Trump was never going to win the popular vote

[edhopper]

His road to victory was always winning those few key States again.

So Biden's national lead is not a good indication for the election.

It's great that he is so much ahead, but we must focus on winning WI, MI and PA, and maybe take one or two of the States he won in 2016.

Let's keep our eyes on the prize.

[Under The Radar]

1. You are correct, he has lost many voters and has gained none

The electoral map started eroding for him on Inauguration Day. However the margin must be huge to overcome the expected voter suppression, the interference with voting machines, suppressing turnout because of the virus concerns and the close calls in numerous counties that will be challenged in court that included absentee ballots, and accusations of fraud. Election night will stretch out for weeks but not out of concern for the president, it will be for the senate seats.

[edhopper]

5. This is so true

and so important!

[Botany]

2. Hillary "won" WI, MI, PA, NC, and FL in 2016 .... the exit polls don't lie.

We should work on winning every state this year and most should be by double digits.

Yes, OK, AR, OK, WY, UT, WV, KY, and TN are long long shots. However much of the south is
now dead to Trump. The reason? No SEC football this year.

Vote early, vote on paper, vote by mail, and always find one down ticket race to vote for the
GOP in.

[unblock]

3. Exit polls are the gold standard for election integrity

In every country except the u.s.

For some inexplicable reason, the only people in the world who systematically lie to exit pollers are Americans who vote for republicans but tell exit pollers they voted for democrats.



Some decades from now, we'll discover the truth and learn that Republican cheating is even greater than we now know, and already we know that probably 1984 and 1972 are the only presidential elections since Eisenhower they would have won without cheating. And of course, even in 1972 they cheated big time anyway.

[Baked Potato]

4. I think Roger Stone knows.

[Botany]

6. With the old fashioned mechanical voting machines ..

... where you pulled down tabs and then voted by opening the curtain with the big lever the exit polls matched the vote. The Republicans know they have been cheating with electronic voting machines even before HAVA (help America vote act).

You can tell the panic in Trump and company when voting by mail cuts out their "cheat factor" but you can't rig a landslide.

Btw dick Nixon help to rig the 68 election by going to both the North and South Vietnamese and telling them they would get a better deal if they got out of the deal LBJ had after he got into the White House but in the end Nixon and Kissinger signed the exact same deal in '73.

[unblock]

8. Yup that's why I didn't include 1968. Another republican cheat.

As if mr. watergate was going to cheat in 1972 when he was widely expected to win anyway but not cheat in 1968 when it was far from certain.

Some people applauded him for "gracefully accepting defeat" in 1960 when he lost to jfk while some people claimed some jfk supporters cheated. Who knows if this is true or not, but I'm quite confident the reason Nixon didn't make a stink is because he wasn't sure jfk chested but he damn sure knew his side cheated. He couldn't stand having any kind of investigation.


I remember my parents taking me on Election Day back when they had those giant mechanical machines. They were awesome and yeah, hard to cheat.

[Drunken Irishman]

9. This is not true. Exit polls have a MOE just like normal polls.

Don't spread misinformation. Exit polls are a generally good indicator of providing a narrative for a race but not a good indicator of predicting a race, especially if it's a close race, because, like polling, it's still susceptible to that pesky MOE. And this isn't a recent phenomenon and it's NOT just a US issue, either.

In the 1992 UK General Election, two exit polls predicted a hung parliament. But the exit polls were off and the party in power kept their majority.

There are plenty of examples, even in past elections, of exit polls being off. In the 1992 GOP primary, the exit polls showed Pat Buchanan doing much better in the New Hampshire primary than he eventually did.

Exit polls become less reliable the closer the results. In fact, exit polls are becoming much more of a challenge due to early polling and people refusing to take the surveys. It's not a gold standard - it's a snapshot. Just like a poll is.

Exit polls accuracy is a legitimate issue. In 2016, the exit polls for New York had Hillary winning by 16 points. She won the state by 21.4. That's a dramatic swing. It doesn't mean she won by a larger margin due to fraud. National exit polls had her leading Trump 47.9% to 44.7%. She won the popular vote 48.02% to 45.93%. Not a significant difference.

Yes, Wisconsin and Pennsylvania were off - but Michigan showed the race a statistical tie in their exit polls, and it basically ended up being a tie. Same with Florida.

