Trump has a Hillary 2016 problem

[GeneralMacArthur]

In 2016, Hillary consistently outpolled Trump, but she also consistently ran into polling ceilings both nationally and in the states.  The fact that she was never able to actually hit 50% should have been a warning sign of how baked-in the anti-Hillary sentiment was.

Trump has the same problem.

Trump's ceiling nationally is 44%
In national polls, Trump has a 44% ceiling.  FiveThirtyEight has 101 polling results in September so far and Trump only breaks 44% in the following polls:
9/13 Fox:  46% (x2)
9/9 Rasmussen:  46%
9/9 Harris:  46%
9/4 Harris:  47%
9/2 Rasmussen:  45%
9/1 AtlasIntel:  46%

That's only 7 out of 101 polling results in which Trump breaks his 44% national ceiling, and Fox News is the only legit one (Rasmussen is GOP, Harris has 0 undecideds, AtlasIntel is a fake poll)


Trump's ceiling in Wisconsin is 45%
Trump has the same problem in Wisconsin, where of the 25 polls conducted in September, not a single one shows him above 45%.


Trump's ceiling in Pennsylvania is 46%
Pennsylvania hasn't received as much polling attention, but if we go back to the start of August, there are 31 polls, and the only ones showing him above 46% are:
9/3 Rasmussen:  49%
9/2 Monmouth:  47%
8/11 Emerson:  47%


Trump's ceiling in Michigan is 44%
Michigan similarly suffers from a lack of polling.  Since August 1 there have been 21 polls and the only ones showing him above 44% are an 8/21 Civiqs poll with Trump at 46% and an 8/28 Trafalgar poll where he's ahead of Biden with 47%.


Trump's ceiling in Arizona is 47%
Go back to the start of July, 34 polls, Gravis gives him a 48% and OnMessage gives him a 51% but that's it.


Trump has no ceiling in Florida
The above states all follow the same pattern.  There have been plenty of polling results but none of them get Trump the numbers he needs to beat Biden.  He always runs into a ceiling.  This is not the case in Florida.  In the last month there have been several polls putting Trump at 48, 49, even 50%.  Those are the kinds of numbers he needs if he wants to win a state and lends credibility to the idea that Florida is the best state for Trump to make a play.



In 2016, Trump received
46.1% of the NPV
47.22% in Wisconsin
48.18% in Pennsylvania
47.5% in Michigan
48.67% in Arizona
49.02% in Florida

He is president today not because he was strong in those states but because Hillary Clinton had a hard ceiling of support, and a lot of her haters voted for third-party candidates.  In Wisconsin, for instance, 6.33% of the vote went to third-party candidates.  In Michigan, 5.23%.  In Arizona, 6.2%.

Biden does not have the Hillary Clinton problem.  He is consistently able to reach or exceed 50% in all of these polls.  It seems very unlikely that Trump will be able to win Pennsylvania with 48% again this year, because 2016 was such an anomaly.  In 2012, for instance, only 1.44% of PA voted for third-party candidates.  A similar number this year, with Trump matching his 2016 result, would be a 2-point Trump loss (Biden 50.36%).

But Trump would be lucky to even match his 2016 results in these states, because he has a Hillary Clinton problem.  Even if people don't like Biden, they refuse to vote for Trump.  In hundreds of polls he's consistently blocked at 2-3% below his 2016 results.  The main difference from 2016 seems to be that instead of voting third-party, the voters he's lost are simply moving to Biden.

[Tender Branson]

Could be said of Biden/Harris too.

In fact, Biden is polling right where Hillary was polling at the time in 2016 in almost all swing states. Except TX and AZ.

We are in a Schrödingerˋs Cat situation: both things could be possible right now. We only know if the votes get counted.

[GeneralMacArthur]

Could be said of Biden/Harris too.

In fact, Biden is polling right where Hillary was polling at the time in 2016 in almost all swing states. Except TX and AZ.

Not true!  Let's take Wisconsin for example.

https://projects.fivethirtyeight.com/2016-election-forecast/wisconsin/

Hillary Clinton never polled above 50%.  Not in a single poll from the hundreds of polls conducted!  On this day in 2016, she was barely polling above 40%.

