14 October 2016

Battleground/National Report, 13 October, 2016 - T-minus 26 days (Clinton 344 / Trump 194)

Thursday, October 13th, 2016, was exactly 26 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 10/13/2016 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.

Polls were released for the following 13 states on 13 October, 2016: AK, FL, GA, MO, NH, NV, NC, OH, OR, PA, TX, UT and VA.

Based on polls released on 13 October, 2016, two states exited the battlegrounds: GA and NH, leaving us with 7 statistical battlegrounds. But I also included two new states onto the battleground chart overall, although they are not in the "battleground" (+5% or less in margin) category:






Two new polls were released from OHIO, with conflicting results: the NBC poll showing Trump +1 and the Emerson poll showing Clinton +2. The funny thing about that is that most of the time, those polls have been showing the opposite of this. That being said, the Ohio margin still rose, a lot of it bouyed by the very large margins from the BW poll from the day before. Clinton is now ABOVE Obama's end polling average in Ohio from 2012 but the state is surely very competitive.

North Carolina is looking even more competitive, and yet, the numbers are also very, very stabile. The two newest polls, also from NBC and Emerson, both show Clinton +4 in a three way. Only, in the two-way statistic there are only two polls and her average there is just +0.75. Average it all together and her aggregate raises only so slightly to +1.91 because one poll fell out of the bottom of the statistic. So, this shows how lean but yet resilient her lead is in this state.

An FAU poll was released for Florida, showing Clinton +6 over Trump. Note: this poll was completed before the sex-tape was released last Friday. This moved Clinton's aggregate up to +3.18, which is larger than Obama's 2008 margin over John McCain and decidedly better than the end-statistics from 2012, both in the end-polling and the actual result.

Conversely, a poll from a Democratic pollster camed in showing a 43/43 tie in Nevada, thus tightening the statistic there, down to +2.18 for Clinton.

Not surprisingly, all four of these battleground states are right next to each other in the +2 to +4 range of the battleground table.

However, New Hampshire is now out of the battlegrounds, posting a healthy +6.09 aggregate for Clinton. Three polls were released on 13 October, all showing Clinton leading Trump with margins between +3 and +9. The GQR poll (Green-Quinlan-Ross), from a Democratic firm, is perhaps the most telling. BTW, GQR was the only firm to absolutely nail Pres. Obama's national average in 2012: on election eve, GQR predicted Obama +3.9 and he won with +3.9 - an absoute bullseye.

And in Georgia, a Landmark (R) poll was released showing Trump +5.5 over Clinton, moving Georgia out of the battlegrounds, at least for now.

In Pennsylvania, a Bloomberg/Selzer poll was released showing Clinton +9 over Trump in both the 2-way and the 4-way matchups. This is the 5th poll in 9 days to show a landslide margin for Clinton in the Keystone State and furthermore, Selzer is a high-end polling company that does serious work and is considered a gold-standard for many states. With one poll dropping out of the bottom of the statistic and Bloomberg/Selzer now in the statistic, Clinton's aggregate jumps to +9.9 over Trump - way out of the real battlegrounds, als Obama 2008.

In Missouri, Remington Research, a strong Republican pollster, released a poll showing Trump up only +5 on Clinton, thus reducing the aggregate from Trump +9 to Trump +7 overall. The state is still not really competitive, nor is Clinton making a play for it - yet - but the fact that Trump's margins in many places are shrinking while Clinton's margins continue to increase is yet another piece of statistical evidence that the Clinton campaign is on the winning side of this election. Missouri, next to Ohio and Nevada, used to be one of the three great bellwether states of the nation, but since 2004 it has drifted more and more to the Right with each cycle.

In Virginia, Emerson College released a poll showing Clinton up only +3 on Trump. This, coupled with one poll falling out of the statistic from 10/12, shrunk her aggregate from +8.58 to +7.09, a pretty big jump, but considering that this was once a Bush +8 state, alone a steady +7 for a Democrat in the Old Dominion is a big story waiting to be told.  Plus, Trump pulled his resources out of this state yesterday, a sure sign that he has ceded it to the Clinton campaign.

Now, on to some non-battleground states that may really become battlegrounds.

I mentioned UTAH in the battleground report from 10/12. On 10/13, yet another poll was released, again showing a very competitive race and Evan McMullin, the Independent (4th party) candidate, at 20%. Well, this makes for quite a conundrum and I want to show you why:



First, not sensing that there could really be a credible fourth-party bid anywhere, I have not yet programmed my excel table for McMullin, but if this trend continues, I may have to.

Again, the Y2 poll from 10/12 showed Trump 26 / Clinton 26 / McMullin 22 / Johnson 14 - a mathematical absolute tie between Trump and Clinton and a statistical tie among the top three.

The Monmouth poll, released on 10/13, shows Trump 34 / Clinton 28 / McMullin 20 / Johnson 9 , margin = Trump +6 in a state that is usually R+40!

If you average the two values, in Utah, in the four man calculations, it's Trump +3 - which is obviously a battleground statistic.

Only, Ipsos/Reuters and UPI, in their 50 state polling, only do 2-way numbers and those numbers are much higher for Trump, although a +15 is, in light of the electoral history of this state, relatively unimpressive but yet enough to guarantee a landslide win. For this reason, I have placed Utah at the top of the battleground table, with a current average of Trump +13.88. I never, ever in my life thought I would see the words "Utah" and "battleground" in the same sentence, and yet, here in 2016, the real possibility for a 3-way dogfight exists in the Beehive state. So, I think that the prudent thing to do is to watch how things unfold. if history is our guide, in this case, I think we can learn something from 1928 (Massachusetts, Rhode Island) and 1924 (Wisconsin) when it comes to a sudden paradigm-shift for a state.

In Alaska, a state that many people tend to forget, an Alaska Survey Research poll was released showing Trump at only 36% and only +5.5 on Clinton, similar to the polling we just saw out of Utah. However, there may be a long-term trend at work here. In 2000, Bush easily won Alaska by +30.95 and in 2004 by +25.55. In 2008, even with Alaskan Sarah Palin on the GOP-ticket, McCain won the state by less: +21.54. And in 2012, while Romney gained most everywhere else, he won Alaska by just +13.99. Now, +13.99 is still a big landslide margin, no doubt about it, but that is an almost 17 point drop in margin over 4 cycles. So, the somewhat narrower margins in our northernmost state may now be par for the course and yet, no one expects a Republican to only be at +5.5 in this state.

Also, in Texas, a SUSA poll was released on 10/13 showing Trump only +4 points ahead of Clinton in the Lonestar State.  That is pretty much the shock-poll of the day.  For now, Texas goes to the top of the battleground table next to Utah, but I don't suspect that it will stay there. This could maybe be an outlier, only, SUSA has a pretty good reputation overall.



The electoral map from yesterday to today has not changed, it is still at:






On the national level, here the numbers:



Every national poll except Rasmussen shows Clinton leading nationally and her aggregate has rising more than one point since 10/12, up to +6.61.


The tracking list:

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

*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 3,970 matchups from 1,887 individual polls. You can find the breakdown to this stuff at the EXCEL table, in the tab that says "poll totals".

For past reference, here is the BATTLEGROUND REPORT from 26 days before the GE 2012, from 10/11/2012. And here is the BATTLEGROUND REPORT from 2008: 10/09/2008.

-Statistikhengst

No comments:

Post a Comment

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