So, this is the last NORMAL update. The next update I'll do on monday, and then I'll revise it tuesday as the data changes before the polls close. That will be the FINAL update, similar to my 2016 and 2020 final updates.
Presidential
So, I dont trust the national popular vote average at all. I suspect herding there. I mean, this entire election cycle, the popular vote has been 1.5-2 more D than the tipping point vote and then they converge last minute? Nah, pollsters are hedging their bets and dont wanna be wrong. It's dishonest, it's hacky, and i condemn it. People should post the data, whatever it is, don't do funky things with it, and let it speak for itself. That's how you get truth. These guys are just copying off of each others' homework. Bleh.
Heck, the same can be said for the election in general. I don't trust the polls much at all right now. I think most pollsters are herding, although attempting to counter for that by looking at the polling ranges and averaging the two most extreme results...gets me an average not all that different from what we see here. WI and MI are more blue, but otherwise it's kinda close.
So the electoral college. I would say outside of the swing states, and even among the sun belt, we're seeing a shift toward trump. MN,NM,NH, and VA have all become a bit less left leaning. IA, TX, FL, OH and ME2 are all a bit more right leaning.
As far as the swing states go, we're seeing a bifurcation that goes in the direction I've suspected the race would fall all along. The sun belt is going red. AZ and GA have been on that trend for weeks. NV and NC in the past couple days have started following. But at the same time, a path remains available for Harris. WI and MI are now in the blue column (and if anything I think the polling underestimates their margins due to my "anti herding" exercise, and these states are actually lean D, not tilt D). And that leaves PA.
PA...it all comes down to us (I say this, as someone from this great state). This election is probably either gonna be 251-287 Trump, or 270-268 Harris. Other outcomes are possible, see my simulation exercise, and a systematic overperformance in either direction will radically change the map.
If Harris overperforms even by 1 point, it's 270-268 Harris, gg no re. If Trump overperforms by 1 then WI and MI fall to trump and he sweeps all 7 226-312. If Harris fails to take PA, she's gonna have to rely on some combination of the other 4, with NV/NC being the easiest path. This is significantly less likely than just winning PA.
And of course, this is why Trump is already throwing up signals of the vote being rigged in PA. He's trying to push his "big lie" again by casting doubt on the election results, meaning if PA doesn't go to him, then he's gonna throw a hissy fit and scream the whole thing was rigged again. And I saw a poll yesterday saying 72% of trump supporters WONT accept the result if trump won't win. So yeah, he's gonna try to pull a January 6th again. Just you wait.
But yeah. Still, despite Trump having the advantage, it's a very thin one. It comes down to a 0.4% average in his favor in one state. If Harris overperforms, she wins the election. It really is almost 50-50 here. And even though I kinda like to count the sun belt out, you technically can't. Because if Harris overperforms by 3 points, she sweeps all 7 states herself. See the exercise i did the other day.
And yeah, that's where we're at.
Senate
We're screwed in the senate. The most likely outcome is 49-51R, the osborn polling dropped since switching to RCP's numbers, and it's in the same category as texas and florida. We would need an 4+ point overperformance to win it, while if the GOP did that to us, we'd be wiped out and the senate would be like 44-56R. The GOP has the same likelihood of crushing us and wiping out the rust belt + arizona, as we do of taking texas, nebraska, or florida. We're screwed here. I'm just gonna say it.
House
In the house things seem...cautiously optimistic. We actually have a nominal lead in the house right now, with us winning 218-217 with a 51% chance, but...let's face it, it's 50-50. Still. Just like with the presidency, we have a shot here.
Simulations
I did 100 for each...
Presidency- 47D,53R,0T
Senate- 2D,96R,2T
House- 24D,76R
It's weird, for the presidency the race is a lot closer than it was in my simulation of 100 screenshotted outcomes. Not sure what's up with that. I did them in samples of 10 so that may have made a difference, but yeah it will prefer trump in the short term, but if you keep spamming the button in one session it comes out even.
And yeah, it's about as split as you'd expect. The race really is 50-50 roughly.
The senate, on the other hand, has the GOP dominating the outcomes. As I said, the odds of us winning the senate, or even cracking a tie, are REALLY BAD. And the simulator makes it even worse.
As for the house, it keeps favoring the GOP despite a nominally tied forecast. Might be the probabilities of individual states. I keep saying it but on the chart the GOP simply has more of a chance to flip stuff, since i assume states are safe if not considered "in play" and backed by actual polling data.
Conclusion
Well, it looks like in the house and the presidency, the race is coming down to the wire. In the electoral college, the fate of the free world is in the hands of Pennsylvania and the other rust belt states. In the house, the chamber is split almost 50-50 and I expect a close result either way.
The only thing I'm certain about this point is the senate looks REALLY BAD for democrats. They CAN win or tie this thing. They still have that 1 in 7 chance, but it would require a massive overperformance and I just dont have enough hopium to predict that. It's within the possible range of results my prediction implies, but that outcome is not likely. I'm just gonna call it for the GOP there.
And yeah. That's where things are at. I'll start my final prediction on monday, although I may edit it as the day goes on or into tuesday if the data changes. You know how it is.
No comments:
Post a Comment