Just a bit more about those polls
I’ve already written about how the polls, particularly the national polls, weren’t all that bad this year. But I’m going to add just a bit more.
As of this writing, Clinton is ahead in the popular vote by about 2 million. This amounts to a 1.5% margin, and it could get even larger. It’s not relevant in terms of the election, because the Electoral College still determines the winner. But it is relevant in terms of the national polls, which only deal with the popular vote, and the average of which (according to the person I consider the best poll analyst, Nate Silver, writing two days before the election) predicted a 1.9 margin for Clinton.
Not so far off at all. In fact, quite close.
What this illustrates is several things:
(1) Polls change many times during the campaign year. You can see how much they changed during the 2016 campaign year in a chart here, with the lead changing several times over the twelve months before the election. Clinton was more consistently ahead, though, and was ahead right before the election—and she won the popular vote.
(2) Although national polls seem quite accurate in predicting the winner of the popular vote (especially the closer it gets to Election Day), they are not especially useful in a race that’s at all close, because the Electoral College is the determining factor.
Another important element in this election (and in the difficulty predicting the outcome) was late-deciding voters. Let’s have some figures for how those “undecideds” broke:
Some of the polls were wrong to a degree, yes, but there was also something at work in the final days of the election: People who decided late broke strongly for Donald Trump in the states that mattered, according to exit polls. And without this apparent late surge, Hillary Clinton would be our president-elect ”” not Trump.
In fact, if you look at the four closest states where Clinton lost ”” or, in the case of Michigan, where she’s expected to lose ”” exit polls show late-deciding voters in each of them went strongly for Trump in the final days. In Florida and Pennsylvania, late-deciders favored Trump by 17 points. In Michigan, they went for Trump by 11 points. In Wisconsin, they broke for Trump by a whopping 29 points, 59-30…
And these weren’t small groups of voters. The number of undecided and third-party-supporting voters who were still free agents in the final week was as many as 1 in 8 voters nationally — an uncharacteristically high number for the eve of an election. (As Nate Silver noted, it was just 3 percent in 2012.)
In Florida, 11 percent said they decided in the final week. In Pennsylvania, it was 15 percent. And in Michigan and Wisconsin ”” states where Trump made a late push ”” fully 20 percent of voters said they arrived at their choice in the last seven days.
That’s a lot of people breaking for Trump, and in battleground states it helped turned the tide.
I find these things interesting, and what’s more the subject of polling is bound to come up in the next election (next election? Ugh!) As always, polling is flawed. But not necessarily as flawed as people think.
But what I’m really looking forward to at this point is putting this tedious, stressful election (and its polls) behind me, and hoping that the future is much brighter.
Wouldn’t that be nice? Wouldn’t that be great?
The US election system is not designed to produce an accurate popular nation-wide vote. It is up to 10’s of thousands of precinct captains whether to count absentee and provisional votes or not. Many don’t if the electoral college result is definitive. Many overseas military votes come in late but are still valid.
And in spite of CA’s electors being definitive, you can bet that every single possible CA absentee and provisional is being counted right now in Dem leaning precincts. And if additional provisional votes were being fabricated today in CA, would anyone investigate it?
Popular vote numbers are nonsense.
Dems are always waging the permanent campaign. Republicans are too stupid to understand.
I looked up Sean Trende’s post election analysis of the RCP battleground state-by-state polling averages; the only polls that really matter. He says that these were just as good or even better than the last pres. election. But if you take the sum of the 3 worst state polling average errors, the 2016 sum is 42% worse than the 2012 sum. Not huge (oops not yuuuge), but substantial.
Pollsters usually like to focus on likely voters, or at least give them extra weighting, thinking that this will improve accuracy. But this only works well when the election at hand unfolds in the mold of those that went before. (How do you know that the person on the phone is or is not a reliable voter? Because they were or weren’t before.) But break the mold, and likely becomes unlikely and vice versa.
TommyJay, exactly right.
There are no controls over who may vote in California. Perhaps hundreds of thousands voters who are not citizens voted. Perhaps tens of thousands of democrats voted multiple times.
We know that California politics is corrupt. Anyone assuming the numbers from California are anything but fabricated is engaged in wishful thinking.
I’m sure that Hillary won enough legitimate votes in California to win the state. That isn’t the issue.
The false premise that somehow she won the legitimate popular vote nationally is used by those WISHING to believe that over sampling democrats creates an accurate polling result.
Precision is not Accuracy.
When all the ballots, legal or not, are finally counted it will be interesting to compare each of the states return by actual turnout percentage. My guess would be that fewer democrats voted than the polls used in their estimates.
