Election Update 10/27

So, we are in the final stretch of the race, and arguably the most integral part of it all. In 2016, the last 3-7 days of the race were where things changed radically, and if Trump is going to make a comeback, it has to be now. That said I'm going to report on the state of the race today, and then report on it again around Friday/Saturday, and finally start on my final projections on Monday, with me finalizing them Tuesday morning or so. That said, let's get into it. 

2 way: Biden +7.4%

4 way: Biden +7.7%

So Biden has declined a little bit and is under 8 points. Since national polls seem to generally speaking be more reliable and have a much lower margin of error, where I would use the number 3 instead of 4, Biden is still expected to win the popular vote with a very high degree of confidence though. Like think 99%+. Of course the popular vote isn't what wins the election, it's the electoral college, so let's take a look at that. I could theoretically remove Michigan from my projections, but I it actually takes quite a bit of time to reformat and rewrite the table when I add/remove states, so I'm going to keep everything as it is and just report the probabilities as they exist. 

States	Winner/Margin	Z score	% D Win	% R Win	EV D win	EV R win
Michigan	Biden +9.4%	-2.350	99.0%	1.0%	232	322
Nebraska CD2	Biden +6.5%	-1.625	94.8%	5.2%	233	306
Minnesota	Biden +6.0%	-1.500	93.4%	6.7%	243	305
Wisconsin	Biden +5.5%	-1.375	91.5%	8.5%	253	295
Nevada	Biden +4.6%	-1.150	87.5%	12.5%	259	285
Pennsylvania	Biden +3.8%	-0.950	82.9%	17.1%	279	279
Arizona	Biden +2.4%	-0.600	72.6%	27.4%	290	259
North Carolina	Biden +0.8%	-0.200	57.9%	42.1%	305	248
Iowa	Biden +0.8%	-0.200	57.9%	42.1%	311	233
Maine CD2	Biden +0.3%	-0.075	53.0%	47.0%	312	227
Florida	Trump +0.4%	+0.100	46.0%	54.0%	341	226
Georgia	Trump +0.4%	+0.100	46.0%	54.0%	357	197
Ohio	Trump +0.6%	+0.150	44.0%	56.0%	375	181
Texas	Trump +2.6%	+0.650	25.8%	74.2%	413	163
Alaska	Trump +6.0%	+1.500	6.7%	93.4%	416	125
South Carolina	Trump +6.3%	+1.575	5.8%	94.2%	425	122
Kansas	Trump +7.0%	+1.750	4.0%	96.0%	431	113
Missouri	Trump +7.5%	+1.875	3.0%	97.0%	441	107
Montana	Trump +8.0%	+2.000	2.3%	97.7%	444	97

Yikes. Much like my predictions from last week, we're back to the race hinging on PA with a 3.8% polling average. That raises Trump's chances back to 17.1% again with Biden's being 82.9%. So we're going down again. Florida and Georgia have also gone back to the red. Trump is gaining ground. Much like 4 years ago. My latest projection is Biden 312 - Trump 226. What's going on? Well, polls being replaced with other polls basically. Some like Georgia and Florida shifted Trump simply from older polls being removed from the count. I'm not sure that it really reflects that much of an actual change in the race. Pennsylvania, the deciding state, is also kind of weird here. Its averages are so low because the two newest polls are Trafalgar group, which leans several points more conservative than other polls, and an "insider advantage" poll, which when I click on it brings me to the "Center for American Greatness". Yeah, that seems objective /s. If we removed those two polls, Pennsylvania would be at 5.6%, which would raise Biden's odds to 92% there. This would shift the election to Wisconsin at 5.5%, which would yield a 91.5% chance of a Biden win and a 8.5% chance of a Trump win. Despite this I will not be correcting statistics as it's not what I do here. I'm not 538, if you want to see more curated averages with corrections made, check out Nate Silver's predictions. I'm just reporting the raw averages. 

One more interesting thing I want to touch on before moving on though is that Texas is declining for Trump. It's also moving to the center with recent shifts moving it to 2.6%, which means that Biden has a 26% chance of taking it, and thus getting a 400 point electoral blowout. So electoral projections are all over. Biden could only get 232 electoral votes, or he could get 413. Who knows. That's uncertainty for you. In all fairness, despite all the talk of doom and gloom and Trump winning again, if not for 2016 I don't think I would be taking it this seriously. Statistically the Biden blowout scenario is far more likely to happen at 26% vs 17% for ANY Trump win.

 Senate forecast

States	Winner/Margin	Z score	% D Win	% R Win	Seats D win	Seats R win
Michigan	Peters 6.0%	-1.500	93.4%	6.8%	46	55
Minnesota	Smith +5.8%	-1.450	92.6%	7.4%	47	54
Arizona	Kelly +4.4%	-1.100	86.4%	13.6%	48	53
Maine	Gideon +4.0%	-1.000	84.1%	15.9%	49	52
Iowa	Greenfield +1.8%	-0.450	67.4%	32.6%	50	51
North Carolina	Cunningham +1.8%	-0.450	67.4%	32.6%	51	50
Georgia 1	Perdue +1.5%	+0.375	35.3%	64.7%	52	49
South Carolina	Graham +2.3%	+0.575	28.2%	71.8%	53	48
Georgia 2	GOP +2.9%*	+0.725	23.4%	76.6%	54	47
Montana	Daines +3.3%	+0.825	20.5%	79.5%	55	46
Kansas	Marshall +4.0%	+1.000	15.9%	84.1%	56	45

 *Georgia 2 is a special election with ranked choice voting with multiple democrats and republicans on both sides. I graded the results according to party affiliation rather than a specific candidate.

So, the senate isn't changed all that much. It's divided in a way, with the Georgia seats going more in the GOP direction. Ultimately, the swing point is about the same, with Iowa and North Carolina being the swing states. My model treats the two where if one goes one way, the other does too. That said, there is going to be no tie in the forecast. That's not to say it can't happen, statistics are weird, but my model doesn't predict it happening. All in all, the democrats are given a roughly 67.4% chance of control of the senate, with the republicans given a 32.6% chance. So roughly 2 in 3 for the democrats, with a 51-49 distribution being most likely. That's about what it was last time. I will say I was watching the polls over the past 4 days, and Cunningham's result went way up then way down again and came to almost exactly what it was with my last forecast. It is what it is. 

That said, to wrap things up, this is getting interesting. I really don't know what to make of these results. I mean, when I wrote my final 2016 predictions, I'm going to be honest, Florida shifting to the Trump column like that was my first sign something was going seriously wrong for Clinton. And I was forced to change my predictions in the middle of writing my final write up of 2016 as a result. That was the first sign. The North Carolinas and Floridas of the race shifting red by less than a point. The next sign were rust belt states declining to only a few points. Pennsylvania was only up 1.9% on election day for Clinton. Michigan at 3.4, Wisconsin at 6.5. Currently, Pennsylvania is up 3.8%, which the "real" number potentially being as high as 5.6% if we remove some skewed polls from that average. Wisconsin is at 5.5%, so worse than it was in 2016. Michigan is way up though, probably because the people are mad the MAGA idiots were plotting to kidnap the governor. Those are Trump's "very fine people" for you. I don't know. We'll have to see how this continues to unfold. It's not a slam dunk for Biden, despite everyone seemingly acting like a 2016 can't happen again. At the same time the odds still skew in Biden's favor. See you in a few days, where hopefully I can make more sense of this.