Election Electoral College predictions (10 Viewers)

[Richard]

To get our minds off of the virus, debates and campaign theatrics for a moment, I thought it might be an interesting exercise to post our predictions of the final Electoral College margin. Of course, any prediction made would be a moment-in-time snapshot of what we think will happen given the current circumstances. Any number of factors could influence the vote over the final month of the campaign.

There are sites that can help you arrive at your prediction. Real Clear Politics and 538 have polling data that can help you come up with a number. I use RCP mostly because they have a good interactive national map that's easy to use and they don't include polls conducted by companies like SurveyMonkey. They do, however include partisan polling, like Trafalgar and Change Research.

My prediction started with looking at a best case scenario for a Trump reelection. I started by putting all of the states shown to lean in one direction in that camp. RCP shows states like Indiana and Missouri leaning Trump and Oregon and Virginia leaning Biden when those states and others like them are certainly going to end up in those columns. That leaves 11 "battleground" states (and two one-vote congressional districts) that theoretically could go either way and gives a picture of a 226-125 Biden lead before assigning a winner in those states. In looking at Trump's best case scenario, I gave him every state in which he is marginally ahead and those he is behind by less less than five points, as well as the two congressional districts. That would provide Biden with a slim 278-260 win. To come out with a better scenario for Trump, you would need to assign one or more of Michigan, Nevada, Pennsylvania, Wisconsin or New Hampshire to Trump and he is trailing in all of those states by more than five points.

However, I don't believe that Trump will achieve his best case numbers and Biden could win several more states, like Arizona, Ohio, Florida, Iowa, North Carolina, Georgia and Texas. I still think Trump will likely keep Texas, Iowa and Georgia, so my prediction as of today is 353-185 in favor of Biden.

If you choose to participate in this exercise, you are welcome to give different predictions over the course of the next month. I intend to post a final prediction on Nov. 2/3 and may even post a weekly prediction, depending on the level of participation.

 

[bdb13]

The latest polls are tightening significantly. It seems the effect of Trump's really bad awful week are wearing off. Biden needs another performance in the debate this week like Trump had in the last one. However, I'm actually not sure that will work this time because in typical Trump fashion people are getting numb to his lunacy.

Eh, guess it's how you look at it.. I think I'd moreso say they've tightened a little bit, with Trump still running a little behind relative to where he was prior to his awful week. Still need more polling by the big players to have a better idea though (NBC still had Biden up +11 nationally 3 days ago), which I'm sure we'll get in the next couple of days.

 

[CoolBrees]

Where are you seeing tightening?

If you look at the RCP 14 day averages compared to the 7 day for both the national and battlegrounds it is really static. The numbers of undecideds is certainly going down but is seems to be fairly even evenly portioned

 

[bdb13]

Pertinent to what we're talking about here:

 

[Heathen]

And despite all of this, the message needs to be to get out and vote. The polls were wrong last time, so don't get complacent. For the sake of this country we call home, please please please vote.

Absolutely.

Turned my ballot in the morning after I got it. Won't make much of a difference here in Utah that is solid red but I'm hoping over the next decade or so we'll get closer to flipping.

 

[SaulGoodmanEsq]

Where are you seeing tightening?

If you look at the RCP 14 day averages compared to the 7 day for both the national and battlegrounds it is really static. The numbers of undecideds is certainly going down but is seems to be fairly even evenly portioned

Pennsylvania, Wisconsin, North Carolina, and Arizona have gotten closer over the last week or so. The first two still show 'comfortable' leads for Biden on paper but since I'm subtracting 2-3% given issues with mail-in ballot spoilage I'm not quite confident those two states are a sure thing.

 

[Saint by the Bay]

Since Oct. 12 aggregate polling in Pennsylvania has tightened by 3 pts, Michigan by 1 pt, and Florida by 2 pts.

That's right in line with my expectations from the Hunter Biden stupidity. It's going to have a 2-3 pt effect on the race which may be just enough. I think Biden can get that back if Trump makes a fool of himself again in the debate this week, but this race is closer than I think a lot of people want to admit.

