When will somebody put up a new state PVI chart?

[Torie]

Yes, both for just 2016 (needed since the map was just so altered this year, and may well be the harbinger for the future given the nature of Trumpism), and per the usual mythology, an average of 2012 and 2016? I want it, and I want it now. Thank you.

Do it yourself you say? Response: I'm too lazy and presently distracted to do that. Thanks again.

[Orser67]

I've thought about doing it, but I'm not going to until we have the final election results. There are still lots of uncounted ballots

[🦀🎂🦀🎂]

will Cook do it?

[100% pro-life no matter what]

I already did.  The numbers may go slightly more Republican if Hillary's margin in the PV goes up more.

https://uselectionatlas.org/FORUM/index.php?topic=251899

Using the Cook formula, which considers a Republican winning 51-49 in a 50-50 election R+1, not R+2, and which goes back two cycles.

I'm using a national PV of Clinton +0.5, but it's unclear of its exact margin, of course!  And, I'm baking the national margins into these!

Alabama: Romney +24.1, Trump +28.8; AVG: 26.45 --> R+13
Alaska: Romney +17.9, Trump +15.7; AVG: 16.8--> R+8
Arizona: Romney +13.0, Trump +4.8; AVG: 8.9--> R+4
Arkansas: Romney +27.6, Trump +27.1; AVG: 27.35--> R+14
California: Obama +16.2, Clinton +27.7; AVG: 21.95--> D+11
Colorado: Obama +1.5, Clinton +1.6; AVG: 1.65--> D+1
Connecticut: Obama +13.4, Clinton +11.7; AVG: 12.55--> D+6
Delaware: Obama +14.7, Clinton +11.0; AVG: 12.85--> D+6
Florida: Romney +3.0, Trump +1.8; AVG: 2.4--> R+1
Georgia: Romney +11.7, Trump +6.2; AVG: 8.95--> R+4
Hawaii: Obama +38.8, Clinton +31.7; AVG: 35.25--> D+18
Idaho: Romney +35.8, Trump +32.1; AVG: 33.45--> R+17
Illinois: Obama +13.0, Clinton +15.5, AVG: 14.25--> D+7
Indiana: Romney +14.1, Trump +19.8, AVG: 16.95--> R+8
Iowa: Obama +1.8, Trump +10.1: AVG: 4.15--> R+2
Kansas: Romney +23.6, Trump +21.5; AVG: 22.55--> R+11
Kentucky: Romney +26.6, Trump +30.3; AVG: 28.45--> R+14
Louisiana: Romney +21.1, Trump +20.2; AVG: 20.65--> R+10
Maine: Obama +11.4, Clinton +2.2; AVG: 6.8--> D+3
Maryland: Obama +22.2, Clinton +24.7; AVG: 23.45--> D+12
Massachusetts: Obama +19.2, Clinton +26.8; AVG: 23.0--> D+11-12
Michigan: Obama +5.6, Trump +0.8; AVG: 2.4--> D+1
Minnesota: Obama +3.8, Clinton +0.9; AVG: 2.35--> D+1
Mississippi: Romney +15.4, Trump +19.0; AVG: 17.2--> R+9
Missouri: Romney +13.3, Trump +19.6; AVG: 16.45--> R+8
Montana: Romney +17.6, Trump +21.0; AVG: 19.3--> R+10
Nebraska: Romney +25.7, Trump +26.8; AVG: 26.25--> R+13
Nevada: Obama +2.8, Clinton +1.9; AVG: 2.35--> D+1
New Hampshire: Obama +1.7, Trump +0.3; AVG: 0.7--> EVEN
New Jersey: Obama +13.9, Clinton 12.3; AVG: 13.1--> D+7
New Mexico: Obama +6.3, Clinton +7.8; AVG: 7.05--> D+4
New York: Obama +24.3, Clinton +20.8; AVG: 22.05--> D+11
North Carolina: Romney +5.9, Trump +4.3; AVG: 5.1--> R+3
North Dakota: Romney +23.5, Trump +36.8; AVG: 30.15--> R+15
Ohio: Romney +0.9, Trump +9.1; AVG: 5.0--> R+2-3
Oklahoma: Romney +37.3, Trump +36.9; AVG: 37.1--> R+19
Oregon: Obama +8.2, Clinton +10.1; AVG: 9.15--> D+5
Pennsylvania: Obama +1.5, Trump +1.7; AVG: 0.1--> EVEN
Rhode Island: Obama +23.6, Clinton +14.1; AVG: 18.85--> D+9
South Carolina: Romney +14.4, Trump +14.6; AVG: 14.5--> R+7
South Dakota: Romney +21.9, Trump +30.3; AVG: 26.1--> R+13
Tennessee: Romney +24.3, Trump +26.7; AVG: 25.0--> R+12-13
Texas: Romney +19.7, Trump +9.7; AVG: 14.7--> R+7
Utah**: Romney +51.9, Trump +19.5; AVG: 35.7--> R+18
Vermont: Obama +31.7, Clinton +28.0; AVG: 29.85--> D+15
Virginia: Tie, Clinton +4.4; AVG: 2.2--> D+1
Washington: Obama +11.0, Clinton +17.0; AVG: 14.0--> D+7
West Virginia: Romney +30.7, Trump +42.7; AVG: 36.7--> R+18
Wisconsin: Obama +3.1, Trump +1.5; AVG: 0.8--> EVEN
Wyoming: Romney +44.7, Trump +48.1; AVG: 46.4--> R+23
**Utah could be off by a little more due to high third party vote making my easier, but less exact, calculations slightly further off.

