Polls for several months have consistently demonstrated that Hillary Clinton would have difficulty in a general election, while Bernie Sanders would do much better. Head to head polling frequently showed Clinton losing to all candidates except Donald Trump, and now that Trump has probably clinched the GOP nomination, even that race has become a virtual tie. I have seen several Clinton supporters brag about polls showing that Clinton has a chance to flip Georgia. What they ignore is that this is not due to any great degree of support for Clinton. It is due to demographic changes in Georgia, and Bernie Sanders has an even better chance of carrying the state. From AJC polling (emphasis mine):
Democrat Hillary Clinton and Republican Donald Trump are locked in a statistical tie in Georgia, according to an Atlanta Journal-Constitution poll that laid bare the deep divide over the presidential race.
Trump’s 4-point lead over Clinton — he’s at 45 percent — is within the poll’s margin of error, meaning neither can confidently claim a state that’s voted for the GOP nominee since 1996. Sprinkled throughout are reminders of the challenges both face in capturing Georgia: dim voter enthusiasm, high unfavorability ratings and deep skepticism from voters.
Perhaps the most telling sign of all: Vermont U.S. Sen. Bernie Sanders polled higher than both in one-on-one matchups, winning a potential contest with Trump 47 percent to 42 percent. Although Clinton seems poised to win her party’s nomination, the AJC poll is among a string of surveys bolstering Sanders’ case that he poses the bigger threat to Trump.
Sanders does better than Clinton both nationally and in the battleground states, along with showing a much better chance of expanding the playing field.It would place the Republican nominee at a disadvantage if Trump had to fight just to hold onto Georgia. In contrast, Clinton is vulnerable in several states which have voted Democratic in recent years.
As David Niose, legal director of the American Humanist Association, wrote at Salon, This is one weak nominee: Hillary Clinton’s problem isn’t Bernie Sanders. It’s Hillary Clinton.
No matter what you think about Hillary Clinton as the presidential primaries wind down, there is one undeniable fact that lingers in the background. Despite having had enormous advantages from the start of the campaign—no serious competition from within the party, solid support from national party leaders, a massive war chest and a nationwide grassroots network built over the course of decades in national politics—Clinton has struggled to put away a 74-year-old Jewish socialist who has had almost no establishment support.
Clinton lost to an inexperienced Barack Obama eight years ago, and is barely beating Sanders despite a system heavily rigged in her favor. She has been fortunate that Sanders has not used her major scandals against her, but we can be sure Trump will.
Seeing Sanders do better than Clinton in a general election match-up, despite her landslide victory in the Democratic primary, also shows how Democratic primary results are not representative of all potential Democratic voters. This might also indicate a decrease in support for Clinton and increase in support for Sanders since the date of the primary. While I would not normally like to see superdelegates decide the nominee, this is the time for them to intervene to correct the problems caused initially by the system giving such an advantage to Clinton. If the superdelegates system was initiated to protect against nominating an unelectable candidate, this very well could be such a situation. Plus we have not encountered a situation in which the leading candidate was as unethically unfit to be president as Hillary Clinton, along with currently being under FBI investigation (and not a security inquiry as Clinton falsely states).
Awesome article my friend. Shared on facebook at Bernie Sanders for Prez.
You are essentially comparing a statistical poll by the Atlanta Journal-Constitution, in which Sanders polled at 47% against Trump and Clinton polled at 41%, to the actual results of the Democratic primary, in which Clinton received 71% of the vote and Sanders 28%.
The AJC probably had about 1000 people in their sample; they don't say, but the MoE they mention indicates approximately that. The Democratic primary had a sample size of 757,000. It's pretty obvious which measurement, poll or vote, is apt to be more accurate.
No, a poll of preferences in the general election is what is accurate. A primary vote does not have meaningful predictive value towards the general election while this poll does. We are talking about two entirely different samples. Clinton does well among hard core Democratic voters (those who turn out in primaries). She does much more poorly among all voters compared to Sanders due to opposition to her from a large percentage of Democratic-leaning independents. Plus Sanders does better in getting cross over votes from Republicans.