Did any polls measure how many women voted for HRC just because she was female, and for no other reason? (Uninformed voters……!)
Yep. There must be a pony in there somewhere! Have a Happy Thanksgiving, all.
I love the schadenfreude after the hyperventilating on the Left (and the leftish Right) over Trump’s riff on waiting to see how the election turned out before deciding he would accept it.
Good times.
http://libertyunyielding.com/2016/11/17/los-angeles-county-decide-presidential-elections-us/
(the actual numbers are out of date, but the thesis point stands)
“In recent decades, the Electoral College has served to solidify the political credibility of presidents who were elected with less than 50% of the popular vote, including John F. Kennedy and Bill Clinton (in both of his elections). It is by no means an institution structurally biased against Democrats.
But Democrats are now pushing variously to abolish the Electoral College, and even to circumvent its purpose right now, for this election, by pressuring the electors to contravene their states’ popular votes, so as to prevent Trump from getting the more-than-sufficient 290 electoral votes he is already assured by the current vote count.
In that context, it is very interesting to look at the 2016 vote through the prism of how it is distributed.
…
Quite a few commentators have pointed out, as the Wall Street Journal did on Tuesday, that California alone accounts for Hillary’s entire lead in the popular vote.
And that’s true. But it doesn’t go far enough.
Los Angeles County alone accounts for Hillary’s lead in the popular vote.
…
If we removed the LA County vote from both candidates in the national total, and left the rest of the California vote in place, Trump would lead Hillary right now nationally, with 60,001,544 to her 59,723,194.
That’s about as regionally specialized as narrow majorities get. “
I was just listening to the radio, and the person mentioned that the popular vote lead is kind of nonsense BECAUSE:
1. Absentee ballots generally break 2-to-1 for Republicans nationwide (i.e., “in most states” …even Dem leaning states …because: didn’t say, but I’m guessing “old folk” lol).
2. Absentee ballots aren’t generally counted UNLESS the Electoral vote appears unusually close (as TommyJay mentioned above).
3. At this point, CA alone has 4½M uncounted absentee ballots (per rule 1, which would likely add 1½M to Hil’s 1½M “lead”, for 3M total nationally …but would give Trump 3M …so Trump “tied/too close to call” in the popular …at which point multiply Rule 1 by all the state’s that don’t tally absentee ballots, yada-yada …and it don’t change a damn thing, because his EC vote is now – Michigan reporting today – 306.)
So TommyJames FTW.
I didn’t know this. I’m not googling it to confirm. Neo’s better at this sort of research than me, anyways lol.
DISCLAIMER: I suck at math. So, YMMV even if the above is true.
Oh yeah: and a Happy Thanksgiving everyone.
If you subtracted out dead voters and those who voted multiple times, I suspect that Hilary’s lead in the “popular vote” would vanish.
3 million fake votes by Democrats. Just enough. Not too much, since even the idiots can figure it out if you win by too much.
It also costs ACORN less to Get out the Vote, and more of that money can be money laundered through Planned Profit or other Leftist grassroots insurgencies and Gramsci marchers.
If they used more of their strategic reserve, they could get 15 million, via cracks/hacks/bribes, but their grassroot organizations are limited to certain states and cities. Their buses can’t reach all of the land. The strategy Trum won by can be described as “if it isn’t close, they can’t win via cheating”.
So many Democrats overturned their own votes for Trum, that it would have taken a strategic adjustment on the part of OFA and other orgs that focused on cities. Because obviously, if a state’s electoral vote goes entirely for whomever wins the most precincts or popular vote in a state, then by controlling a few key cities in that state, one can obtain all the electoral votes. But that was only for states that didn’t reliably vote Democrat. This time, they needed to cover the entire board in Go, not just a few key places. Since HRC was distracted by smashing Sanders, and many of the get out the vote volunteers were Sander loyalists, that most likely threw a wench into the Left’s efforts. Nothing like the effort they put for Hussein, their Messiah King.
I have my doubts about Hillary’s lead in the popular vote opening up with the counting of absentee ballots. Many of the absentee ballots are military, and from what I’ve read, went for Trump about 8:1.
Of course, in Dem states, they count the votes, don’t they?
Many of the absentee ballots are military, and from what I’ve read, went for Trump about 8:1.
Of course, in Dem states, they count the votes, don’t they?
They count to make sure to discard the military absentee votes or convert them to Democrat. This time they might not bother, since it isn’t worth it.
My own favorite factoid
If Hillary had convinced the right 200K people to vote her way (22K in MI, 56K in WI, and 116K in PA) we’d be talking about her winning those states by Trump’s margin.
Of course the Dems would still be crowing about their meaningless margins in CA and NY.