Compare to the 2020 Wisconsin polls:

https://projects.fivethirtyeight.com/polls/president-general/wisconsin/

50% is Biden's floor!  Fox has him at 49 and Siena and Marquette have him at 48.  Those are his worst results in September.


Let's also look at Pennsylvania.

https://projects.fivethirtyeight.com/2016-election-forecast/pennsylvania/
https://projects.fivethirtyeight.com/polls/president-general/pennsylvania/

once again, Hillary Clinton never polled above 50% in Pennsylvania.  Biden's numbers here are a little more shaky but he's still polling over 50% in about 3/4 of the polls.

[Tender Branson]

Of course did Hillary not poll much above 50 in the swing states ... because the 3rd party polling was much higher.

But the margins themselves are pretty similar between 2016 and today.

[Da2017]

I will add Biden is more liked. Biden does not have Clinton's baggage.

[Kuumo]

This is why I feel that this election is different than 2016. Biden has had a much more stable lead than Hillary, and has hit 50% in numerous national and state polls. Candidates rarely underperform the percentage support they receive in polling, even if they don't win any undecideds. Trump polling at close to 50% in Florida is why I think Florida stick with him by less than 1% while Wisconsin, Michigan, Pennsylvania, and Arizona deliver the election for Biden.

[Sir Mohamed]

Of course did Hillary not poll much above 50 in the swing states ... because the 3rd party polling was much higher.

But the margins themselves are pretty similar between 2016 and today.

The margins are less important when you're at 50% or above. It's a huge difference between leading a poll 45-41% or 50-46%. Both is a 4 pt margin, but in the latter, you still win even if the undecideds break mostly against you. In the first, you have a problem.

Additionally, Biden has not just been at or above 50 in numerous state and nationwide polls, polling data so far also suggests Biden is winning voters having a negative opinion of either candidate. Said group massively broke in Trump's favor in 2016.

[GeneralMacArthur]

Of course did Hillary not poll much above 50 in the swing states ... because the 3rd party polling was much higher.

But the margins themselves are pretty similar between 2016 and today.

The margins are less important when you're at 50% or above. It's a huge difference between leading a poll 45-41% or 50-46%. Both is a 4 pt margin, but in the latter, you still win even if the undecideds break mostly against you. In the first, you have a problem.

Additionally, Biden has not just been at or above 50 in numerous state and nationwide polls, polling data so far also suggests Biden is winning voters having a negative opinion of either candidate. Said group massively broke in Trump's favor in 2016.

Right.  In 2016, Hillary would lead Wisconsin by 7, but it would be like Clinton 44 - Trump 37.  Clinton had a ceiling and was never able to get past 45-46% in Wisconsin.  On election day, she got 46.45%.

In 2020, a Wisconsin +7 poll would be like Trump 44 - Biden 51.  It's Trump who's stuck below a 45% ceiling, while Biden has no ceiling and can poll well above 50%.

The point being, in 2016 Clinton needed undecided/3P voters to break her way, but poll after poll showed that this simply wasn't happening.  54% of Wisconsin was simply determined not to vote for Hillary, whether that was by voting for Trump or by writing in Harambe because lol.  In 2020, Trump not only has the exact same conundrum, but he also needs to pull voters away from Biden, since Biden is at or above 50% in all of these swing states.  Even if every single undecided/3P votes Trump's way in that +7 poll, Biden wins 51-49.

[ElectionsGuy]

You realize I'm going to come back and roast the crap out of you when this inevitably doesn't happen. This is perhaps the worst prediction I've seen in a while. I mean, 44% nationwide? He's at 45% approval right now!

Of course did Hillary not poll much above 50 in the swing states ... because the 3rd party polling was much higher.

But the margins themselves are pretty similar between 2016 and today.