 

[CoolBrees]

While I agree with you it is close and will certainly not be a walk, the polls that moved all three of the states you named were Trafalger and Insider Advantage. IA only republishing the results from the Center for American Greatness- Trumps polling service.

While I believe both service’s Trump polling numbers, around 46-48%, they heavily under sample Dems. And looking at the averages they undersample by about 1-2%, which is enormous. So if you simply and rudimentarily add 1-2 points to Biden’s average numbers, they fall almost exactly in line with the rest of the polling services.

All this said Trafalger was right the last election- the only one. So they certainly cannot be discounted. I just believe that the undecideds at this point are so minimal they won’t swing the election like last time. That is my hope anyway.

 

[JimEverett]

While I agree with you it is close and will certainly not be a walk, the polls that moved all three of the states you named were Trafalger and Insider Advantage. IA only republishing the results from the Center for American Greatness- Trumps polling service.

While I believe both service’s Trump polling numbers, around 46-48%, they heavily under sample Dems. And looking at the averages they undersample by about 1-2%, which is enormous. So if you simply and rudimentarily add 1-2 points to Biden’s average numbers, they fall almost exactly in line with the rest of the polling services.

All this said Trafalger was right the last election- the only one. So they certainly cannot be discounted. I just believe that the undecideds at this point are so minimal they won’t swing the election like last time. That is my hope anyway.

I just looked at the Michigan poll by Trafalger. Not going to read it in detail, so maybe they account for the following discrpancies, but these seem to be major:

The age participation in their poll was 6.7% 18-24 year olds. CNN Exit polls from 2016 show 11% of voters were that age.
The poll had 23.1% of participants at 65+ CNN exist polls from 2016 show 19%

So Trafalger is skewing older than the 2106 vote.

The poll has 81.6% of its participants as white, with 13.3% black. Exit polling from 2016: 75% white, 15% black.

I did not see a breakdown by party preference.

One way it might skew Democrat: Trafalger poll has 54.1% female participation. Exit polls showed 52% of voters were female in 2016. Not a big difference, certainly not as big as age and race.

 

[B4YOU]

I just looked at the Michigan poll by Trafalger. Not going to read it in detail, so maybe they account for the following discrpancies, but these seem to be major:

The age participation in their poll was 6.7% 18-24 year olds. CNN Exit polls from 2016 show 11% of voters were that age.
The poll had 23.1% of participants at 65+ CNN exist polls from 2016 show 19%

So Trafalger is skewing older than the 2106 vote.

The poll has 81.6% of its participants as white, with 13.3% black. Exit polling from 2016: 75% white, 15% black.

I did not see a breakdown by party preference.

One way it might skew Democrat: Trafalger poll has 54.1% female participation. Exit polls showed 52% of voters were female in 2016. Not a big difference, certainly not as big as age and race.

Trafalger also polls at least 1k people, keeps polls to as few questions as possible, and doesn’t just call people. I wouldn’t blow off their polls and they probably should be more heavily weighted in the RCP average based on past performance.

 

[Nebaghead]

Trafalger also polls at least 1k people, keeps polls to as few questions as possible, and doesn’t just call people. I wouldn’t blow off their polls and they probably should be more heavily weighted in the RCP average based on past performance.

I read up on Trafalgar and they account for shy Trump voters. They did predict a few winners in 2016 that most pollsters missed but they were off by 3-4 points in most cases. The shouldn’t be weighed anymore or less than others in an aggregate. Nate Silver talked about them and weighs them as a C, the same as Rasmussen since they skew regardless of results.

 

[CoolBrees]

Let’s pump the brakes on Trafalger being right- especially a lot. They aren’t. They absolutely have a (heavy) right leaning formula.

They were right during 2016 because their skews aligned with the way the undecideds broke. You could make the argument that was more luck than pure extrapolation. As my grandfather used to say- “The sun shines on every dog’s arse once in a while”

But we should remember why our buddies at fivethirtyeight.com are currently considered the standard bearer. Nate Silver has the best formula going.

 

[DaveXA]

Let’s pump the brakes on Trafalger being right- especially a lot. They aren’t. They absolutely have a (heavy) right leaning formula.

