One potential miscalibration in the conventional wisdom on redistricting is not accounting for the partisanship of states that have/not passed maps. In 2020, the median partisan lean of districts in states which HAVE passed new boundaries was D+1. Was R+14 in states that haven't.
This just mean that we should expect the seats that haven't been redrawn to be muuuuch redder than the seats that have -- and the plans there maybe more biased toward Republicans. Court cases in OH, NC make this a little tricky, but if you look at the most likely maps...
... then you get the following changes v 2020:
In states that *HAVE* passed new maps: Democrats have gained 12 seats that lean toward them (counting competitive seats).
After adding likely maps is the remaining states: Dems will gain around 8 versus their 2020 map.
That puts the number of seats Democrats would win the presidential popular vote in in a completely tied election at 213, up from 205. So, big Dem against — the fairest map in decades, probably — but still biased towards Rs.*
*Depending mainly on new maps in NC, OH, and FL.
Yes, “gains, sorry I have bad eyes and fat fingers — a bad combo for tweeting on my phone
NEW: Here is our big story on US Congressional redistricting so far.
Nationally, Democrats have fared surprisingly well in the maps passed so far. But they are still biased toward Republicans — the median seat will be about one point to the right.
Much will depend on what happens in the 14 states that haven't passed new maps (or were told by courts to redraw them). Here is how one measure of fairness — the efficiency gap — has changed in each state (or, for states without final maps in the average plan under consideration)
A big story of the 2022 redistricting cycle is Democratic gerrymandering. But the data show Republicans are playing an entirely different game when it comes to packing and cracking partisans into seats. Nationwide, about 3% (8m) more Democratic votes are "wasted" than GOP votes.
.@baseballcrank, if you think Democrats and Republicans shouldn’t be doing partisan gerrymandering, there are plenty of proposals for reforms from the last ~decade that you + your mag wrote against that you might reread and find newly fruitful. Your hypocrisy here is my point
Am hearing more and more that Dems have gerrymandered their way to a balanced national House map. Will have more on this next week, but 2 points are (a) this ignores the role of litigation + independent commissions & (b) that there is still huge asymmetric bias at the state level
As a reference pt, states that Biden won by more than 5pts in 2020 had an efficiency gap of 5% under the old lines. Republicans rates had an average efficiency gap closer to 15 (2x the proposed threshold for litigation). After redistricting, Ds might be closer to 9, and Rs 14-15.
(Note, of course, that lots of maps aren't finalized yet so these calculations require some educated guesses, and numbers will change slightly)
This morning, @TheEconomist launches a polling model for the upcoming French presidential election. Polls there are pretty good, and Macron has a good chance of winning.
We have tried some new (and some old) things in communicating uncertainty to readers: economist.com/France2022
You can read an explanation of our methods (and see some cool additional charts) here:
The data are president-day pairs, so actually quite large from a raw n perspective -- the problem is that we have a limited data set to capture changing voter psychology (we have ~2.5 presidential terms of high partisanship)
the biden supporter version of january 6 is nancy pelosi kneeling while the cast of hamilton, sitting on the barrier of the gallery, yell-sings “history has its eyes on you” at kevin mccarthy