ID-4
(29)
HI-4
(115)
VA-13
(207)
NC-15
(335 | 218)
NE-2
(93)
AR-6
(25)
DE-3
(118)
NH-4
(211)
GA-16
(203)
IN-11
(91)
OK-7
(19)
WA-12
(130)
NM-5
(216)
OH-18
(187)
UT-6
(80)
ND-3
(12)
CT-7
ME CD1-1
(138)
MN-10
(226)
IA-6
(169)
MS-6
(74)
WV-5
(9)
RI-4
(142)
MI-16
NE CD2-1
(243)
TX-38
(163)
LA-8
(68)
WY-3
NE CD3-1
(4)
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.

Sunday did not lack for changes here at FHQ. The map had its changes as discussed above and so, too, did the order on the Spectrum. But that realignment in the order was concentrated outside of the middle column of most competitive states. In fact, from Michigan at the bottom of the second column from the left to Texas, there was no change in the alignment of states (meaning that Pennsylvania remains the tipping point). That is an important point. Granted, that group of states is among the most polled of the cycle, and the addition of four new polls would have less of an impact there than if four new polls were added to one existing poll in Vermont, for example.

The bigger change might be to the Watch List below. Of the ten states that were on the List a day ago, only Georgia, Ohio and Pennsylvania remain within a point of changing categories. And they are joined by just two other states -- New Hampshire and New Mexico -- today. That sort of paring down of the List is indicative of a certain calibration in the order. The vast majority of states have pushed firmly into the categories in which they now reside. That is not to say that those other states cannot change, but that there is some insulation there now that was not yesterday.



Where things stood at FHQ on October 4 (or close to it) in...
2016
2012
2008


--
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
State
Potential Switch
Georgia
from Toss Up Trump
to Toss Up Biden
New Hampshire
from Strong Biden
to Lean Biden
New Mexico
from Lean Biden
to Strong Biden
Ohio
from Toss Up Trump
to Toss Up Biden
Pennsylvania
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:
The Electoral College Map (10/3/20)

The Electoral College Map (10/2/20)

The Electoral College Map (10/1/20)


Follow FHQ on TwitterInstagram and Facebook or subscribe by Email.

Saturday, October 3, 2020

The Electoral College Map (10/3/20)

Update for October 3.


Changes (October 3)
StateBeforeAfter
Georgia
Toss Up Trump
Toss Up Biden
After a rollercoaster ride of a work week that included the first presidential debate and the president testing positive for Covid-19, the weekend began with a surprising number of poll releases. And perhaps, more importantly, Georgia jumped the partisan line back into Toss Up Biden territory for the first time since early July. That shift comes with the typical caveats. Before the pair of new surveys of the Peach state were added, Georgia was already the most competitive of the Trump states. Now, it is the most competitive of the Biden states. And that change in color -- and classification -- is likely less consequential than the fact that Georgia remains as close as it has. In other words, this is to be expected. Georgia was basically tied, albeit tipped slightly toward the president, and remains that way, only tilted toward the former vice president.

And again, as has been said around here a number of times -- often in the context of the movement in Ohio -- if Georgia and/or Ohio are among the most competitive states on November 3, then states like Michigan, Pennsylvania and Wisconsin are likely more firmly in Biden's column, and the former vice president's electoral vote total is at or over 270.

On to the polls...


Polling Quick Hits:
Florida
(Biden 47, Trump 42)
[Current FHQ margin: Biden +3.52]
Siena/NYT Upshot were last in the field in the Sunshine state back in June and in the time since has transitioned from a registered to likely voter sample. Despite that change, the numbers have barely budged over those three plus months. Biden still holds the support of 47 percent of respondents and Trump's share only increased by a single point. In a state that the president is absolutely going to have to have to get to 270, that is not the sort change he needs. It is positive for the Trump campaign that the gap closed, however marginally, but it is not outside the realm of possibility that that is simply statistical noise.


Georgia
(Biden 45, Trump 44)
[Current FHQ margin: Biden +0.04]
There is a Hart Research Associates poll among the wave discussed below, but Redfield and Wilton Strategies also recently conducted a survey in the Peach state. Both show Biden narrowly ahead, but only the R&W poll is part of a series. And the last poll found Trump to be just ahead of the former vice president. Like the Florida poll above, Biden was stable (at 45 percent) while only Trump's number changed since mid-September, dropping a couple of points. That still leaves a rather healthy share of undecideds at this late stage of the race in a state that has been reliably red over the last generation. Regardless, it is Biden and not Trump who has now inched into the smallest sliver of an edge in the battle for the 16 electoral votes on the line in Georgia.


