So, against my better judgment I decided to update my house model.
So, to discuss it. I went by Cook PVI's house ratings, and included all districts not considered solid/safe. I labeled all districts, TU means tossup according to cook, LR/LD are leans, YR/YD are likelies, and yeah.
Then I basically looked for obviously gerrymandered districts that go the wrong way, like if something from California or Florida is like lean/likely red/blue and then goes the other way, I flipped them back the way cook had them. This is an measure to compensate for gerrymandering. Since my model takes 2024's districts and shifts them uniformly so many points, and some districts arent even the same any more, I declared a few results invalid and flipped them back. This gives the GOP a net +2 advantage over the raw data, which I included in the results.
With that said, my model has the dems winning at 231-204, with a 96% chance of victory. A bit lower than my 235/98% stats in the old model, but that's why i rebuilt it, to update it considering how districts have changed. This seems like a possible result.
No comments:
Post a Comment