Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

Doctor Jack

(3,072 posts)
Sun Aug 7, 2016, 05:01 PM Aug 2016

Reminder: 92% of Candidates Behind At This Point in August Lost the General Election

I like to bring this up from time to time. In the past 82 years, only 2 candidates that was behind at this point in the polling (papa bush and truman) came back to win. All others ended up losing. And if we go just 2 more weeks and Trump is still behind,he will be in a position that that only 1 candidate ended up winning has ever been in. Every candidate for the past 82 years that has been consistently ahead by August 21st has ended up winning, except in 1948.

And based on our current polling, I would say trump making up a near double digit deficit and actually pulling ahead in a week or two is about as close to 0% as you can get.



Keep in mind, his position right now is already extremely dismal. History says he only has an 8% chance of winning right now. In 2 weeks, it drops to 4%. His position is quickly leaving "virtually no chance of winning" to "complete catastrophe".


Edit: Updated the information based on the 1948 election. I blame the person that made the cart for using nearly identical colors for Truman and Dewey.

17 replies = new reply since forum marked as read
Highlight: NoneDon't highlight anything 5 newestHighlight 5 most recent replies
Reminder: 92% of Candidates Behind At This Point in August Lost the General Election (Original Post) Doctor Jack Aug 2016 OP
Do you have the data for this op? andym Aug 2016 #1
True, I guess I got that one wrong Doctor Jack Aug 2016 #2
Maybe not. See my reply below. The polls that year don't mean much. nt stevenleser Aug 2016 #4
Let's go with the best case scenario for trump Doctor Jack Aug 2016 #5
Great work here... so Gore was really never ahead of Bush!?! WTF?! Jebb really fucked us in Florida uponit7771 Aug 2016 #9
Gore was briefly ahead in September 2000 but was well behind most of the election Doctor Jack Aug 2016 #12
1948 would be an interesting asterisk to that because of the polling issues stevenleser Aug 2016 #3
Bush Pere was winning after both Conventions concluded DemocratSinceBirth Aug 2016 #6
That is the problem with these early conventions this year Doctor Jack Aug 2016 #7
+1, the GOP is already turning towards the down ballots ... Clinton has to stay on the offensive uponit7771 Aug 2016 #8
Her team is extremely experienced Doctor Jack Aug 2016 #10
I agree, we were saying the other night that Clinton would not only keep the trains running on time uponit7771 Aug 2016 #13
I hate these stats even if it is absolutely true Jon Fogerty Aug 2016 #11
The Brexit polls were pretty accurate Doctor Jack Aug 2016 #15
The Brexit Analogy Needs Calrifying louis c Aug 2016 #17
Don't forget, polling has become far more accurate in the past 25 years, too. eom MohRokTah Aug 2016 #14
If he trades in his wife for an American and gets a recognizably human haircut... yurbud Aug 2016 #16

andym

(5,441 posts)
1. Do you have the data for this op?
Sun Aug 7, 2016, 05:11 PM
Aug 2016

It would be quite interesting. What about 1948? I thought Dewey was outpolling Truman by a lot?

Doctor Jack

(3,072 posts)
2. True, I guess I got that one wrong
Sun Aug 7, 2016, 05:23 PM
Aug 2016

The colors on the graph are so close for Dewey and Truman that I got the two confused. Ok, so 4% of candidates have come back from being behind in the fall to win.




Keep in mind, there were 2 elections, 1980 and 1960, where there was some back and forth in the fall but A. the candidate that was usually ahead ended up wining and B. any election where the lead was consistent by now was over and done by this point. Most elections are consistent by now with no real back and forth.

Here are some random elections showing this point












http://www.gallup.com/poll/110548/gallup-presidential-election-trial-heat-trends.aspx

Doctor Jack

(3,072 posts)
5. Let's go with the best case scenario for trump
Sun Aug 7, 2016, 05:29 PM
Aug 2016

Polls in the 1940's were still lacking in any sophistication but lets lump them into the overall picture because it helps Trump's numbers, and we can see the absolute worst case scenario for our side. 1948 probably doesn't mean shit for this argument and Trump is likely much weaker than the 4% chance of winning that including that election would give him. However, I am the type of person that likes to see just how weak our position could possibly be to get an idea how good it actually is. And at a 96% chance including fudging the numbers a bit for Trump still means he is as close to hopeless as he can get.

I don't think that Clinton should be measuring the drapes in the oval office yet but maybe she should at least go out and buy a tape measurer.

Doctor Jack

(3,072 posts)
12. Gore was briefly ahead in September 2000 but was well behind most of the election
Sun Aug 7, 2016, 05:58 PM
Aug 2016

However, he likely won any way. But the official result was that Bush won, so take it as you will.