The problem isn't theft. It's voter suppression. But it does no one any good starting a point with a faulty premise. And the idea that exit polls have always been right just isn't true. There are multiple points, in primaries and general elections, presidential races and mayoral races, races in other countries, too, where the exit polls were wrong.

[unblock]

10. i agree with everything you said other than the suggestion that my post is inconsistent with yours.

by touting exit polls, i only mean that this is one of the most important monitoring and measuring tool international observers use in evaluating if an election is legitimate or not. they are not the only tool, they don't detect every manner of cheating, nor are they always 100% accurate. but they are worth their weight in gold in flushing out rigged elections and cheating.

i agree, for instance, that exit polls don't catch people who don't even show up because they're not allowed to vote (e.g., felony disenfranchisement, voter registration purges, id requirements or other effective poll taxes, etc.) and yes, of course they have a margin of error, which means they become less useful the more precise you're trying to be.

however, they're generally far more useful than pre-election polling because they catch people who actually showed up to vote and generally it doesn't have the risk of people changing their mind later because it happens minutes after they cast their ballot.

if you thought i meant that exit polls are always 100% correct and always more accurate than a legitimately run official vote, then of course, no, i don't agree with that strawman argument either.

[Drunken Irishman]

11. The problem with American exit polls compared to a lot of other countries...

...is that we're dealing with 50 different exit polls. We're also dealing with a recent phenomenon of close elections. The fact is, for a huge chunk of our modern election history, elections often weren't close. So, exit polling data was less likely to be wrong.

Exit polling data came into play in the 1960s. From 1960 to 1996, a span of of nearly 40 years, there were only two really competitive elections:

1960, Kennedy vs Nixon
1976, Carter vs Ford

Every other election was an electoral college blowout.

But since that 1996 Clinton blowout win, we've had extremely competitive elections in 2000, 2004, 2012 and 2016 (though, 2012 is the most lopsided of these mentioned). So it only reasons the exit polling data is going to be less accurate because we're dealing with extremely tight races.

In 2000, Bush 'won' by a paltry amount in Florida. But people also forget New Hampshire. Had Gore taken New Hampshire, he's elected president, even without Florida. So, take Florida out of the equation and three of the last five presidential elections have essentially come down to:

2000: 7,211 votes (Bush's margin in NH)
2004: 118,601 votes (Bush's margin in OH)
2016: 79,646 votes (Trump's margin in PA, WI and MI combined)

With how close those races proved to be, exit polling data is almost irrelevant at that point because everything falls into the MOE. For a typical state-based exit poll, the MOE for a 95% confidence interval is +/-4%.

Here was the exit poll margin for Hillary in WI, MI and PA at their initial poll closings (so, before adjusted based on actual vote trends):

PA: 4.4
MI: 0.0 (it was a tie in the exit polls)
WI: +3.9

Remember: the average MOE is +/- 4%. Every single one of those states fell into that MOE for exit polls.

BUT the exit polls were actually MORE accurate than the actual polls in WI & MI, as they showed each state closer than the average lead Clinton had on election day (3.4 in MI, 6.5 in WI), while exit polls were off in PA, where Clinton's average polling lead was 1.9.

So, where I disagree with you is that it's evidence of fraud. I think some people lie, but that's not why the exit polls were wrong. They were only wrong because these elections are becoming consistently, and ridiculously, close, which makes it difficult to predict based on exit polls only.

[unblock]

13. but you can still do analysis on the exit polls vs. official votes

there are enough precincts and different elections (not just presidential elections but also congressional elections, etc.) to come to some meaningful conclusions.

i don't know how good this guy is but he is claiming the exit poll analysis supports the argument that there is election fraud. this is from 2012, so obviously it doesn't say anything about 2016.

https://richardcharnin.wordpress.com/2012/06/25/election-fraud-an-introduction-to-exit-poll-probability-analysis/

[Awsi Dooger]

16. You linked to a guy who got booted out of here a long time ago

That is the famous/infamous TruthIsAll, the guy who asserted Kerry was 99.99% certain to win 2004 based on undecideds to the challenger.

I had many battles with him, as did others here. Everything is a conspiracy with him.

I heard he switched sides after Sanders didn't win the 2016 primary and became an unhinged Trump supporter.