What people fail to recognize is that it wasn't just the undecided vote that gave Trump the win in these states. Undecided voters would've had to go to Trump 80-100% in order for those poll averages to make sense. That didn't happen, exit polls show undecided voters leaning 50-60% Trump. Clinton's % in poll averages would've been ABOVE her vote share if there were fewer undecideds but the same margin. That's why Biden being at 50% is not the safe spot so many people think it is. He could easily get 49% if he's polling at 50%, that goes for Trump or any candidate too in any race like this that only has around 5% undecideds and relatively little third party support.

General MacArthur is using the 'Trump only won because of undecideds' type of logic to justify Hillary's loss from 2016 and using it to apply all or most undecideds to Biden.

[Pollster]

This is why I always say that vote share is far more important to watch in polling than margins.

It's far more common for sitting incumbents to have polling ceilings - Cruz, Donnelly, Heller, Manchin, McCaskill, and Tester all had one in 2018 that they consistently polled within the margin of error of.

In open seat races, all bets tend to be off. Bredesen had a very hard ceiling in his polling, whereas both Sinema and McSally fluctuated around the 40's in many post-primary polls.

[Beefalow and the Consumer]

I realize this is a distillation of several of the previous comments, but...

If Biden is polling north of 50% consistently in polls that allow 3rd parties and undecideds, Biden will win.

[GeneralMacArthur]

Bumping this thread because in the last two weeks, we've had lots more polls.

Overall, the results are as follows:

National:  There have been tons of polls, and Trump has gotten 45-47% in a few but the vast majority have him at or below 44%.

Wisconsin:  He's gotten a handful of results at 1% above his ceiling, but again it mostly holds.

Pennsylvania:  Firmly holds.

Michigan:  If we ignore Trafalgar, it holds -- only one poll at 1% above his ceiling.

Arizona:  Just the outlier ABC/WaPo poll.


National 44% ceiling
Trump has only broken this in the following polls:
9/29 RMG:  45%
9/28 Monmouth:  45%
9/28 Harris:  45%
9/23 Emerson:  47%
9/23 Rasmussen:  47%
9/17 Targoz:  46%
9/15 Rasmussen:  47%

Wisconsin 45% ceiling
9/29 Susquehanna:  46%
9/24 YouGov:  46%
9/16 ABC/WaPo:  46%

Pennsylvania 46% ceiling
Holding firm

Michigan 44% ceiling
9/29 Trafalgar:  47%
9/24 Trafalgar:  47%
9/24 YouGov:  45%

Arizona 47% ceiling
9/23 ABC/WaPo:  49%

[ProgressiveModerate]

I think the real problem for him is Biden is polling above 49.5% in states collectively worth 272 EVs (Hillary 2016 states - NV and plus the big 3). That means if literally all undecides broke for Trump, he would still have to dig into Biden's current support, which is not a good place to be as an incumbent who people already know.

[Alben Barkley]

Of course did Hillary not poll much above 50 in the swing states ... because the 3rd party polling was much higher.

But the margins themselves are pretty similar between 2016 and today.

The 3rd party polling was higher because the 3rd party support was higher because people didn’t like either candidate but they do like Joe which is why he’s getting above 50 which means HE WINS!

Jesus Christ, how dense can you possibly be?

[Alben Barkley]

You realize I'm going to come back and roast the crap out of you when this inevitably doesn't happen. This is perhaps the worst prediction I've seen in a while. I mean, 44% nationwide? He's at 45% approval right now!

Of course did Hillary not poll much above 50 in the swing states ... because the 3rd party polling was much higher.

But the margins themselves are pretty similar between 2016 and today.

What people fail to recognize is that it wasn't just the undecided vote that gave Trump the win in these states. Undecided voters would've had to go to Trump 80-100% in order for those poll averages to make sense. That didn't happen, exit polls show undecided voters leaning 50-60% Trump. Clinton's % in poll averages would've been ABOVE her vote share if there were fewer undecideds but the same margin. That's why Biden being at 50% is not the safe spot so many people think it is. He could easily get 49% if he's polling at 50%, that goes for Trump or any candidate too in any race like this that only has around 5% undecideds and relatively little third party support.

