Sunday, October 23, 2016

Clinton vs Trump.
23 Oct. 2016,
16 days from the election

Here is the list of states and districts, from most likely to go Trump to most likely to go Clinton.  RED indicates a state moving from one category to another in Trump's direction, while BLUE means the state moved in Clinton's direction.
 
Solid Trump (more than 95% Confidence of Victory [CoV]):
ND NE-3 NE-1 WY WV NE LA AR AL OK ID TN MT KS NE-2 KY MS SD SC
Total: 96

Leaning Trump (between 50% and 95% CoV for Trump): IN MO TX AK GA
Total: 78

Toss-up: OH UT
Total: 24
 
Leaning Clinton (between 50% and 95% for Clinton):
IA ME-2 AZ NC NV FL PA
Total: 88

Solid Clinton (more than 95% CoV for Clinton): NM WI NH MN CO ME CT ME-1 WA NJ OR DE MI VA RI IL NY MD VT HI MA CA DC
Total: 252

The current count 

CLINTON 340
TRUMP 174
TOSS-UP 24

The best news this week gives to Trump is that deal is not sealed in Ohio.

The worst news for Trump is that the deal is also not sealed in Utah and the manner in which it is no longer clear makes the news even worse for the Republican nominee. 

There have been seven polls in the past two weeks in Utah that have checked the numbers for Evan McMullin, a last ditch candidate put up by the Never Trump people, a candidate without even a party affiliation. While McMullin gets less national press than Gary Johnson or Jill Stein, in Utah his poll numbers are rising quickly, the most recent poll from Emerson showing him in the lead. The Confidence of Victory method can deal with races with three or more fairly even competitors, and the current numbers give Trump a 47% chance to win, Clinton a 28% chance and McMullin at 25%. If any candidate gets over 50% in the three-way race, my system will give them the nod.

It's added a little spice to a contest that doesn't look like it has many surprises left.

And now the Probability of Victory numbers.
Trump's chances can still be read off in parts per million, specifically 590 ppm. Nate Silver's idea that he has a 13% chance this late in the game mystifies me. Right now, he has to keep every state where he has a lead - not very hard right now - and then take Utah, Ohio, Arizona, North Carolina, Nevada, Florida and then one bonus state, like Pennsylvania, New Mexico or Wisconsin. There is no way in hell the odds in all his possible paths adds up to even 1%, let alone 13%.

 Here are the 15 states considered this week's by the algorithm.

Trump's non-battleground count of electors: 128
The battleground states ranked by pqn:
(Clinton %, Trump %, electors)
1. OH: 50%, 50%, 18
2. FL: 85%, 15%, 29
3. NC: 70%, 30%, 15
4. TX: 8%, 92%, 38
5. AZ: 60%, 40%, 11
6. GA: 19%, 81%, 16
 7: UT: 53%, 47%, 6 (53% is Clinton + McMullin)
8: PA: 92%, 8%, 20
9. IA: 59%, 41%, 6
10. NV: 77%, 23%, 6
11. MO: 9%, 91%, 10 
12. IN: 6%, 94%, 11
13. AK: 9%, 91%, 3
14. WI: 97%, 3%, 10
15. ME-2: 59%, 41%, 1


Current probability of victory if the election were held today:
Clinton 99.94%
Trump    0.06%
 
 Just 16 days left until this mess is officially over. I have no idea how the House is going and as you can see from yesterday's post, there is still some drama in the Senate. But when it comes to the race that is taking up the vast majority of the discussion, the drama is nearly non-existent.
Posted by at
Labels: , , , , , ,

8 comments:

  1. MarcOctober 23, 2016 at 2:37 PM

    Professor I was wondering what you think of this notion of a big "silent" Trump vote the polls aren't reflecting?

    Myself I think it's possible there's some (key word) of that just as much as there's a hidden Clinton vote. I'm also tired of the Brexit comparison as it's not the same thing.

    ReplyDelete
    Replies
    1. Prof. HubbardOctober 23, 2016 at 3:07 PM

      Polls in the general election are fairly reliable. The overly long American election cycle tends to calcify people's view. I've found that polls taken in races that are over in a few weeks - like American primary races or Brexit - aren't as reliable in general.

      This is the fourth election in which I've gathered these numbers. I haven't seen that much "silent voter" effect, and not enough to get Trump into the lead in eight or more states where he doesn't currently have the advantage.

      Thanks for stopping by.

      Delete
    2. DerFarmOctober 24, 2016 at 8:36 AM

      Does your experience include the fatigue factor? I'm getting vibes of "why bother" from neighbors who were talking of Trump. That could easily be from only MA, but if it is also in deep blue VA and PA it could make a difference in the congressional makeup

      Delete
    3. Prof. HubbardOctober 25, 2016 at 4:16 PM

      I can't tell about GOTV, but Trump's team looks as amateurish at this as it does at near every aspect.

      It's my experience that Dems show up during general elections and get insanely lazy during the midterms. Both sides call the other side "low information voters", a slightly euphemism for "stupid", but I'm willing to say my side shows low information during midterms. I don't understand why they can't get the fire lit if there isn't a president to vote for.

      Thanks for stopping by.

      Delete
  2. EricaOctober 23, 2016 at 9:44 PM

    Are the %s for WI inverted?

    ReplyDelete
    Replies
    1. Prof. HubbardOctober 24, 2016 at 3:56 AM

      Yes, they were! Thank you for pointing it out. I'm glad to have people reading carefully enough to catch a mistake like that, which is now rectified.

      Delete
  3. GOctober 26, 2016 at 1:04 AM

    Hi Professor, thank you for your blog. Nate Silver posted today explaining one of their reasons for giving Trump a 13% chance, I wonder if you could comment:

    "About 15 percent of the electorate isn’t yet committed to Clinton or Trump, as compared to just 5 percent who weren’t committed to President Obama or Mitt Romney at this point in 2012. That’s one of the reasons why our models still give Trump an outside chance at victory. In theory, with Clinton at “only” 46 percent of the vote, he could beat her by winning almost all of the undecided and third-party voters. (In practice, there’s no particular indication that these voters have Trump as their second choice.)".

    ReplyDelete
    Replies
    1. Prof. HubbardOctober 26, 2016 at 3:33 AM

      Hi, G. I will readily admit Silver's model is more complex than mine, but I agree with a statement made at the PollyVote website.

      "Keep it simple: There is no evidence that complexity improves accuracy."

      Part of my simple system is the rejection of the idea that the people who make up their minds late in the game will have a distribution wildly at odds with the way people have made their choices earlier. This could be one of those elections when the minor candidates get over 5% of the vote, and there is even an outside chance one outsider, Evan McMullin, might pick up some electors in Utah, though currently I think Trump should be favored there.

      Thanks for stopping by.


      Delete

Subscribe to: Post Comments (Atom)