 

stevenleser

(32,886 posts)
3. 1948 would be an interesting asterisk to that because of the polling issues
Sun Aug 7, 2016, 05:23 PM
Aug 2016

Gallup had Truman 17 points behind in late September and its final pre election poll had Truman losing 50-45. In fact Truman won 50-45 so Gallups final poll was off by 10%. Other polling agencies had similar issues see http://www.math.uah.edu/stat/data/1948Election.html .

So was Truman behind by a lot in August and September? As the above link indicates there was a now known problem with sampling issues in those polls. For all we know Truman was never behind because if your samples are screwed up, you really can't draw any conclusions from the result of your poll.

DemocratSinceBirth

(99,704 posts)
6. Bush Pere was winning after both Conventions concluded
Sun Aug 7, 2016, 05:31 PM
Aug 2016






The Republican Convention concluded on August 18th.

Another way to think about conventions is that they help reset the race to equilibrium. In 1988, for example, George H.W. Bush headed into the conventions trailing Michael Dukakis despite conditions that seemed relatively favorable for Bush: The outgoing Republican president, Ronald Reagan, was fairly popular, and the American economy was in good shape. The conventions produced a big swing in the polls toward Bush, and he never looked back

http://fivethirtyeight.com/fea...

Doctor Jack

(3,072 posts)
7. That is the problem with these early conventions this year
Sun Aug 7, 2016, 05:33 PM
Aug 2016

In most elections, the conventions wouldn't take place until mid to late august. They were really early this year. That makes comparisons a bit difficult when matching up specific dates. Our place in the election right now is probably closer to early September in most years, at least in terms of polling.

As I said in another post, I want to base this on our worst case scenario, which is Trump still having a small chance of pulling a George HW Bush and pulling ahead by mid-August (despite the differences in convention dates) and including 1948 which had a lot of polling problems. Odds are, Trump is much worse off than the 4-8% chance I said in the original polls but at his very best, his situation is extremely dismal.

Doctor Jack

(3,072 posts)
10. Her team is extremely experienced
Sun Aug 7, 2016, 05:55 PM
Aug 2016

I'm not too worried about them staying on top of everything. Really though, what we see on the public stage is only a small fraction of the campaign. The behind the scenes work is what will win the election. Clinton's team is more experienced and much larger than even Obama's 2008 or 2012 operations. Trumps, on the other hand, doesn't even exist.

Clinton's campaign is playing 3D chess while Trump's is sitting in the corner eating paste. I'm not too concerned that her campaign is going to make a serious of fatal mistakes.

uponit7771

(90,225 posts)
13. I agree, we were saying the other night that Clinton would not only keep the trains running on time
Sun Aug 7, 2016, 05:59 PM
Aug 2016

... but is the most likely to build more tracks.

She said in an interview that the DNC gets are asses handed to us in off years, hopefully they're already implementing their off year strategy.

Few would be shocked if tRump comes out of this whole thing with more money than he went into it with... and the GOP is sunk

Jon Fogerty

(45 posts)
11. I hate these stats even if it is absolutely true
Sun Aug 7, 2016, 05:57 PM
Aug 2016

because it will reduce voter turnout (heck, she's going to win anyway so why should I bother), encourage "protest" voting (well, she's going to win anyway so I will vote for Trump to send a message) and it hides what many have noted: the "stealth" Trump supporters who publicly denounce him, but privately love and vote for him.

Brexit, anyone?

Doctor Jack

(3,072 posts)
15. The Brexit polls were pretty accurate
Sun Aug 7, 2016, 06:02 PM
Aug 2016

As the polls got closer to the actual vote, it showed a dead heat and it barely passed. It wasn't far off at all. However, polling in Britain isn't really relevant to the U.S.. British polls tend to be far more inaccurate than those in the U.S.

As for the "secret trump voter", they say that about the losing party every year and it never comes to pass. In 2008, there was a lot of speculation about a massive number people saying they would vote for Obama but would actually vote for McCain because they were afraid of looking racist. That group of people didn't end of existing .

If anything, Trump tends to underperform his polls. It happened all of the time in the primaries, where he would be polling at something like 36% but only get 31% when the state actually voted.

 

louis c

(8,652 posts)
17. The Brexit Analogy Needs Calrifying
Sun Aug 7, 2016, 07:56 PM
Aug 2016

Whites voted 53% to 47% to exit the EU. Britain is 90% white. If, in the election in the United States in November, Hillary gets 47% of the white vote, she will win by 12 points.

yurbud

(39,405 posts)
16. If he trades in his wife for an American and gets a recognizably human haircut...
Sun Aug 7, 2016, 07:48 PM
Aug 2016

it still wouldn't help, but it would be funny to see.

Latest Discussions»Retired Forums»2016 Postmortem»Reminder: 92% of Candidat...