That being said, I am going to "herd" here a little bit with my final numbers.
I have spent most of the election cycle under the statistical assumption that the final voting numbers would show a slightly improved Democratic turnout from the 2020 election. This was based on some previous formulas that maybe do not make much sense anymore. With registration numbers and more people identifying as Republicans I am coming to the conclusion that we are probably headed towards something similar to the 2020 election when Democrats held a one-point advantage. In fact, in looking closely at the pollsters, many of them are assuming something closer to an even number of both Democrats and GOP voters or even a smallish GOP advantage. I may still be erroring on the side of greater Democratic turnout.
So these are probably my final numbers. I may not even update anything on the spreadsheet anymore since I seem to have more than enough data (way more than any previous election cycle) to feel comfortable. Call it burnout or call it a lot of work to see the model move by 100's of a percent. Either way. So let's start with this:
According to the average of the 35 pollsters, Trump is polling just under 91% of the GOP vote, Harris is polling just slightly better at 91.25. Trump is polling at about 5.5% of the Democratic vote, while Harris is polling at a slightly higher 6% of the GOP vote. Trump leads the independent vote by just over a half percent. What all of this means is that if the election was exactly split with the same number of Democrats and Republicans showing up with a typical number of independents, then the election would be almost a wash (Harris is up 0.16%). So with sort of partisan voting average, the entire election really comes down to turnout. Using those same averages, these would be the numbers based on various voter turnout models. If either side at this point goes higher than 2% then it is likely that side wins across the board.
D+2 - Harris + 1.90
D+1 - Harris + 1.05
Even - Harris + 0.16
R+1 - Trump +0.72
R+2 - Trump +1.58
As you can see, there is a slight advantage for Harris, simply because she is picking up that half percent of GOP voters and keeping that half percent of Democratic voters. The half percent advantage for Trump with the Indies does not entirely make up for the larger demographics of the two parties. Before adding the new crosstabs into the formula from all the last minute, this had been almost even, but a couple of pollsters showed some changes in the breakdown of GOP and DEM voters that shifted towards Harris.
Now this is where there has been a bit of a disconnect. For the first time in this election cycle, the Hill/DDHQ is now with RCP in showing Trump winning in the national popular vote (albeit only by 0.2%). But while the pollsters are moving in Trump's direction, my spreadsheet has moved ever so slightly (0.2%) towards Harris. This tells me that some of these pollsters are moving their projected partisan breakdown a bit towards the GOP. This could be based on an increase in Republicans responding, it could be based on the "likely voter" calculations becoming slightly better for Trump, or they could be just correcting an earlier belief that they no longer have.
So looking back on my final numbers, I am hedging a bit by supplying all of the possible partisan advantages in both directions. But if you put a gun to my head right now, I would say that the D+1 number of 1.05% in favor of Harris is my best guess. I would also argue that her winning the popular vote by around 1% will not be enough for her to carry the battleground states she would need. I still am not convinced that she has done enough in those states to make up for the 4% difference between the national and battleground partisan make-up. In fact, according to every polling aggregate she would have to "overperform" the polling in the battleground states to get to 270. Trump leads everywhere in Georgia, North Carolina, Arizona, Nevada and most importantly... Pennsylvania.
At the end of the day, there are real problems for Harris in several battleground states, including some last-minute problems. There have been some big Palestinian groups who have openly endorsed Trump over the past few days as well as other Palestinian Democrats who have refused to endorse their own candidate. The early voting numbers are especially problematic in North Carolina, Pennsylvania, and Nevada, and not particularly good for them anywhere. There is an overall lag in minority early voting across the board. The comedian October surprise that Harris and gang tried to manufacture has come back to bite them with a couple of really stupid statements by the sitting Democratic President about trash and about how he apparently wants to slap MAGA republicans on their collective asses. Harris has been playing the run-away from those comments game at exactly the wrong time.
Lastly, and maybe it is just me, but watching the various college football games where I have seen the two candidates air their ads, I am seeing a lot of Trump ads featuring real life examples of what he has done versus what Biden/Harris has done. Meanwhile, you get the Harris ads of "promises" about what she would do as well as "predictions" about what Trump will do. Again, perhaps this is just my opinion, but Trump ads (both positive and negative) seem more effective when put into context that he is President Trump while she is just Kamala Harris. For Trump, he just wants you to compare the past four years to the previous four. For Harris, she wants you to forget about those eight years and listen to how different she will be if elected.
I understand I have a certain bias here and perhaps it is the cynic and troubleshooter in me that takes sides on this. But I believe in someone who can show me what has actually been done, over someone who wants to tell me what they promise to do. I would guess that many Americans also hold that sort of logic.
Comments