Now that the votes are in, let's re-visit the swings in relative vote margin for states from 2016 to 2020. See this thread for a primer, which describes the swing relative to the national vote margin.

https://twitter.com/MJRosenbergDad/status/1328768268891467776

1/

Remember: we're looking at the difference between the Republican margin and the national margin ("R+[]") and the change between elections. This is different from the Cook PVI that uses difference in vote share instead of margin (my numbers are ~2x larger, but same idea).

2/

Here are the numbers, sorted by the shift in Republican vs. national margin between 2016 and 2020 (negative number means a state moved towards the Democrats more than the national margin)

3/

More numbers

4/

OK, what are the takeaways?

Between 2016 and 2020, states that became the most increasingly favorable for Dems are largely in the northeast (New England + DE/MD/VA), plus highly-educated suburban areas like NE-2 and Colorado (my favorite state, long gone as a swing state).

5/

Minnesota and Oregon also became more Dem-friendly, with MN especially bucking the trend between 2012 and 2016.

Kansas and Alaska also continued a leftward jog, though not enough to matter.

6/

The big impact here was obviously Georgia, which needed every bit of its 3% pro-Dem swing, on top of the national improvement of Biden over Clinton.

7/

Though big swings were expected in Arizona and Texas, they only became about 1-2% more Dem-friendly relative to the nation as a whole (vs. 7-9% swings from 2012 to 2016), which was enough to swing AZ, but not nearly enough in TX, which remains R+10.

8/

Interestingly, the Blue Wall states of MI/WI/PA did not shift much at all (MI 0.7% towards the left, PA 0.5% and WI 1.0% towards the right). These all went back to Biden on the strength of his overall national performance, rather than particular strength in that region...

9/

...though perhaps you could argue Biden successfully arrested the rightward slide of those states, which went 4-8% to the right between '12 and '16.

North Carolina also showed no change over the last 3 cycles, sticky at R+6. This was unexpected after Obama won it in '08.

10/

Iowa and Ohio continued a march to the right by 1-2% each, now around R+12.5%. The days of these states mattering in a national race are long gone. These are the anti-Colorado.

11/

Several other states had notable swings towards Trump: irrelevantly, DC, NY, IL, HI, and CA. More relevantly, NV (two cycles in a row swinging towards the Republicans). Most relevantly, Florida, which at nearly R+8 has to be considered a light-red state rather than a tossup.

12/

Brief aside: This data highlights just how cruel it was that the first state to report was Florida, by far the Dems' worst and most consequential shift of this cycle...

13/

...Had PA reported a 1.2% win for Biden out of the box, we would have had a very clear snapshot of the race and known very early that it was likely to be a ~4-5% popular vote margin and a narrow win in the likely tipping point states.

14/

Where does this leave us for 2024?

First, what states are likely to be in play? The last 6 popular vote margins were between R+3 and R-7, so any states between about R-4 and R+8 should be considered in play.

15/

If there are no shifts, that would be any state between NH and FL on this list sorted by 2020 R-margin.

But let's look at shifts on the fringes. NM, VA, and CO have been steady or moving towards the Dems. Those are out of play.

16/

Maine has been very jumpy and could be in play.

The Blue Wall states are slightly redder than the nation, but the trend is ambiguous. These will be highly contested. Nevada is also trending slightly redder.

17/

Georgia and Arizona are slightly redder than the nation, but getting bluer, but not clear how much farther blue they can go. These will also be hotly contested.

18/

North Carolina and Florida are at the edge of this range. NC isn't budging, so is winnable for a D, but not likely to be the tipping point. Florida is a tough ask for Dems and we may find it out of reach in 2024.

19/

Texas is getting bluer, but not nearly fast enough. It may be in play in a Dem landslide scenario, but again won't be near the tipping point.

20/

So where is the tipping point likely to be? In 2012, this was Colorado, R-1.5. In 2016, it was Wisconsin, R+2.9. In 2020, it was WI or PA, around R+3.5.

Based on these numbers and trends, it is likely to be an R+2-4 state, one of MI/WI/PA/NV/AZ/GA.

21/

A big caveat, of course, is that these trends are not definitive. Based on 2012 --> 2016, you would have expected Michigan to continue racing to the right. It actually drifted back slightly to the left.

22/

Dems did well "fighting the last war" in 2020, winning back the Blue Wall. They could do it in 2024 by going hard after Latino voters and countering disinformation, putting Texas or Florida in play.

23/

But overall, given how far the fringy states are from the core tipping point states, and the lack of major swings into or out of that range, it is likely that the 2024 race will be decided by largely the same states that decided 2020.

24/24