Missouri
(Trump 51, Biden 46)
[Current FHQ margin: Trump +6.11]
Although there have not been very many updates on the presidential race in Missouri in calendar 2020, Remington Research has carried much of the load. The firm has accounted for 40 percent of the survey work in the Show-Me state and the trend line has broken with the established trajectory of the presidential race. When, during June and July, Biden was surging in national and state polls across the country, there was barely any movement through Remington's lens in Missouri. But now that things have regressed to the pre-surge mean elsewhere, the Remington series continues to show a gradually decreasing lead for the president. Whereas Trump has stayed in the low 50s, Biden's support has continued to track upward at Remington over time. No, Biden is not likely to win the Show-Me state unless the bottom truly drops out for the president, but the shift since 2016 is indicative. Trump's support has trailed off by nearly six points as Biden has gained a nearly equivalent amount on Clinton's share in the state in November 2016.


New Jersey
(Biden 52, Trump 38)
[Current FHQ margin: Biden +17.74]
In the Garden state, DKC Analytics has consistently had Biden underperforming Clinton's pace in the state from four years ago. But that has been overshadowed by Trump's own decline in New Jersey. The president topped 40 percent in the state in 2016 but has only hit (or exceeded) that mark just once in calendar 2020. And Trump sits more than five points below his 41 percent share from 2016. This poll did nothing to change the established dynamics for both candidates.


Pennsylvania
(Biden 49, Trump 42)
[Current FHQ margin: Biden +5.36]
There was more of an interregnum between Siena surveys in Florida, but Siena/NYT Upshot was just in the field in the Keystone state a week ago before the first debate. Sure, the change is small, but the direction runs counter to much of the analysis in the wake of the first showdown between the two major party candidates in Cleveland. It was Trump who gained a couple of points while Biden again -- as in Florida and Georgia above -- remained stationary but all the while hovering around 50 percent.  And that has been the story in this series from Siena all year: Biden around the majority mark and Trump mired in the low 40s. In the pivotal tipping point state, one would much rather be in Biden's position than Trump's with just more than four weeks until election day.


Wisconsin
(Biden 47, Trump 44)
[Current FHQ margin: Biden +6.13]
Just as there was little movement in a pair of recent Siena surveys in Pennsylvania, there was not much to speak of in Wisconsin from Trafalgar Group either. The only difference was that the two back-to-back polls from Trafalgar were both conducted before the first debate. The firm shows in their release numbers from both before and after the debate, but the dates the survey was in the field precede the debate. In any event, Biden maintained his three point advantage as both candidates lost a point each. And that maintains a shift in the series from the right-leaning Trafalgar. Narrowly ahead there in August and before, Trump has fallen and stayed behind in the series in the time since the national conventions (movement that breaks with the return to pre-surge polling overall in the race).


Hart Research Associates (late September wave of battleground polls)

Georgia: Biden +3
Iowa: Trump +2 [Current FHQ margin: Trump +0.50]
North Carolina: Biden +2 [Current FHQ margin: Biden +1.48]
Ohio: Trump +2 [Current FHQ margin: Trump +0.33]
Texas: Trump +2 [Current FHQ margin: Trump +1.23]

Last but not least is an additional wave from Hart Research that pairs with a prior wave in five of the six core battlegrounds released earlier in the week. While the previous set askew as compared to where each state has tended to fall in the order of states on the Electoral College Spectrum below -- Arizona and Pennsylvania had wider than typical Biden leads and Michigan narrower -- this latest wave in mostly Trump toss ups is more in line with the order if not the margins as well. The one exception is Georgia. The Biden +3 in the Peach state is not out of the ordinary, but that the race stands there while North Carolina, for example, is Biden +2 stands out. The Tar Heel state has consistently been to the Biden side of Georgia in the order throughout the year. Now, of course, the Peach state has rejoined North Carolina in the Biden side of the partisan line.


NOTE: The finalized September data from Survey Monkey on Michigan, Ohio, Pennsylvania and Wisconsin have been added to their respective dataset and are reflected in the averages above. Prior to today, the data had only extended through September 28. The last two days have now been added.


