2024 Election Polls

Robert and Gare take a look at the polls post Trump's felony conviction.

See omnystudio.com/listener for privacy information.

Episode Transcript

Speaker 1 (00:01):
Also media, welcome back to it could happen here a
podcast where Robert Evans is lying down on a couch
because he just feels exhausted from sleeping a leven full hours. Garrison,
you were much younger than me and don't seem to

(00:23):
feel exhausted because you just woke up after staying up
all night, did you? Yeah?

Speaker 2 (00:29):
No, not as exhausted.

Speaker 1 (00:31):
I hate do not hate you?

Speaker 2 (00:33):
Do you do? You know what is exhausting? Robert?

Speaker 1 (00:35):
Elections?

Speaker 2 (00:36):
The twenty twenty four presidential election.

Speaker 1 (00:39):
The twenty twenty four presidential election. Yeah, I hate it.
I hate it, Garrison, I hate it. But also I
have made a commitment. I have made a commitment to
making a prediction about the election this year and sticking
to it even though it's going to make everybody angry.
And I have a good reason for doing so. It's
because I want to try one of the rarest drugs

(01:01):
that exists in the world today. That Nate Silver Shit. See,
everyone's been wondering since like twenty twenty's what's up with
that guy? Did he like lose his mind? Was he
always kind of like out there and we just didn't
notice because he he got lucky a couple of elections
in a row. And the answer to that is no,

(01:24):
Nate was a pretty reasonable guy. He comes out of
like not politics. He only got into politics in two
thousand and six because they banned online gambling and he
got angry about it. And then he accurately predicted the
two thousand and eight and twenty twelve elections.

Speaker 2 (01:40):
Which wasn't hard to be frin, It wasn't hard.

Speaker 1 (01:42):
No, it was not. I mean, he got all the
states right, but it was just a matter because people
have pointed out he didn't seem to be nearly as
accurate in twenty sixteen or twenty twenty. There's a degree
of fairness to that, but like twight and twenty twelve
were our last non smartphone elections where they're wasn't this
like big you know, demon of social media kind of

(02:04):
hiding behind everything and making everything a lot weirder? And
I think part of you know, I think what ultimately
caused Nate's madness, though, is that in twenty sixteen he
did pretty well. He like laid out he was a
twenty nine percent chance of Trump winning, and when he
like explained what that chance was, how Trump might sweep
the blue Firewall States and whatnot. It's basically what wound

(02:27):
up happening, and as like a reward for being more
or less correct. While the election was going on, all
of the Democrats hated him because the news sources they
liked said that Trump had only a two percent chance
of winning, and then when the election was over, it
became like mainstream kind of reality to just say, Yeahate,

(02:47):
Nate fucked that one up. He finally screwed up, And
I think that that mix of things is what's driven
him insane. So I've decided to predict that there's a
twenty nine percent chance that the election is basically the
same as twenty twenty. And now, unlike Nate, I don't
have any kind of math to back that up. It's
just a gut feeling. But I'm I'm calling that now

(03:09):
because I want people to get really angry at me now,
and then ideally, when I'm right, they'll get even angrier
at me, and then I can go and say it
on social media and just gradly become completely unhinged and
see what it's like to be Nate Silver, the ultimate
high Garrison.

Speaker 2 (03:25):
See, I thought you were gonna say you thought there
was a twenty nine percent chance that Nate Silver would
just completely completely lose it and do some like like you.

Speaker 1 (03:36):
Do a major terrorism.

Speaker 2 (03:38):
Yeah, yeah, absolutely what I thought.

Speaker 1 (03:40):
Yeah, he drives a double decker bus into the end
of the Lincoln Memorial. God, that's my that's my hope.

Speaker 2 (03:47):
He storms the five thirty eight headquarters.

Speaker 1 (03:50):
Yeah, he's gonna take it back once and for all.

Speaker 2 (03:57):
Okay, so today we're going to be talking about election polling.
The debate is very very soon here in Atlanta, Georgia,
and as a little bit of a preparatory measure, we
want to go over some of the actual poll numbers
for the twenty twenty four presidential election. I like to
start with this Iowa poe from Seltzer and Co. Now,

(04:19):
Iowa's a weird one, right. Iowa has has gone read
pretty consistently the past two years, although twenty twenty was
closer than twenty sixteen. In twenty twenty, Trump won the
state by fifty three point one percent to Biden's forty
four point nine percent. But the numbers right now are
much much worse for Biden. Not good, No, it's it's

(04:40):
it's pretty bad. Trum Trump is leading Biden in the
general election in Iowa by eighteen percentage points, and third
party candidates, including Kennedy and the libertarian candidate Chase Oliver,
are receiving a combined fifteen percent support. It's it's pretty bad.
It hasn't been this bad in a while. Now. People
like to use this specific Iowa poll as kind of

(05:01):
a barometer for the Midwest in general, and that's, you know,
not completely accurate all the time, but it is something
that people do consistently point to as a general barometer
for Trump's possible success in the Midwest. Now we have
Yes Minnesota.

