28 October 2016

Battleground/National Report, 27 October, 2016 - T-minus 12 days (Clinton 323 / Trump 180 / Tossup 35)

Thursday, October 27th, 2016, was exactly 12 days before the General Election on November 8, 2016. T-minus less than two weeks.... final sprint....

Complete polling results for all states for 2016 to-date are here as an EXCEL table for 10/27/2016 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 the first 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.

8 national polls came in on 10/27 plus polls for the following states: CA, FL, GA, IA, MA, MI, MO, NC, PA, TX, VA. That's eleven states, all of which received one poll except TX, which received 2. So, 20 polls were logged on 10/27.

Among the polls were battleground polls put out by Quinnipiac (GA, IA, NC, VA) and those polls show exactly what the aggregate has been showing quite consistently, namely, that IA really is all tied-up, GA is a true battleground, Clinton has a lean but stubborn edge in NC and is tromping Trump in VA. VA is definitely NOT a battleground in 2016.

In fact, all of the state polls simply confirmed what the stats and trends have been saying about each of those states and the battleground statistics have barely budged. Especially in the green battleground zone, there was no change in the partisan ranking, based on aggregate:

With this in mind, the electoral landscape has not changed at all in the last 4 days:

Some of the margins shifted every so slightly down for Clinton, due mostly to some polls falling out of the 7-8 day time-frame. 

Interesting is that Florida, with the addition of one poll and the loss of one poll, barely budges. Clinton remains at a +2.42, roughly President Obama's 2008 margin of victory in the Sunshine State.


Today I would like to concentrate on 4 very safe states for the sake of statistical comparison to 2012: CA vs TX, MO vs MA.

Here is the current polling in California:

Clinton's current aggregate in California: +26.65. That is an enormous aggregate.  I want to make a point here. Let's go back to 2012:

Obama's 2012 end-aggregate was: +15.80 (almost 11 points under Clinton's current aggregate) and the exit poll showed him only at +17.5. When all was said and done, he carried the state by +23.12, so the aggregate was off to the RIGHT by more than 7 points. His win in CA translated to just over +3 million raw vote margin, out of just over 13 million votes cast

Please notice that the final PPIC (the gold standard for California) showed Obama at only +12 and the current PPIC (which may also be the final one, notice the date from 2012) shows Clinton +26. That is a huge difference and if the mathematical bias of the polls is similar to 2012, then we could very well be looking at a Clinton +30 or even more victory in our Union's largest state.

Furthermore, voter registration in California is up sharply in 2016 and it is estimated that at least 14.5 million will go to the polls. If that is so, and Clintons current +26.65 aggregate holds - and let's assume that the polls are off only .35 to the RIGHT and estimate her at +27 (yes, I am deliberately lowballing this...), then that would result in a 3.9 million raw vote margin and could easily be a +4 million, when all is said and done. I am writing this to illustrate how hard it is going to be for Trump to even get close to Clinton in the NPV.

Moving on to TEXAS:

The current aggregate is: Trump +7.50, with a large disparity between the two-way and three-way aggregate.  Some networks are classifying TX as a battleground. I am not so sure about this, but wait and see... however, let's compare to 2012:

In 2012, Gov. Mitt Romney's end-aggregate was +17.50, we won by +15.78, so the polling was about 2 points to the Right in that state. His landslide victory in the Lonestar state translated to a +1.3 million raw-vote margin out of almost exactly 8 million votes cast.

Voter registration is also up considerably in TX and it is very likely that at least 8.7 million Texans will vote. If the current aggregate holds, then Trump will win Texas, but by only 650,000 votes.

Now, here is where very simple math comes into play:

2012, exact raw vote margins:

California: Obama +3,014,327
Texas: Romney +1,261,719
Difference: Obama +1,752,608 raw votes

So, if we see the two largest states in the Union as relatively cancelling each other out in terms of margins, Obama still got 1.7 million of his 5 million vote national margin in that year. It was 34% of this national raw-vote margin.

Now, let's go back to the last time a Republican won the GE, 2004:

California, Kerry +9.95%, +1,235,659
Texas: Bush 43 +22.86%, +1,694,213
 Difference: Bush +458,554 votes

The difference between Bush and Kerry in these two behemoth states in 2004 contributed greatly to Bush's. 4.5 million raw votes from the difference of these two state was 15% of his national margin, not as much as Obama's but still, a big raw-vote engine.

Now, let's see what happens if the current polling aggregates in CA and TX hold and the projected voter turnout holds:

California: Clinton +27 of 14.5 million = +3,900,000 raw votes
Texas: Trump +7.5 of 8.7 million = +650,000 raw votes
Difference: Clinton +3,350,000 raw votes.

That is a HUGE difference.  
Let's compare, CA to TX, difference:

2012: Obama +1,752,608 raw votes
2004: Bush +458,554 raw votes
2016: ? Clinton possibly +3,350,000 raw votes

Now, that is just an estimate, but I will be curious to see how well it holds on election night. Yes, Clinton is going to win California's 55 EV and yes, regardless what the press and pundits want to jabberwocky right now, although it may be close, Trump is going to win Texas' 38 EV. But the difference in the raw vote margins between these two largest vote generators in the nation is going to be a story for itself.