NOTE: A description of the methodology behind the graduated weighted average of 2020 state-level polling that FHQ uses for these projections can be found here.


The Electoral College Spectrum1
DC-3
MA-11
(14)2
CT-7
(162)
WI-10
(253)
AK-3
(125)
AL-9
(60)
HI-4
(18)
NJ-14
(176)
PA-203
(273 | 285)
SC-9
(122)
IN-11
(51)
CA-55
(73)
OR-7
(183)
NV-6
(279 | 265)
MO-10
(113)
UT-6
(40)
MD-10
(83)
NM-5
(188)
FL-29
(308 | 259)
MT-3
(103)
KY-8
(34)
VT-3
(86)
ME-2
(190)
AZ-11
ME CD2-1
(320 | 230)
NE CD1-1
KS-6
(100)
SD-3
(26)
NY-29
(115)
CO-9
(199)
NC-15
(335 | 218)
LA-8
(93)
ID-4
(23)
WA-12
ME CD1-1
(128)
VA-13
(212)
GA-16
(351 | 203)
MS-6
(85)
ND-3
(19)
RI-4
(132)
MN-10
(222)
OH-18
(187)
AR-6
(79)
OK-7
(16)
DE-3
(135)
NH-4
(226)
IA-6
(169)
NE-2
(73)
WV-5
(9)
IL-20
(155)
MI-16
NE CD2-1
(243)
TX-38
(163)
TN-11
(71)
WY-3
NE CD3-1
(4)
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.

And it is that change -- Georgia jumping the partisan line into Biden territory -- that highlights the changes on the day. In fact, it is the only change, but a big one. The Peach state slipping into Biden's column raises his projected total back above 350 electoral votes, but note how small the margin is. With the Peach state that close to tied, it moving back across the partisan line in the coming days as new polling data is released is not out of the question. But Georgia is now the palest shade of blue on the map and on the Spectrum. And its potential shift has changed on the Watch List below. Other than the Peach state, everything else remained on Saturday the same as it was on Friday to close the work week.



Where things stood at FHQ on October 3 (or close to it) in...
2016
2012
2008



--
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
State
Potential Switch
Arkansas
from Strong Trump
to Lean Trump
Georgia
from Toss Up Biden
to Toss Up Trump
Iowa
from Toss Up Trump
to Toss Up Biden
Louisiana
from Lean Trump
to Strong Trump
Mississippi
from Strong Trump
to Lean Trump
Nevada
from Toss Up Biden
to Lean Biden
Ohio
from Toss Up Trump
to Toss Up Biden
Pennsylvania
from Lean Biden
to Toss Up Biden
South Carolina
from Lean Trump
to Toss Up Trump
Virginia
from Strong Biden
to Lean 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:
The Electoral College Map (10/2/20)

The Electoral College Map (10/1/20)

The Electoral College Map (9/30/20)


Follow FHQ on TwitterInstagram and Facebook or subscribe by Email.

Friday, October 2, 2020

The Electoral College Map (10/2/20)

Update for October 2.


Well, this has been a week.

The work week comes to a close with a handful of new surveys from five states and another wave of polls from Redfield and Wilton Strategies in the six core battlegrounds. Of those five, surveys of Georgia, Michigan and New Hampshire were all in the field after the first presidential debate on Tuesday night. But of those only the polls in Georgia and Michigan had pre-debate surveys from the same firms. In both there was movement in Biden's direction following the debate. The swing in Georgia from an August Landmark Communications survey to now was more pronounced, but that preceding survey last month, in retrospect, was a fairly clear outlier. It, too, had a survey that preceded it. And using that one as the point of comparison rather than the outlier shows a smaller shift, but still one in the former vice president's direction.  The change in Michigan was more modest from one Public Policy Polling survey (on behalf of Progress Michigan) to another. All in all, the changes were small enough to have potentially just been statistical noise than from anything stemming from the debate itself. Still, this is likely something worth tracking heading forward.

On to the other surveys...


Polling Quick Hits:
Arizona
(Biden 50, Trump 46)
[Current FHQ margin: Biden +3.47]
The Suffolk poll of likely voters in the Grand Canyon state is the firm's first conducted there in calendar 2020. And while there is no natural point of comparison, the survey is not inconsistent with the established averages at FHQ. The margin is on target, but both candidates are running a couple of points ahead of their FHQ average shares of support. As has been noted a few times in this space, that sort of dynamic should continue as election day approaches and more undecideds come off the board. That should increasingly have one or both of the candidates out in front of their averages.