Speaker 1 (05:18):
Day because one of the most probably the most viable
path to Biden winning involves holding that quote unquote blue firewall,
which doesn't include Iowa obviously, it does include Michigan and Wisconsin,
both of which are generally within the margin of error
in most polls but looking very sketchy for Biden compared
to how he would like them to be at this point.

Speaker 2 (05:38):
Wisconsin's not looking great. Minnesota, according to a Survey USA
poll from just a few days ago, Biden is up
six points.

Speaker 1 (05:45):
Yes, Yes, Michigan is I think the one I was
saying is a little closer.

Speaker 2 (05:49):
Yeah, So that's kind of the situation with this poll.
And I'm not going to get into any of the
more specific numbers because the numbers in this Iowa poll
are going to be actually pretty reminiscent of more more
of the general election numbers which we're going to get into,
especially when we're going to start factoring in things like
the conviction and Trump's popularity among independence, which could very

(06:09):
well be a major deciding factor in this election. So
I'm going to quote from Forbes here. Quote. Trump leads
Biden by one point fifty to forty nine percent in
a CBS poll released Sunday that comes after a streak
of surveys found Trump's lead has slipped since his felony
conviction Manhattan last month, including a Fox News survey released Wednesday,
the nineteenth that shows Biden up by two points, a

(06:33):
three point swing since the network's May survey. This was
among a streak of five polls since mid June that
show Biden beating or tied with Trump unquote. So Biden
has made some considerable progress in the polls in the
past month. Biden and Trump are now tied in the
Morning Consults Weekly survey, as Biden has now been leading
Trump by a point in for two weeks in a row.

(06:55):
A month prior, Trump was way way ahead of Biden,
and the two are also tied in the Economist YouGov
survey released last Thursday, as well as a PBST Marrist
pull from Tuesday the eighteenth.

Speaker 1 (07:08):
Yeah, and there's a couple of things. I mean, like,
it's easy to say that's probably due to the conviction,
because that's the biggest thing that's happened since then. But
I also think there's a decent chance that some of
that is just the result of the fact that Trump
is now definitely the nominee, which was a little more
up in the air previously, so now people are kind
of forced to consider what that really means. But it

(07:29):
does seem in general like there's been motion and like
things have been moving in Biden's favor since the conviction.
So I don't think it's wrong to say that probably
overall the evidence suggests helps Biden at this point.

Speaker 2 (07:42):
Yeah, And the Fox News survey is really interesting because
they have this They have it on a graph here,
and you can see Biden steadily moving upwards on the
graph very consistently, and Trump has largely flatlined. If not
is actually kind of moving a little bit down. Robert F.
Kennedy Junior is also moving quite down. Yeah, not completely

(08:06):
surprising considering the whole brain brainworms thing.

Speaker 1 (08:09):
He's going to be the most interesting thing. I mean,
not the most interesting thing, because whether or not Trump
wins could mean whether or not we are able to
continue doing what we do. But RFK is kind of
the most interesting thing for me in terms of like,
are is it going to Is there going to be
any kind of evidence that there's actually real hunger for
a third party which everyone keeps talking about. Is this

(08:30):
constant topic of discussion in US politics, but it never happens,
and people were getting very RFK is obviously a bad
guy to pin your hopes on a viable third party on,
but I am interested to see if if it because
there's a decent evidence that the primary chunk because when
you factor in RFK, Biden's lead doesn't go down, right,

(08:53):
because RFK is really popular among a lot of the
independence that Trump is already strong with. And yeah, so
the big question is like, is he going to drain
votes from Trump or just kind of fizzle out? And
I think right now the smart money is kind of
on fizzling out. But it's it's a little hard to say.

Speaker 2 (09:11):
Do you know what we can say for sure?

Speaker 1 (09:12):
Though, Robert that uh, Robert F. Kennedy Junior is the
primary sponsor of this podcast.

Speaker 2 (09:18):
God, I hope, so, I hope, I really hope we
start getting some RFK ads on here.