Our two largest states, California and Texas, both trended Democratic massively, allowing most of the rest of the country to trend Republican relative to the nation as a whole.  Iowa flipped all the way from D+1 to R+2 (R+5 just based on 2016).  Pennsylvania, Wisconsin, and New Hampshire are now even.  PA is technically just to the right of the nation as a whole over the last two elections, while the other two are just to the left of the nation as a whole.  The swing states lined up favorably for Trump, with Iowa, Pennsylvania, Michigan, Wisconsin, and New Hampshire (plus ME-02) all flipping from more Democratic than the nation as a whole in 2012 to more Republican than the nation as a whole in 2016.

[Nyvin]

That's really terrible for Democrats....it literally looks like every state that already has a huge dem margin (CA, MA, MY, WA, etc) got even more Democratic,  while a lot of the states that were on the fringe became more Republican.    The Democrats really need to be more than the Party of California.

A few of the bright spots are VA, FL, TX, AZ, and GA though.  

[cinyc]

I'll probably update the county map PVI project when we get official results.  No sense doing so before then, as it would have to be redone, and it gives me something to do in the off-season.  

Most PVI bellwether counties weren't good bellwethers this cycle, anyway, so I question its usefulness.  I suspect we're going to get a lot of trending in the new PVI counties from last cycle to this, too. 

I haven't checked all of the national PVI bellwethers, but Washoe County, NV may have done better than many of the others at predicting the national margin.  Educated counties like Loudoun, VA and Los Alamos, NM were way off as those with college degrees skewed Clinton.

[muon2]

Here's my table based on the current numbers on Leip's page.

State	2012 PVI	2016 PVI	Net
AL	R+14	R+14	0
AK	R+12	R+9	D+3
AZ	R+7	R+5	D+2
AR	R+14	R+15	R+1
CA	D+9	D+12	D+3
CO	D+1	D+1	0
CT	D+7	D+7	0
DE	D+8	D+6	R+2
DC	D+40	D+43	D+3
FL	R+2	R+1	D+1
GA	R+6	R+5	D+1
HI	D+20	D+18	R+2
ID	R+18	R+19	R+1
IL	D+8	D+7	R+1
IN	R+5	R+9	R+4
IA	D+1	R+2	R+3
KS	R+12	R+13	R+1
KY	R+13	R+15	R+2
LA	R+12	R+11	D+1
ME	D+6	D+3	R+3
MD	D+10	D+12	D+2
MA	D+10	D+12	D+2
MI	D+4	D+1	R+3
MN	D+2	D+1	R+1
MS	R+9	R+9	0
MO	R+5	R+9	R+4
MT	R+7	R+10	R+3
NE	R+12	R+14	R+2
NV	D+2	D+1	R+1
NH	D+1	EVEN	R+1
NJ	D+6	D+7	D+1
NM	D+4	D+4	0
NY	D+11	D+11	0
NC	R+3	R+3	0
ND	R+10	R+16	R+6
OH	R+1	R+3	R+2
OK	R+19	R+19	0
OR	D+5	D+5	0
PA	D+1	EVEN	R+1
RI	D+11	D+10	R+1
SC	R+8	R+8	0
SD	R+10	R+14	R+4
TN	R+12	R+13	R+1
TX	R+10	R+8	D+2
UT	R+22	R+20	D+2
VT	D+16	D+15	R+1
VA	EVEN	D+1	D+1
WA	D+5	D+7	D+2
WV	R+13	R+19	R+6
WI	D+2	EVEN	R+2
WY	R+22	R+25	R+3