General MacArthur is using the 'Trump only won because of undecideds' type of logic to justify Hillary's loss from 2016 and using it to apply all or most undecideds to Biden.

What the f—k are you talking about? No, Trump did not have to win 80-100% of undecideds last time to win. Combination of winning a majority of them down the home stretch, depressed Democratic turnout, and high third party vote was enough for him to just BARELY win. That’s how he won Wisconsin despite getting fewer votes than Mitt Romney. Yes, some of the polls were slightly off, but most were within the margin of error AND pollsters have adjusted for their mistakes since then AND both candidates then were polling low, unlike now, meaning more uncertainty and more room to grow. That is precisely why FiveThirtyEight gave Trump a 30% chance to win despite being down in the polls.

It is obvious now that being at or above 50 is a FAR better position to be in than unable to crack 44. There is no universe where that is not the case. Even if we say all the polls are off in Trump’s favor and Biden’s actually at 49, he’d still need almost all undecideds this time, unlike last time, precisely because there are fewer of them. Because most of them are going to Biden instead of Trump or third parties this time. Because Biden is liked, unlike Hillary or Trump. And even those that don’t like either prefer Biden, unlike 2016 when they preferred Trump. Who is polling worse than any incumbent since HW or Carter, who both LOST.

I swear, this galaxy brain “math” is like if people were arguing how ackshually Bob Dole was ahead in 1996. It’s morbidly fascinating to see the insane lengths people will go to in order to deny the evidence of their own senses and pretend like up is down and black is white. I cannot WAIT for reality to hit you people like a ton of bricks next month. All it would have taken in 2016 was one tiny thing to go wrong for Trump and he would have lost. But because he drew an inside straight at the last minute, people think he’s unbeatable. Completely ridiculous. Now he is in a FAR worse position, and FAR more would have to change FAR more drastically, and yet more people think he’s gonna win now than they did then. Well, it was dumb to think Hillary was a lock in 2016 and it’s even dumber to think Biden is NOT a lock now. Twisting your brain into knots, denying or warping the data to force it to fit your priors... these things do not lead to accurate predictions. They just make you look stupid.

[Amenhotep Bakari-Sellers]

Trump is it found out tried to recruit Mark Cuban to run, so he can take votes away from Biden, Trump is not confident of a win if there isn't a large enough third party share

[Turbo Flame]

It all comes down to consistency. If the poll numbers are strong as of today on Halloween, then Joe Biden wins. No questions asked. But somehow, Trump will get lucky like last time.

[It’s so Joever]

It all comes down to consistency. If the poll numbers are strong as of today on Halloween, then Joe Biden wins. No questions asked. But somehow, Trump will get lucky like last time.

You act like it’s certain.

[Turbo Flame]

It all comes down to consistency. If the poll numbers are strong as of today on Halloween, then Joe Biden wins. No questions asked. But somehow, Trump will get lucky like last time.

You act like it’s certain.

Anything can happen. Maybe an October Suprise favors Biden and gets a boost. I don't know. We'll just have to see how it plays out.

[ProgressiveModerate]

It all comes down to consistency. If the poll numbers are strong as of today on Halloween, then Joe Biden wins. No questions asked. But somehow, Trump will get lucky like last time.

You act like it’s certain.

Anything can happen. Maybe an October Suprise favors Biden and gets a boost. I don't know. We'll just have to see how it plays out.

So shouldn't Biden be favored to win? There is a big difference between Trump being able to win vs being favored to win.

[Turbo Flame]

It all comes down to consistency. If the poll numbers are strong as of today on Halloween, then Joe Biden wins. No questions asked. But somehow, Trump will get lucky like last time.

You act like it’s certain.

Anything can happen. Maybe an October Suprise favors Biden and gets a boost. I don't know. We'll just have to see how it plays out.

So shouldn't be favored to win? There is a big difference between Trump being able to win vs being favored to win.

Yea, I have speech issues. That's what happens when people like me have Autism. Sorry about that. I meant that an October Surprise that lifts Biden that finishes Trump for sure. I know Biden is the favorite, for right now at least.