Speaker 1 (09:22):
Look, folks, if you if you're not sure whether or
not you want to vote for RFK, we get it.
You know, obviously it's a this is a big choice.
But our recommendation is head down to the Gulf of
Galveston and shove your head in that in that Texas
coast water, get a couple of amibas rattling around on
that brainy ores, and then see how you feel about
our FK Junior. You know, all right, we are back.

Speaker 2 (09:55):
Let's talk a little bit about independence, because this voting
block will basically be the deciding factor in this whole election.
So that PBS Marrist poll that found Trump and Biden
tied also found that Trump has lost six points with
independence compared to their poll taken just before his conviction
last May, and Biden has gained eight points with independence

(10:19):
and now leads Trump by two points in that category.
And Similarly, the Fox poll also shows Biden leading by
nine points among Independence.

Speaker 1 (10:29):
And that's a massive, like shift. That's enough of a
shift that I wonder how much polling methodology maybe to
explain for it, Like were they just pulling these people?
Were they pulling them badly before? Or are they pulling
them badly now? Because that's that's quite a lot of movement.

Speaker 2 (10:45):
We'll talk a little bit about pulling methodology here at
the end, because it might. Yeah, it is, certainly the
polling methodology produces a large degree of the numbers. A
lot of these polls have a margin of error of
about three point five percent, but this this finding is
consistent across almost every single poll being done right now.
A political Ipsos poll from mid June found that thirty

(11:08):
two percent of Independence say they are now less likely
to support Trump after his conviction, with twenty one percent
saying it would be an important factor in their vote.

Speaker 1 (11:17):
Yep, and I did. This is the kind when we
were taught would talk in the group chat before the conviction.
I would. I made a note a couple of times
of the fact that there's a sizeable number of Americans
who are not what you'd call high information voters but
just feel really gross about voting for a felon. And
I think these are the kind of people who are
independence a lot of the times. They're not people who

(11:39):
think much about politics. They're people who make.

Speaker 2 (11:41):
The most part of him typically yeah.

Speaker 1 (11:43):
Yeah, yeah, And they can kind of make a swing,
gut decision on either of these guys in a moment.
And if you tell well, he's a felon, that matters
to some people. There's like the this frustrates a lot
of like high information political analysts, the fact that so
many Americans kind of like make almost random decisions like
the slip of a coid calls about what to do, but.

Speaker 2 (12:06):
Which is also what makes polling very hard. Is right,
But all polls also indicate that this will probably be
a much closer election than twenty twenties, and in an
election this close, small shifts among independents could very well
determine the outcome. Now, I'm going to quote from that
Political Episos report on their own poll quote quote a

(12:27):
plurality respondence. In our poll, thirty eight percent reported that
Trump's conviction would have no impact on their likelihood to
support Trump for president. Thirty three percent of respondent said
that the conviction made them less likely to support Trump,
while only seventeen percent said it made them more likely.
These results were worse for Trunk. Among respondents who said
they were political independents, thirty two percent said that the
conviction made them less likely to support and only twelve

(12:49):
said that it made them more likely to support Trump unquote.
And that same poll also found that nine percent of
Republicans say they're now less likely to support Trump.

Speaker 1 (12:59):
Yeah, which is massive, and that that actually makes me
want to bring up one of the guys the analysts
I've been reading, because this is actually the extent that
there's any real basis behind my twenty nine percent chance
things work out. Basically, like twenty twenty one, it's this this,
this fucking dude Helmet Norpoth norpoth Is. He's one of

(13:21):
these guys who's built a model, Like you get these
every now and then, like because they're great content for
TV news. Dudes like, oh, this guy's got a model
predict the election. His model predicted the last forty elections properly,
even though they like ran them through after we knew
how the elections were going to go. And I don't
know how fair that is. Helmet actually did accurately predict

(13:42):
a couple of like the last he's had his model
going the primary model for like the last seven elections,
and it predicted five of them correctly. Now, one it
got right was twenty sixteen, although it predicted how Trump
was going to win wrong, it got that he was
going to win right. I don't know how much credence
you want to give that. And he fucked up in
twenty twenty, although you know the fact that there was

(14:02):
a pandemic, then I'll give him a little bit of grace.
The other one he fucked up was twenty was two thousand,
but he called it. He called it for Gore. So yeah, Well,
and I'll read from his website describing like how this
works because it's relevant to what you're talking about in
terms of independent voters, and it's also relevant to what
I think is another major factor and who's going to
ultimately when, which is likely voters versus like correct If

(14:26):
I feel like it, I'll vote because Biden's lead jumps
substantially when you consider likely voters correct, whereas Trump does
very well with like maybe voters. And I kind of
don't feel like this is going to be a high
turnout election, right, That's that's what I am seeing.