[Torie]

Thanks Muon2. Just looking at MI, how can it be Dem +1%, when it is dead even, and Trump will end up losing the popular vote by about 1.5%?  That translates to me into Pub +1%. Or are both columns averages of two election cycles?

[100% pro-life no matter what]

Thanks Muon2. Just looking at MI, how can it be Dem +1%, when it is dead even, and Trump will end up losing the popular vote by about 1.5%?  That translates to me into Pub +1%. Or are both columns averages of two election cycles?

Cook PVI will be a combination of 2012 and 2016, just as the current one is 2008 and 2012.

[muon2]

Thanks Muon2. Just looking at MI, how can it be Dem +1%, when it is dead even, and Trump will end up losing the popular vote by about 1.5%?  That translates to me into Pub +1%. Or are both columns averages of two election cycles?

The 2012 column has the values on the Cook PVI Wikipedia site. It's based on the average of 2008 and 2012. I used the 2012 votes for each state from Leip's page to get the two-party fractions and averaged them with two-party fractions from the 2016 page. I'll update it after all the states are finalized.

In the case of MI the Pub two-party share was 45.2% in 2012 and 50.2% in 2016. The national Pub shares were 48.1% and 49.6%, with an average of 48.8%. The new MI PVI is (45.2% + 50.2%)/2-48.8% = -1.1% or D+1.

[muon2]

I've updated my chart with the latest numbers from the Atlas. Here's a map version of the shifts. The regional changes are pretty dramatic.

[Torie]

In your chart, Muon2, for PA how does the average of D+1 and EVEN average to R +!? PA wasn't even for 2016. It must be a typo.

[muon2]

In your chart, Muon2, for PA how does the average of D+1 and EVEN average to R +!? PA wasn't even for 2016. It must be a typo.

In my chart PA is EVEN after the 2016 results. The 2012 Pub two party fraction nationally was 48.1% in 2012 and 49.2% in 2016, for an average of 48.65%. The 2012 Pub two party fraction in PA was 47.3% and 50.6% in 2016. The average or PA is 48.95%. The difference is R+0.3 which rounds to 0. Did I misread something off the Atlas data?

The map shows the shift in PVI which comes from the last column in my chart. PA was D+1 after 2012 and moves to EVEN, so the shift is R+1.

[Torie]

Oh, the third column is the delta function. Who knew?

[muon2]

The table tool on the forum doesn't give ideal spacing. I put a heading of "Net" on the last column, but the tool tries to run it into the prior column.

[Nyvin]

How is Virginia just D+1?   Hillary won it by over five points.

[Lachi]

How is Virginia just D+1?   Hillary won it by over five points.

Because he had VA as a tie for Obama in his original calculations, when he really won by 3.8

Extremerepublican has a lot of errors for 2012 in the calculation, with major ones being that he has ROMNEY winning Ohio, Florida, as well as having much weaker margins in general for Obama that what actually happened, while the 2016 ones should be updated.

[Nyvin]

How is Virginia just D+1?   Hillary won it by over five points.

Because he had VA as a tie for Obama in his original calculations, when he really won by 3.8

It was even in 2012, that makes sense.   But Hillary won the two party vote in VA by a larger margin than Obama did while winning the national PV by a smaller margin (almost half).   D+1 just doesn't make sense in 2016.

[Lachi]

How is Virginia just D+1?   Hillary won it by over five points.