Georgia
(Biden 47, Trump 45)
[Current FHQ margin: Trump +0.10]
As noted above, FHQ will throw out the immediately previous Landmark Communications survey in favor of the one that was released two weeks before that in mid-August. It fit in more with the other polls in the field in the Peach state around the same time and represented a four point swing toward Biden (rather than a nine point shift). And really, the two polls -- minus the outlier -- are much more in line with the ranges of support in which both candidates have much more consistently ended up in. Georgia is most often in a Biden +3 to Trump +3 range at the moment and this latest Landmark poll fits well in there.


Michigan
(Biden 50, Trump 44)
[Current FHQ margin: Biden +7.04]
Meanwhile, in Michigan, Public Policy Polling offered an opportunity to examine the possible pre- and post-debate effects. And the movement in the Great Lakes state was toward Biden as well. However, it did not really come at Trump's expense. The president was stable at 44 percent across the two surveys and Biden picked up a couple of points from other-undecided in that time. But again, while it may be easy to chalk that up to some impact from the debate earlier this week, the shift is small enough to just as easily have been some more pedestrian polling variability instead. The bottom line is that the former vice president maintains a smaller than average advantage in these surveys than the full world of Michigan polls.


New Hampshire
(Biden 53, Trump 45)
[Current FHQ margin: Biden +7.96]
Emerson waded into the Granite state to gauge public opinion on the presidential race for the first time in calendar 2020 and found what the slew of other recent survey of the state have found: Biden up by a margin in the upper single digits. As with the Arizona poll above, this survey nailed the margin, but had both candidates a couple of points out in front of their established average shares of support here at FHQ.


New York
(Biden 61, Trump 29)
[Current FHQ margin: Biden +28.35]
Over in the Empire state Siena_ was back in the field, but for the first drew a likely rather than registered voter sample. That shift was beneficial to the Democratic nominee. Biden gained four percent since the June Siena survey to push back above the 60 percent mark. Trump, on the other, hand lost out in the registered-to-likely transition. The president not only lost ground because of Biden's gain, but also saw his own support decay, nudging his support below 30 percent in his former home state for the first time since an April Siena survey there. Look, Biden is unlikely to lose New York on November 3, but he is running a hair above Hillary Clinton's pace in the state from four years ago while Trump has lost nearly six points off his 2016 showing in the state.


Redfield & Wilton Strategies (late September wave of battleground polls)

Arizona: Biden +3 (Biden +/-0, Trump +2  since mid-September wave)
Florida: Biden +5 (Biden +1, Trump -1) [Current FHQ margin: Biden +3.49]
Michigan: Biden +9 (Biden +2, Trump +3)
North Carolina: Biden +2 (Biden +/-0, Trump +/-0) [Current FHQ margin: Biden +1.47]
Pennsylvania: Biden +6 (Biden +1, Trump +/-0) [Current FHQ margin: Biden +5.35]
Wisconsin: Biden +5 (Biden +1, Trump +2) [Current FHQ margin: Biden +6.18]

Finally, there was another wave of battleground surveys from Redfield and Wilton. And again, the changes were minimal from the last wave less than two weeks ago. What continues to stand out in these is the extent to which the order in these polls differs from the established order on the Electoral College Spectrum. Some variability in the margins is to be expected between the two, but the Arizona is now back to being on the low side of Florida and Pennsylvania and Wisconsin are flipped in the order. Those are, however, small differences.



NOTE: A description of the methodology behind the graduated weighted average of 2020 state-level polling that FHQ uses for these projections can be found here.


The Electoral College Spectrum1
DC-3
MA-11
(14)2
CT-7
(162)
WI-10
(253)
AK-3
(125)
AL-9
(60)
HI-4
(18)
NJ-14
(176)
PA-203
(273 | 285)
SC-9
(122)
IN-11
(51)
CA-55
(73)
OR-7
(183)
NV-6
(279 | 265)
MO-10
(113)
UT-6
(40)
MD-10
(83)
NM-5
(188)
FL-29
(308 | 259)
MT-3
(103)
KY-8
(34)
VT-3
(86)
ME-2
(190)
AZ-11
ME CD2-1
(320 | 230)
NE CD1-1
KS-6
(100)
SD-3
(26)
NY-29
(115)
CO-9
(199)
NC-15
(335 | 218)
LA-8
(93)
ID-4
(23)
WA-12
ME CD1-1
(128)
VA-13
(212)
GA-16
(203)
MS-6
(85)
ND-3
(19)
RI-4
(132)
MN-10
(222)
OH-18
(187)
AR-6
(79)
OK-7
(16)
DE-3
(135)
NH-4
(226)
IA-6
(169)
NE-2
(73)
WV-5
(9)
IL-20
(155)
MI-16
NE CD2-1
(243)
TX-38
(163)
TN-11
(71)
WY-3
NE CD3-1
(4)
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.