[GeneralMacArthur]

Yes, we get it.  Anything can happen.  Hillary could have won the 2016 election if she had released a cure for cancer on November 1.

The point of this thread is to act as a counterpoint to the fears about a repeat of the 2016 polling errors that led to Trump's victory.

[ElectionsGuy]

In 2016, Hillary consistently outpolled Trump, but she also consistently ran into polling ceilings both nationally and in the states.  The fact that she was never able to actually hit 50% should have been a warning sign of how baked-in the anti-Hillary sentiment was.

Trump has the same problem.

Trump's ceiling nationally is 44%
In national polls, Trump has a 44% ceiling.  FiveThirtyEight has 101 polling results in September so far and Trump only breaks 44% in the following polls:
9/13 Fox:  46% (x2)
9/9 Rasmussen:  46%
9/9 Harris:  46%
9/4 Harris:  47%
9/2 Rasmussen:  45%
9/1 AtlasIntel:  46%

That's only 7 out of 101 polling results in which Trump breaks his 44% national ceiling, and Fox News is the only legit one (Rasmussen is GOP, Harris has 0 undecideds, AtlasIntel is a fake poll)


Trump's ceiling in Wisconsin is 45%
Trump has the same problem in Wisconsin, where of the 25 polls conducted in September, not a single one shows him above 45%.


Trump's ceiling in Pennsylvania is 46%
Pennsylvania hasn't received as much polling attention, but if we go back to the start of August, there are 31 polls, and the only ones showing him above 46% are:
9/3 Rasmussen:  49%
9/2 Monmouth:  47%
8/11 Emerson:  47%


Trump's ceiling in Michigan is 44%
Michigan similarly suffers from a lack of polling.  Since August 1 there have been 21 polls and the only ones showing him above 44% are an 8/21 Civiqs poll with Trump at 46% and an 8/28 Trafalgar poll where he's ahead of Biden with 47%.


Trump's ceiling in Arizona is 47%
Go back to the start of July, 34 polls, Gravis gives him a 48% and OnMessage gives him a 51% but that's it.


Trump has no ceiling in Florida
The above states all follow the same pattern.  There have been plenty of polling results but none of them get Trump the numbers he needs to beat Biden.  He always runs into a ceiling.  This is not the case in Florida.  In the last month there have been several polls putting Trump at 48, 49, even 50%.  Those are the kinds of numbers he needs if he wants to win a state and lends credibility to the idea that Florida is the best state for Trump to make a play.



In 2016, Trump received
46.1% of the NPV
47.22% in Wisconsin
48.18% in Pennsylvania
47.5% in Michigan
48.67% in Arizona
49.02% in Florida

He is president today not because he was strong in those states but because Hillary Clinton had a hard ceiling of support, and a lot of her haters voted for third-party candidates.  In Wisconsin, for instance, 6.33% of the vote went to third-party candidates.  In Michigan, 5.23%.  In Arizona, 6.2%.

Biden does not have the Hillary Clinton problem.  He is consistently able to reach or exceed 50% in all of these polls.  It seems very unlikely that Trump will be able to win Pennsylvania with 48% again this year, because 2016 was such an anomaly.  In 2012, for instance, only 1.44% of PA voted for third-party candidates.  A similar number this year, with Trump matching his 2016 result, would be a 2-point Trump loss (Biden 50.36%).

But Trump would be lucky to even match his 2016 results in these states, because he has a Hillary Clinton problem.  Even if people don't like Biden, they refuse to vote for Trump.  In hundreds of polls he's consistently blocked at 2-3% below his 2016 results.  The main difference from 2016 seems to be that instead of voting third-party, the voters he's lost are simply moving to Biden.

Trump's ceiling nationally is 44% ... gets 47%
Trump's ceiling in Wisconsin is 45% ... gets 49%
Trump's ceiling in Pennsylvania is 46% ... gets 49%
Trump's ceiling in Michigan is 44% ... gets 48%
Trump's ceiling in Arizona is 47% ... gets 49%

Good job General MacArthur.