They were right during 2016 because their skews aligned with the way the undecideds broke. You could make the argument that was more luck than pure extrapolation. As my grandfather used to say- “The sun shines on every dog’s arse once in a while”

But we should remember why our buddies at fivethirtyeight.com are currently considered the standard bearer. Nate Silver has the best formula going.

Agreed. As long as they use a consistent and reliable methodology, then I wouldn't discount them. The various polling groups are trying to guess as to who actually is going to win and each has their way of getting the end results.

I tend to give more weight to Silver's than others though. Good debate to be had on what the best methodology is. I think Trump voters are more difficult to nail down because they're voting booth voters. Meaning they won't tell anyone their vote until they cast a ballot. Those are more difficult to account for in polling because they generally don't participate in those polls.

 

[Sandman]

Agreed. As long as they use a consistent and reliable methodology, then I wouldn't discount them. The various polling groups are trying to guess as to who actually is going to win and each has their way of getting the end results.

I tend to give more weight to Silver's than others though. Good debate to be had on what the best methodology is. I think Trump voters are more difficult to nail down because they're voting booth voters. Meaning they won't tell anyone their vote until they cast a ballot. Those are more difficult to account for in polling because they generally don't participate in those polls.

I really don’t think the Trump voters are “shy” anymore. If anything, I think you have “shy” Biden voters. I know I am not putting any signs out in my yard. I have a lot of Halloween decorations and I live in a strong Tea Party area. My decorations would be trashed.

To the contrary, the Trump voters go out of their way to tell you they are voting for Trump. While they may be missed by conventional polling, there is nothing shy about them anymore.

 

[DaveXA]

I really don’t think the Trump voters are “shy” anymore. If anything, I think you have “shy” Biden voters. I know I am not putting any signs out in my yard. I have a lot of Halloween decorations and I live in a strong Tea Party area. My decorations would be trashed.

To the contrary, the Trump voters go out of their way to tell you they are voting for Trump. While they may be missed by conventional polling, there is nothing shy about them anymore.

I really think it depends on where you are. If you're in a heavily Democrat district, they're more shy. People just act differently when they're in the minority than when in the majority in a lot of cases. I do think there are not as many shy Trump supporters, but I still think a disproportionate number loathe polls and don't really participate in them.

That said, this election is going to be about turnout as much as the last one was. Hillary lost becuase the turnout didn't match what Obama turned out. This time, Biden will have to turn out. I think he will. The question is whether Trump can improve his turnout numbers over 2016. That remains to be seen.

 

[Mr. Blue Sky]

I live in a strong Tea Party area. My decorations would be trashed.

You live in a big MAGA area.

The Tea Party was quaint by today’s standards.. it’s members would be considered ‘RINOs’ and would be laughed out of the Republican Party for not being far enough to the Right.

 

[Sandman]

You live in a big MAGA area.

The Tea Party was quaint by today’s standards.. it’s members would be considered ‘RINOs’ and would be laughed out of the Republican Party for not being far enough to the Right.

This is true. My local Tea Party is full-on MAGA, so for me, there is no difference.

 

[Saint by the Bay]

You live in a big MAGA area.

The Tea Party was quaint by today’s standards.. it’s members would be considered ‘RINOs’ and would be laughed out of the Republican Party for not being far enough to the Right.

Justin Amash gives you a big thumbs up on this post.

 

[Sandman]

I really think it depends on where you are. If you're in a heavily Democrat district, they're more shy. People just act differently when they're in the minority than when in the majority in a lot of cases. I do think there are not as many shy Trump supporters, but I still think a disproportionate number loathe polls and don't really participate in them.

That said, this election is going to be about turnout as much as the last one was. Hillary lost becuase the turnout didn't match what Obama turned out. This time, Biden will have to turn out. I think he will. The question is whether Trump can improve his turnout numbers over 2016. That remains to be seen.

Honestly, no one wants to do a poll. When was the last time you answered a phone call from a number you don’t know? I never do. For young people (20s), they never even use the phone part of their “cell phone.”

I agree on the turnout, but I believe that Trump has done nothing to grow his base. They are enthusiastic, but their numbers haven’t grown. The Republicans now depend on depressed turnout numbers to win elections. I just don’t see that happening this year.