Like yesterday's update, the polls added to the dataset today did nothing to alter either the map or the Watch List. But the did have some impact on the order among four important toss up states. Arizona and Florida again changed places with the Grand Canyon state moving one cell closer to the partisan line. And in Trump territory, there was more shuffling among the the group of states clustered within a point of the partisan line. The new Georgia poll brought the average margin down enough in the Peach state to jump it over Ohio in to the most competitive position right up against the barrier separating Trump from Biden states. But again, both are so close to one another right now that one is likely to see more of this moving forward. And importantly, both states are within two-tenths of a point of shifting over onto Toss Up Biden turf.

That is no small thing. As has become the mantra of sorts around here, if the chatter about the most competitive states on election day is about Georgia, Iowa and Ohio rather than, say, Michigan, Pennsylvania and Wisconsin, then one will have a pretty good idea of where the election is likely headed.



Where things stood at FHQ on October 2 (or close to it) in...
2016
2012
2008



--
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
State
Potential Switch
Arkansas
from Strong Trump
to Lean Trump
Georgia
from Toss Up Trump
to Toss Up Biden
Iowa
from Toss Up Trump
to Toss Up Biden
Louisiana
from Lean Trump
to Strong Trump
Mississippi
from Strong Trump
to Lean Trump
Nevada
from Toss Up Biden
to Lean Biden
Ohio
from Toss Up Trump
to Toss Up Biden
Pennsylvania
from Lean Biden
to Toss Up Biden
South Carolina
from Lean Trump
to Toss Up Trump
Virginia
from Strong Biden
to Lean 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:
The Electoral College Map (10/1/20)

The Electoral College Map (9/30/20)

The Electoral College Map (9/29/20)


Follow FHQ on TwitterInstagram and Facebook or subscribe by Email.

Thursday, October 1, 2020

The Electoral College Map (10/1/20)

Update for October 1.


Yesterday's update focused on states currently outside of toss up territory and while that is still the case today, there are a few battleground toss ups thrown into the mix. But even in those four states -- Arizona, Iowa, North Carolina and Ohio -- the additions served to confirm what was already known in each: Iowa and Ohio are barely on the Trump side of the partisan line and Arizona and North Carolina, while toss ups, are further away from the partisan line and in Biden's column. And as today marks 33 days until election day on November 3, that alignment is not lost -- or should not be -- on those observers of the process. A couple of Trump states are closer to changing hands (and adding to Biden's already winning projected electoral vote total) than two states that the president is absolutely going to have to have to get to 270. Each day that the narrative does not change from that -- each day that the alignment stays the same -- is a day lost for President Trump.


Polling Quick Hits:
Arizona
(Biden 49, Trump 45)
[Current FHQ margin: Biden +3.47]
Data for Progress was back in the field looking at competitive Senate races and in the case of Arizona has been less than a week since the firm last tested the presidential race. A one point Trump lead then is a four point Biden advantage now. And really, this survey is much more in line with where the FHQ averages currently peg the race in the Grand Canyon state: Biden up 48-44 (rounded).


Iowa
(Trump 47, Biden 44 via Data for Progress | Biden 48, Trump 46 via RABA Research)
[Current FHQ margin: Trump +0.39]
One can take one's pick in the pair of new Iowa polls added today. Each of the two major party candidates had a lead in one and they average out to a half a point lead for the president. In the context of all of the other calendar 2020 polling in the Hawkeye state, however, and through the lens of the graduated weighted average used here at FHQ, the president's edge is 0.39 points. There is some difference there, but not much. Likewise, Trump leads in Iowa, but not by much.