Seeing Sanders perform significantly better than Clinton in this head to head general election poll is consistent with all the other polls showing the same.
If the goal is to stop Trump, Sanders is by far the best choice.
We're talking about the relative strengths of Sanders and Clinton. In what way is an overwhelming victory by Clinton over Sanders in the primaries not indicative of their relative strength among Democratic voters? And remember, multiple polls by veteran pollsters at this point in 2008 showed McCain winning Georgia by 14 points; he won by five. Why is it you think the AJC poll is accurate again?
Entirely different? What evidence do you have that none of those polled by the AJC were among the 757,000 who voted in the Democratic primary?
Exactly none of that information comes from the AJC poll. It is information from national polls that says nothing whatsoever about these 600 to 1000 Georgians. You are doing what the Republicans did leading up to the 2012 vote; they called it "unskewing" the polls. Ron, I realize you know very little about probability and statistics, but you're drawing conclusions about differences between numbers within the margin of error. You really can't do that when the numbers are within 1-sigma — the "margin of error" in polls like this — and you've a low probability of being correct is you do it within 2-sigma — twice the "margin of error." (Note that the margin of error of the Democratic primary is zero.)
I suggest you read the Wikipedia article on margin of error. It's somewhat simplistic, but it covers the issue.
Of course all this is moot; Hillary Clinton will be the Democratic candidate for President.
We are talking about the relative strengths in the general election, among all voters, not relative strength among primary voters. The ACJ poll measures a general election match-up. The primary does not. This certainly can change quite a bit by election day, and probably will, but at this point the polls consistently show that Sanders does significantly better than Clinton against Trump. This comes from multiple polls, not just this one.
Considering how badly you botched your interpretation of the data, you certainly are in a poor position to say that someone else knows very little about probability and statistics.
The primary measured at least 20% of the people who will vote in the general. That showed overwhelming support for Clinton.
Polls of Georgia voters? Citations needed.
Given that the Applied Mathematics Department of Brown University was willing to have me teach a graduate course in probability and statistics, I tend to think I know a fair amount about the subject. It's purely your opinion that I botched it; guess how much credence I give to your opinion about such things.
I don’t care if you taught a graduate course in statistics. When you try to take a primary vote from over a month ago (over a period in which Clinton’s support has declined nationally) and compare it to a current general election poll, you are talking nonsense. You botched it, period.
It is polls from other states and national polls which demonstrate that Sanders does better with independents than Clinton. This is the most likely explanation for Sanders doing better than Clinton in the head to head poll in Georgia.
Let me get this straight Bob. You think that the primary vote has anything to do with the general election vote? You know what Mark Twain said about statistics? If not, google it, while you are looking up how much better Bernie does with independents than Clinton does.
Alex, didn’t Twain say something about liars, damn liars, and Clinton supporters? 🙂
Bob does have a long history of denying the facts when it comes to Clinton.
As for statistics, I can still recall an old statistics prof at Michigan starting out the class by explaining that statistics is the science which proves that the average human has one testicle and one breast.
Bob is not only mangling statistics (or misusing statistics to lie). He is woefully ignorant of basic political science if he things that you can say anything about the general election based upon primary results. Did this wanker really teach at Brown?
Bob is hardly the first person to make bogus extrapolations based upon a primary to support their candidate in the general election. He only claims to have taught statistics, not poli sci.
I don’t really know whether he taught at Brown. I only know him from his on line antics, and have caught him in lies in the past so anything he says has to be taken with a big grain of salt. He will say virtually anything to try to cover when he has been shown to be wrong on an argument.
I think I remember Wanker Bob from Brown. The way I recall things, he flunked out sophomore year. I'm pretty certain he flunked both statistics and political science. #bernieorbust
No comment.