Regardless, I hope he is okay. He had some serious health problems when he posted here.

[unblock]

17. Cripes! I didn't recognize the name

I think I was even the mod who pulled the plug. I thought his statistical analysis was pretty good but he had a thin skin when it came to feedback....

[Awsi Dooger]

18. Definitely good math but thin skinned

Particularly thin when it came to exit polls and anything related to the Kennedy assassination.

I remember the name primarily because after he got booted out of here another DUer corresponded with me emphasizing TruthIsAll wanted me to join him on some other site so we could debate. Many related PMs. The premise was ridiculous. He just wanted to flaunt his math. I never claimed to have his math skills. But I knew more than I needed to know to win at sports and politics and stock market. Still true.

[Drunken Irishman]

19. There are some major issues with his writings...

It compares one data point which isn't exactly perfect (pre-election polling) to another data point that isn't exactly perfect, either (exit polling). Again, we are dealing with margin of error percentages. I mentioned two states where Hillary actually did better in the initial exit polling data as opposed to the pre-election polling average (MI and WI). This speaks to that issue: all these measurements are just snapshots. The reality is that, the only tangible numbers we have are actually counted votes. Everything else is, unfortunately, a statistic based on a specific set of probabilities.

And that's the problem when using this data to prove much of anything. Remember, exit polls HAVE a MOE for a reason - and it's because, at the end of the day, it's still a survey and not actual results.

But this goes to my own overall point: with how narrow the elections have been the last 20 years (five election cycles), that random, weird events MAY be enough to alter the data and expose one of those fundamental issues that you discussed in your OP: Republicans being less inclined to want to take exit polls. It's not a significant issue - unless we're talking a very close election (say, within that MOE). Then it can absolutely have an impact.

But this article was written pre-2012 election (came out in June, 2012).

It's hard to find any initial exit polling data from 2012. There were some leaks that I found on an old DU post, though:

And those numbers were WAY more favorable to Romney than the actual results.

BUT doing searching, I did find CNN's initial exit polling data (which changes as votes come in).

In Virginia, their initial exit polls showed it tied, 49-49.

But Obama won Virginia 51% to 47 - so, again, a significant surge toward Obama.

Initial exit polling also had Obama and Romney tied in North Carolina. Romney won the state 50-48.

Ohio's exit poll was in favor of Obama 51-48 . Obama won the state ... 50-47.

Florida was 50-49 and that's about what Obama won the stat with.

So, again, this hits to my point that they aren't perfect. Neither NC or VA were tied in the final results and yet, the initial exit polls showed that.

One shift favored Romney (NC) and the other Obama (VA).

Was that fraud?

Or just the margin of error?

Beyond all that - for this guy's point to be correct, there would have to be wide-spread election fraud at every level of every state and I just can't get into the idea that something like that has happened, consistently, for many, many years. Sorry.

[Drunken Irishman]

7. Absolutely Biden's lead is a good indicator - just as Hillary's was in 2016.

Hillary's lead suggested the election was far narrower than 2012. Everyone should focus on the national popular vote because it's often the best indicator of how things are going in those swing states.

Take this as an example:

Had Hillary won the national popular vote by just five-points, without any change in, say, California (still winning it by 30%, which was a higher margin than Obama won it in 2012), her margin over Trump nationally would have been 6,842,875, or 3,974,357 MORE than in 2016 over Trump.

That near-4 million vote has to come from somewhere. It would have come from those exact states and since Hillary lost those states by a combined 77,000 or so, having won 4 million more votes on top of what she won in 2016 would have been enough.

So, you're wrong. The national lead is absolutely a good indicator of the election. In fact, it's the best indicator we have because national polls tend to be more accurate than local polls (larger sample size, often smaller MOE and more frequent polls).

[Demsrule86]

12. I am really tired of these sort of posts...Biden is doing great and Trump can't win the EC

with numbers like this...Biden is ahead in the blue wall states too...a landslide is what we need. WE can walk and chew gum at the same time.

[pecosbob]

14. If there is another stolen election, I wonder how will the disenfranchised seventy percent respond?

[smirkymonkey]

15. We are in the streets.

We will not accept another illegetimate election. We can not let it ever happen again. We have more power than we think we do.