[Skill and Chance]

In 2016, Hillary consistently outpolled Trump, but she also consistently ran into polling ceilings both nationally and in the states.  The fact that she was never able to actually hit 50% should have been a warning sign of how baked-in the anti-Hillary sentiment was.

Trump has the same problem.

Trump's ceiling nationally is 44%
In national polls, Trump has a 44% ceiling.  FiveThirtyEight has 101 polling results in September so far and Trump only breaks 44% in the following polls:
9/13 Fox:  46% (x2)
9/9 Rasmussen:  46%
9/9 Harris:  46%
9/4 Harris:  47%
9/2 Rasmussen:  45%
9/1 AtlasIntel:  46%

That's only 7 out of 101 polling results in which Trump breaks his 44% national ceiling, and Fox News is the only legit one (Rasmussen is GOP, Harris has 0 undecideds, AtlasIntel is a fake poll)


Trump's ceiling in Wisconsin is 45%
Trump has the same problem in Wisconsin, where of the 25 polls conducted in September, not a single one shows him above 45%.


Trump's ceiling in Pennsylvania is 46%
Pennsylvania hasn't received as much polling attention, but if we go back to the start of August, there are 31 polls, and the only ones showing him above 46% are:
9/3 Rasmussen:  49%
9/2 Monmouth:  47%
8/11 Emerson:  47%


Trump's ceiling in Michigan is 44%
Michigan similarly suffers from a lack of polling.  Since August 1 there have been 21 polls and the only ones showing him above 44% are an 8/21 Civiqs poll with Trump at 46% and an 8/28 Trafalgar poll where he's ahead of Biden with 47%.


Trump's ceiling in Arizona is 47%
Go back to the start of July, 34 polls, Gravis gives him a 48% and OnMessage gives him a 51% but that's it.


Trump has no ceiling in Florida
The above states all follow the same pattern.  There have been plenty of polling results but none of them get Trump the numbers he needs to beat Biden.  He always runs into a ceiling.  This is not the case in Florida.  In the last month there have been several polls putting Trump at 48, 49, even 50%.  Those are the kinds of numbers he needs if he wants to win a state and lends credibility to the idea that Florida is the best state for Trump to make a play.



In 2016, Trump received
46.1% of the NPV
47.22% in Wisconsin
48.18% in Pennsylvania
47.5% in Michigan
48.67% in Arizona
49.02% in Florida

He is president today not because he was strong in those states but because Hillary Clinton had a hard ceiling of support, and a lot of her haters voted for third-party candidates.  In Wisconsin, for instance, 6.33% of the vote went to third-party candidates.  In Michigan, 5.23%.  In Arizona, 6.2%.

Biden does not have the Hillary Clinton problem.  He is consistently able to reach or exceed 50% in all of these polls.  It seems very unlikely that Trump will be able to win Pennsylvania with 48% again this year, because 2016 was such an anomaly.  In 2012, for instance, only 1.44% of PA voted for third-party candidates.  A similar number this year, with Trump matching his 2016 result, would be a 2-point Trump loss (Biden 50.36%).

But Trump would be lucky to even match his 2016 results in these states, because he has a Hillary Clinton problem.  Even if people don't like Biden, they refuse to vote for Trump.  In hundreds of polls he's consistently blocked at 2-3% below his 2016 results.  The main difference from 2016 seems to be that instead of voting third-party, the voters he's lost are simply moving to Biden.

Trump's ceiling nationally is 44% ... gets 47%
Trump's ceiling in Wisconsin is 45% ... gets 49%
Trump's ceiling in Pennsylvania is 46% ... gets 49%
Trump's ceiling in Michigan is 44% ... gets 48%
Trump's ceiling in Arizona is 47% ... gets 49%

Good job General MacArthur.

The whole concept of ceilings is bogus unless you are talking about a country like Australia or Brazil with legally mandatory voting and 98%+ turnout.  Turnout was ~2/3rds of registered voters in the US, and that's the highest it has been in over a century.  There are always more voters to bring out for both sides. 

[One Term Floridian]

Trump, like Hillary, is stuck at 232 electoral votes