2 more safe states: MA and NE

In MASSACHUSETTS, an absolute Democratic "lock":

Less analysis, just some quick observation: in October there have been four polling results showing Clinton at +32 or above in the Bay State. That is a margin that harkens back to Bill Clinton's  +33.4% 1996 margin.  Remember, the principle is that "a rising tide lifts all winning boats". And the margin in the Suffolk poll represents a major jump over the last Suffolk results from September.

Let's go back to 2012:

The end-aggregate was: Obama +19.00, he won by +23.14, so the end-polling was off to the RIGHT by 4 points. To be fair, the Suffolk poll also had Obama at +32, while a Republican pollster (Kimball) had him only at +13, but the MassINC poll also had him under his actual margin and right now, MassINC is showing Clinton +26. The point is: if the polls are still off to the Right by four as they were in 2012, then Clinton is very likely at +30 here. You can see from the margins comparing 2012 to 2016 that Clinton's values are overall higher than they were for Obama in 2012, across the board.

In NEBRASKA, an absolute "lock" for the GOP, the polling looks like this right now:

The current aggregate: Trump +13.17. This is decidely less than Mitt Romney's +21.78% win in the state, but tracks very close to John McCain's +14.93% win in 2008, a year where Obama won by +7.26% nationally.

Let's compare to 2012:

Mitt Romney's end-aggregate was: +17.25, he won by +21.78, so the aggregate in this case (one of the very few from 2012) was off to the LEFT by 4.5 points. This is also one of the few moments in 2012 where Republican pollster Gravis had a bullseye, BTW.  So, even if we assume that the polls in NE are still 4.5 points to the left, then that would mean perhaps a Trump +17.8, a margin between Romney's 2012 and McCain's 2008 statistic, and not anywhere close to Bush 43's +28.99% win in 2004. Remember, for a Republican to win nationally, he must run up unbelievably high margins in the middle and smaller "red" states in order to get there, precisely because of states like California, New York and Illinois. The current Trump statistic does not, however, show this happening. BTW, as further proof, the current NE statistic (Trump +13.17) is right next to the current KS statistic (Trump +12.17) and those two states were right next to each other and had almost identical margins in both 2012 and 2008. So, the statistic we see in NE is also the statistic we see in KS - further proof of consistency in polling.

In other words, and this is not just anecdotal: in most safe-BLUE states, we are generally seeing higher polling aggregates for Clinton compared to 2012, and in most safe-RED states, we are generally seeing lower polling aggregates for Trump compared to 2012. You can see it quite clearly illustrated at the EVERYTHING TABLE in the excel data.  It's not just that the battleground polling is showing a clear electoral win for Hillary Clinton, but rather, the aggregate patterns in the so-called "safe" states are showing the same.


That's 8 new polls that came in on 10/27, so between 10/25-10/27, the polling "gene-pool" has been almost completely replaced, and yet, Clinton's aggregate remains very steady, right now, at +6.55.

I will admit, for the first time, I was considering not including a poll: Gravis (R), since it has teamed up with Breitbart, which is so right-wing oriented and where Steve Bannon is now Trump's campaign guru, but for the sake of consistency, I will include it. But the drag from some pollsters whom I am absolutely convinced are deliberately keeping Cinton's margin down leads me to believe that her national margin be considerably larger than the current statistics point to. I will also remind that DEM CORPS (GQR) was the only pollster in 2012 to absolutely nail Obama's winning margin. They had him at +3.9 and he won by +3.9.  So, the +12 value from GQR is not something I would poo-poo at. It is entirely likely that Hillary Clinton is, in reality, easily above +10 right now.

So, after the final results are in, I will be closely examining EVERY pollster, for the end numbers on both side will then be permanent record. 

The tracking list:

12 October 2016, Clinton +5.52, -0.21
13 October 2016, Clinton +6.61, +1.08*
14 October 2016, Clinton +6.75, +0.14
15 October 2016 - no battleground report posted
16 October 2016, Clinton +6,44, -0.31
17 October, 2016, Clinton +7.35, +0.92*
18 October, 2016, Clinton +6.91, -0.45
19 October 2016, Clinton +6.64, -0.27
20 October 2016 - no battleground report posted
21 October 2016, Clinton +6.74, +0.10
23 October 2016, Clinton +6.02, -0.72
24 October 2016, Clinton +6.25, +0.23
25 October 2016, Clinton +6.37, +0.23
26 October 2016, Clinton +6.73, +0.36
27 October 2016, Clinton +6.55, -0.18
*Due to rounding, the difference is off by 0.01

On this corresponding day in 2012the national aggregate shifted from Obama +0.17 on 10/24 back to Romney +0.07% on 10/25, so Clinton is currently 6.62 points ahead of where Obama was four years ago. Yesterday, she was 6.56 ahead, so in-spite of losing a tick on margin, in relation to 2012, she is actually MORE ahead, but only by a tick. Could also be statistical noise, so, pffffft..... 

If you look at the margins, you see a gentle rising and falling - often called the sinus-curve effect. I fully expected to see this.
And some end-statistics: I first started collecting 2016 presidential polls at the beginning of October 2013 and since then I have logged 4,407 matchups from 2,279 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 ELECTORAL LANDSCAPE from 12 days before the GE 2012, from 10/25/2012.  And here is also the corresponding report from 10/23/2008.

One more thing: early voting has begun and there are lots of stats coming out of many states. Soon, I will be collating that data as well.


No comments:

Post a Comment

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