Kansas
(Trump 52, Biden 42)
[Current FHQ margin: Trump +8.66]
There have been some closer than typical polls of the presidential race in Kansas this season, but the last couple of days have brought two that have shown a less competitive contest. The new Civiqs is the second of those two and serves to stretch out Trump's lead in the Sunflower state. Neither poll is unusual in the scope of the full universe of Kansas polling, but Trump is running near the top of his range in both while Biden is toward the lower end of his. Still, there was not much change in the Civiqs series from the last poll in June to now. Biden gained a couple of points and Trump was stable at 52 percent.


Maine
(Biden 53, Trump 39)
[Current FHQ margin: Biden +12.90]
That same sort of dynamic (but in reverse) was true of the update in Maine from Data for Progress. Two months ago, Biden held a ten point lead with 53 percent of the respondents supporting him. In that same timeframe, however, Trump lost four points to fall below 40 percent. And again, that is consistent with where the president sits statewide in the Pine Tree state in the FHQ averages. Those two statewide electors as stake in Maine have never really been in doubt and sadly, this poll did not also include data on the more competitive second district. But judging from the statewide data here and other polls in a similar range, Biden likely would have narrowly led there as well.


Michigan
(Biden 52, Trump 44)
[Current FHQ margin: Biden +7.02]
Another day brings another poll in Michigan, the most frequently surveyed state of calendar 2020. And the latest from ALG Research is like a number of the other polls today: consistent with existing averages -- both candidate shares and margin -- in the state. This is the first time ALG has conducted a poll -- one that has been publicly released -- in the Great Lakes state. The big thing in Michigan now is Biden's proximity to 50 percent. Of the 20 polls conducted in the state in whole or in part in September, the former vice president was at or above 50 percent in 13 of them. Furthermore, he was at 49 percent in four of the remaining three. That is a troubling sign for a president trying to defend the state in 2020.


New Hampshire
(Biden 53, Trump 44)
[Current FHQ margin: Biden +7.96]
There have been several new polls of New Hampshire this week, a veritable flood of new data in a state that has likely been underpolled in 2020 given its position in the order of states (not to mention its close finish in 2016). But the latest three polls including the new one from American Research Group have expanded Biden's advantage in the Granite state, pushing the former vice president's average share of support at FHQ above 50 percent. Near each other on election day in 2016, New Hampshire and Michigan have both exhibited similar shifts since then up to now. Both are right in line with a seven or eight point shift toward the Democrats over the last four years.


North Carolina
(Biden 50, Trump 47)
[Current FHQ margin: Biden +1.46]
In North Carolina, the polls continue to fall in the same tight range from Biden +3 to Trump +3. The latest addition from ALG Research is on the high end of the Biden side of that range, but nevertheless in that range. It is typical then that while both candidates are running ahead of their established FHQ  averages in the Tar Heel state, Biden is running a bit more ahead than is Trump. But as has become the mantra around here, North Carolina is close and this new poll does little to shake the state from its position in the rank order. With Ohio now on the other side of the partisan line, North Carolina is the most competitive of the Biden coalition of states.


Ohio
(Biden 48, Trump 46)
[Current FHQ margin: Trump +0.13]
Speaking of Ohio, ALG Research was also in the field there (and for the first time in calendar 2020) as well. But this poll, dated yesterday on the release, was conducted as August turned to September. The addition is new, but the data is on the old side. But it fits in with other polling around the same time or those conducted since. Sure, Biden is on the high side of his range over that period in this one, but not by a whole lot. The firm nailed the Trump share. And while that cut into the president's advantage, it did not eliminate it. That said, Ohio has once again changed positions with Georgia and is now up against the partisan line, the most competitive state on the Trump side.


Oregon
(Biden 56, Trump 39)
[Current FHQ margin: Biden +14.61]
Oregon has only seen a couple of polls in 2020 and the Civiqs survey added for today is the second. The earlier survey very closely resembled the results from 2016, and although this poll has Trump stationary at 39 percent (as compared to the last survey), Biden has pushed into the upper 50s. That pulls the Beaver state closer in line with a more uniform swing toward the Democrats since 2016. It remains below average, but there is now a swing where there really was not before with just the previous poll. [Note, however, that that initial poll was factored in alongside consideration of the shifts in other states that finished near Oregon in the order four years ago.]


