15 October 2016

Battleground/National Report, 14 October, 2016 - T-minus 25 days (Clinton 344 / Trump 194)

Friday, October 14th, 2016, was exactly 25 days before the General Election on November 8, 2016. Less than one month to go...

Complete polling results for all states for 2016 to-date are here as an EXCEL table for 10/14/2016 in GOOGLE DOCS. All polls are hyperlinked.


So, my methodology for doing averages in this wild-and-wooly-sometimes-3-way election was explained in yesterday's starting Battleground Report, from 11 October, 2016.

The Battleground table is pretty darned detailed. The LEGEND for said table is HERE. In fact, you need to read it in order to develop an "eye" for the Battleground tab.

Only one state received a poll yesterday: Florida. So, the battleground table and the electoral map has not changed one bit:




In Florida, a poll from PPP (D) was released showing Clinton +5 in a two-way and  +4 in a four-way. For Florida, traditionally a very, very tight state, a +5 is a big margin. And there is reason to believe that this margin is here to stay, based on two key poll internals. Now, I won't always have time to fuss with internals, but this time, I think it's worth the effort. And PPP (D) was one of the only firms in 2012 to correctly call FL for President Obama, by +1, and indeed, he wonn Florida by +1.

So, internal number 1:






It is really rare when not even one single undecided voter in a poll has a favorable opinion of one candidate. That is a sure-fire sign that said candidate is in real trouble.

And this internal:



According to PPP (D), Clinton is leading among women in Florida by +14 in a four-way and by +16 in a two-way, making for an aggregate of +15. On the other hand, Trump is winning in the mens vote by +7 in both scenarios.

Let's take a look at the 2012 exit poll from Florida:






In 2012, Mitt Romney won in the mens vote by +6, similar to Trump's current +7. But Obama only won in the womens vote in this state by +7, half of Clinton's current margin. This is going to be a hard statistic for Trump to turn around, since the Florida electorate in terms of gender was Female +10 in 2012, tendency: rising.

But even with this poll, Clinton's Florida statistic only ticked up very slightly (+0.03) as a poll at the bottom of the time-frame used for 10/13 dropped out of the statistic for 10/14, showing a very stable Florida aggregate of Clinton +3.21 in the Sunshine State.

Obviously, the battlegrounds remain at 7 just like they were on 10/13 and the national Electoral vote outlook is the same as well:







NON-BATTLEGROUND polling, or maybe???  Yesterday I spent a good deal of time mentioning UTAH and the fact that the statistics out there are just weird for the state that is provably the most Conservative state of all in presidential politics. With the sudden rise of 4th party candidate Evan McMullin in Utah polling, I mentioned that I may change the Utah tab in the excel table to reflect that, and I did exactly that:






So, those are all 22 Utah polls to-date and the three-way has been expanded to include McMullin (also from Utah, also a Mormon). As you can see, both the Ipsos/Reuters and the UPI/Cvoter polls, which have been doing 50-state polling about once a week, only do 2-way polling and the survey groups are very, very small. On the other hand, the Monmouth and the Y2 groups are not much larger or the same size, but they are not internet-only polls. So, take your pick. But you can see the HUGE disparity between the two-way margin and the four-way margin average in Utah. Were we to take the four-way average, then at Trump +3, this would then be a true battleground state. But it's too early to tell if the McMullin "conundrum", as I choose to call it, is real or not. We need to wait a week or more yet to see. At the risk of sounding political in a numbers series where I really do try my hardest to be apolitical, my gut tells me that if former presidential nominee Mitt Romney, seeing that Trump's path to 270 is pretty much foreclosed, anyway, comes out in direct support of fellow Utahn Evan McMullin, then we really could see an extremely rare independent candidate win here. It happened with Robert LaFollette in Wisconsin in 1924 in spite of Calvin Coolidge winning the general election with over +25% in margin, the second largest margin in our Union's history. So, I suppose anything is possible. In this way, Utahns could vote their conscience without thinking that the 6 EV missing from the GOP column would make the difference between victory or defeat for Donald Trump. On the other hand, there may just well be enough Utahns, including Mormons, who will place platform over candidate and vote "R" anyway, which is of course their good right. But just to show what a HUGE disparity we are seeing in the two-way vs. four-way polling in Utah, which we are seeing nowhere else in the country, here is the polling tab for neighboring Wyoming, usually the 2nd most Conservative state in the Union:




Now, to-date there are no matchups from Wyoming (which is usually not even polled at all in presidential elections) involving McMullin and there is a sizeable contingent of Mormons also living Wyoming, but you can see here with the naked eye that the two-way and three-way margins in Wyoming are pretty much the same and hover at +35 for Trump.

But take a look at the UTAH tab again. As early as September, 2015, polling was showing Trump in trouble in Utah, long before McMullin announced and long before the tape and the allegations of sexual abuse of women on Trump's part. Something has been up in Utah for more than one year now - this is not a new phenomenon. And I suspect that for the first time ever, the TV stations are not going to be calling Utah for anyone immediately on election night. That alone may become a historic first in the TV-era.

On the national level:






A poll fell out of the bottom of the statistic and the newest Dornsife, Ipsos and others came into the statistic, notably PEW. This bumps Hillary Clinton's national average up just a tick, to +6.75, well out of the MoE and well outside of the Battleground zone.


The tracking list:

12 October 2016Clinton +5.52, -0.21
13 October 2016, Clinton +6.61, +1.08*
14 October 2016, Clinton +6.75, +0.14

*Due to rounding, what looks like it should be +1.09 is truly +1.08.

And some end-statistics: I first started collecting 2016 presidential polls at the beginning of October 2013 and since then I have logged 4,004 matchups from 1,908 individual polls. You can find the breakdown to this stuff at the EXCEL table, in the tab that says "poll totals". Now, that's a pretty big jump over 10/13 whereas I am reporting only one state poll. However, I forgot to update a couple of tabs from very safe states, so a little housekeeping went into this statistic, nothing more.

For past reference, here is the BATTLEGROUND REPORT from 25 days before the GE 2012, from 10/12/2012. And here is the BATTLEGROUND REPORT from 2008: 10/10/2008.


-Statistikhengst

No comments:

Post a Comment

Constructive comments and critique are always welcome. Please keep it polite and respectful.