Because he had VA as a tie for Obama in his original calculations, when he really won by 3.8

It was even in 2012, that makes sense.   But Hillary won the two party vote in VA by a larger margin than Obama did while winning the national PV by a smaller margin (almost half).   D+1 just doesn't make sense in 2016.

I updated my original post showing what is wrong in the calculations.

[100% pro-life no matter what]

How is Virginia just D+1?   Hillary won it by over five points.

Because he had VA as a tie for Obama in his original calculations, when he really won by 3.8

Extremerepublican has a lot of errors for 2012 in the calculation, with major ones being that he has ROMNEY winning Ohio, Florida, as well as having much weaker margins in general for Obama that what actually happened, while the 2016 ones should be updated.

Those numbers are relative to the national PV.  Florida and Ohio did vote to the right of the nation in 2012.  It definitely still needs to be updated for changes in 2016 since a couple weeks ago.

[muon2]

How is Virginia just D+1?   Hillary won it by over five points.

VA in 2012: R 47.3%, D 51.2%, R two party share 48.0%
VA in 2016: R 44.4%, D 49.8%, R two-party share 47.1%
Average R two-party share 47.55%

The R national two-party share was 48.1% in 2012 and 49.2% in 2016, for an average of 48.65%.

The difference is D+1.1% which rounds off to D+1.

That's how the Cook PVI is calculated. It's not the winning margin.

[Nyvin]

How is Virginia just D+1?   Hillary won it by over five points.

VA in 2012: R 47.3%, D 51.2%, R two party share 48.0%
VA in 2016: R 44.4%, D 49.8%, R two-party share 47.1%
Average R two-party share 47.55%

The R national two-party share was 48.1% in 2012 and 49.2% in 2016, for an average of 48.65%.

The difference is D+1.1% which rounds off to D+1.

That's how the Cook PVI is calculated. It's not the winning margin.

Okay, but 47.1% - 49.2% is still -2.1% for the 2016 numbers alone,  for 2016 it should be D+2 from what I'm seeing.

Just like in 2012 the Republican two party vote share was 48% both nationally and in Virginia, so the state had a PVI of 0. 

[Nyvin]

Unless I'm missing something Texas should be R+6,  here's the numbers from this website:

TX 2016
Total two party vote: 8,551,643   
Hillary: 3,868,291   (0.452344772)
Trump: 4,683,352   (0.547655228)

National 2016:
Total two party vote: 126771042
Hillary: 64,433,399 (0.50826591)
Trump: 62,337,643 (0.49173409)

so 54.8% - 49.2% = 5.6% (rounded to R+6)

Again this is just for 2016, not 2012 and 2016.

[100% pro-life no matter what]

Unless I'm missing something Texas should be R+6,  here's the numbers from this website:

TX 2016
Total two party vote: 8,551,643   
Hillary: 3,868,291   (0.452344772)
Trump: 4,683,352   (0.547655228)

National 2016:
Total two party vote: 126771042
Hillary: 64,433,399 (0.50826591)
Trump: 62,337,643 (0.49173409)

so 54.8% - 49.2% = 5.6% (rounded to R+6)

Again this is just for 2016, not 2012 and 2016.

Official PVI is a 2-cycle rolling average

[muon2]

How is Virginia just D+1?   Hillary won it by over five points.

VA in 2012: R 47.3%, D 51.2%, R two party share 48.0%
VA in 2016: R 44.4%, D 49.8%, R two-party share 47.1%
Average R two-party share 47.55%

The R national two-party share was 48.1% in 2012 and 49.2% in 2016, for an average of 48.65%.

The difference is D+1.1% which rounds off to D+1.

That's how the Cook PVI is calculated. It's not the winning margin.

Okay, but 47.1% - 49.2% is still -2.1% for the 2016 numbers alone,  for 2016 it should be D+2 from what I'm seeing.

Just like in 2012 the Republican two party vote share was 48% both nationally and in Virginia, so the state had a PVI of 0. 

If you only used 2016 then you would get D+2. The same measure in 2012 alone gives EVEN. The official numbers use a two cycle average to reduce one-time effects of a particular candidate or event in a state or district.