Pennsylvania
(Biden 50, Trump 44)
[Current FHQ margin: Biden +5.34]
Like its poll of Ohio, the final ALG Research survey release on the day from Pennsylvania was actually in the field a month ago and released just yesterday. But it is a poll that rests comfortably within the context of other recent polls of the Keystone state. FHQ currently has the race for those 20 electoral votes at Biden 49-44 (rounded), so this ALG survey is very much in line with that. Biden is approaching 50 percent there as well. Of the 26 polls conducted since (and including) the ALG poll, half of them have had the former vice president at or above 50 percent. All eyes may be on tipping point state Pennsylvania in the Trump campaign, but it is a lot like Michigan above in terms of how Biden is closing in on 50 percent. Pennsylvania is closer than Michigan, but continues to present the president's reelection effort with some significant obstacles.


South Carolina
(Trump 47, Biden 43)
[Current FHQ margin: Trump +5.67]
Lastly, in South Carolina, Data for Progress also sees a close contest for the Palmetto state's nine electoral votes. Only, things are not quite as close as the Quinnipiac poll from a day ago. Still, like that poll, this one has Trump lagging behind the 50 percent (rounded) he carries currently in the FHQ averages. Where the DfP survey is a bit more consistent is on the Biden number. His 43 percent share of support in the poll nearly matches his 44 percent share in the averages. Even with the lead, South Carolina has shifted nearly eight points in the Democrats' direction over the last four years.



NOTE: A description of the methodology behind the graduated weighted average of 2020 state-level polling that FHQ uses for these projections can be found here.


The Electoral College Spectrum1
DC-3
MA-11
(14)2
CT-7
(162)
WI-10
(253)
AK-3
(125)
AL-9
(60)
HI-4
(18)
NJ-14
(176)
PA-203
(273 | 285)
SC-9
(122)
IN-11
(51)
CA-55
(73)
OR-7
(183)
NV-6
(279 | 265)
MO-10
(113)
UT-6
(40)
VT-3
(76)
NM-5
(188)
AZ-11
(290 | 259)
MT-3
(103)
KY-8
(34)
MD-10
(86)
ME-2
(190)
FL-29
ME CD2-1
(320 | 248)
NE CD1-1
KS-6
(100)
SD-3
(26)
NY-29
(115)
CO-9
(199)
NC-15
(335 | 218)
LA-8
(93)
ID-4
(23)
WA-12
ME CD1-1
(128)
VA-13
(212)
OH-18
(203)
MS-6
(85)
ND-3
(19)
RI-4
(132)
MN-10
(222)
GA-16
(185)
AR-6
(79)
OK-7
(16)
DE-3
(135)
NH-4
(226)
IA-6
(169)
NE-2
(73)
WV-5
(9)
IL-20
(155)
MI-16
NE CD2-1
(243)
TX-38
(163)
TN-11
(71)
WY-3
NE CD3-1
(4)
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.

There were scads of new polls again today, but little to show for it in terms of changes at FHQ. Both the map and the Watch List remained unchanged from a day ago on the last day of September. But the beginning of October saw a couple of shifts in the order of states depicted on the Electoral College Spectrum. The new poll of Kansas was enough to push it past Nebraska's first congressional district and deeper into the Lean Trump category. Also on the Trump side of the partisan line and right up against, Ohio jumped Georgia and reclaimed its position as the most competitive state among the Trump coalition. But it is worth noting just how close the trio of Georgia, Iowa and Ohio are to each other and the partisan line. The three states are all within about a quarter of a point of each other in their margins and all within a half a point of the partisan line. All three and Texas are closer to the partisan line than the Biden state next to it on his side, North Carolina. That is 78 electoral votes that are closer to jumping the partisan line into Biden territory than North Carolina's 15 are to moving to Trump's side.


Where things stood at FHQ on October 1 (or close to it) in...
2016
2012
2008



--
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
State
Potential Switch
Arkansas
from Strong Trump
to Lean Trump
Georgia
from Toss Up Trump
to Toss Up Biden
Iowa
from Toss Up Trump
to Toss Up Biden
Louisiana
from Lean Trump
to Strong Trump
Mississippi
from Strong Trump
to Lean Trump
Nevada
from Toss Up Biden
to Lean Biden
Ohio
from Toss Up Trump
to Toss Up Biden
Pennsylvania
from Lean Biden
to Toss Up Biden
South Carolina
from Lean Trump
to Toss Up Trump
Virginia
from Strong Biden
to Lean 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:
The Electoral College Map (9/30/20)

The Electoral College Map (9/29/20)

The Electoral College Map (9/28/20)


Follow FHQ on TwitterInstagram and Facebook or subscribe by Email.