Speaker 2 (14:41):
We have some data on this that I'll talk about later.

Speaker 1 (14:44):
Yeah, yeah, and Helmet's model works that way. So quote,
the primary model gives President Joe Biden a seventy five
percent chance to defeat Donald Trump in November. This forecast
takes account of the performance of the two candidates in
the early primaries. Biden won the Democratic contest in those
states by far larger margins than tru in Republican once.
What also benefits Biden and the general election is an
electoral cycle that fits the sitting president in a nutshell.

(15:07):
A White House incumbent facing no significant challenge in primaries
almost always wins reelection. As for the electoral college, the
most likely outcome of the twenty twenty four election predicted
by the model is that Biden will get three fifteen
and Trump two two three And basically, so part of
why I think this guy's probably a hack, but it's
kind of interesting is he's looking at how they performed
relative to each other in their primaries and could there's

(15:30):
a degree to which you can say, like, well, primaries
are absolutely not general elections. But what it does show
is relative how much Trump's support has faded from Republicans,
and Trump actually did considerably less well in the primaries
than he did in twenty twenty. Right totally, there was
a degree of actual like hunger to vote for Kerry Lake,

(15:54):
whom I think is the Arizona right candidate who was
running a gagainst him, and he showed weakness in a
number of primary states that was not there in twenty twenty,
which suggests, along with the polling you showed, you know
that like nine percent of Republicans are less likely to
vote from after the conviction, an amount of weakness in
his base that could be pretty meaningful when we get

(16:18):
to the election. And I don't think it's been taken
into account enough by, for example, folks on the left
looking at how much everybody hates Joe Biden, which is
also a very real factor. But I think that people
are kind of denying the degree to which a lot
of folks who should be his base don't like Trump anymore.

Speaker 2 (16:34):
Yeah, and this is one of the weird things. Post
to conviction, there were some pundits who were trying to
make an argument that somehow the conviction would actually make
Trump a more popular choice, which maybe works if you're
like a contrarian, but it doesn't really make much sense.
And if you look at like the approval ratings for
the conviction and the verdict, they fall pretty pretty well

(16:56):
on party lines. It's really going to come down to independence.
And like everyone who's going to vote for Trump, who
like really really really want to vote for Trump, are
still going to vote for Trump, right, Like that's that's
that's how it goes.

Speaker 1 (17:09):
Absolutely not, and they will buy the I'm voting for
a felon hats that Facebook keeps trying to sell.

Speaker 2 (17:14):
Mate absolutely right, Like, those are not the people that
are in question, but there is a large number of
other people who do not own a mega hat who
are actually, you know, questionable in who they're going to
vote for. On this note, I'd like to like to
quote again from Forbes quote. Polls consistently show the conviction
is a low priority for most voters in deciding who
to actually cast their ballot for. The political Ipsos poll

(17:35):
found that fifty three percent said it's not important, so
that it's not important to their voting decision, while sixty
one percent in a Reuter's full released last week said
it won't impact their vote unquote. Now, one of the
clearer shifts that we have seen post verdict is a
sizable increase in Biden voters who list stopping Trump as
one of their main reasons to do so. This we
have numbers from March to now, is that the main

(17:57):
reason for supporting Biden. In March, we had forty seven
percent saying it's to oppose Trump. Now it's fifty four
percent saying it's to oppost Trump. Which I think that
number will only increase the closer we get to the election,
because people don't want Trump to be president again, even
though they don't like Biden.

Speaker 3 (18:15):
Like.

Speaker 2 (18:15):
The other thing with these numbers is that the percent
of people who say I like Biden as reason for
supporting him has decreased since March. Yes, yes, it's decreased
by four percent.

Speaker 1 (18:27):
Of course, because he's not likable and he's he shouldn't
be president still. But Trump is even like and people
understand like I that is the number one thing. When
I go out of like my the bubble of my
friends and whatnot and talk to family members or just
like have conversations with right like uber drivers or whatnot

(18:47):
about politics, I have not heard a single person state
a reason for vote. They want to vote for Biden.
That is more important than I don't want Trump to
be president. That is everyone that I encounter basically like
I'm I'm obviously you have other people, but it is
weird to the extent to which that's what this election
is going to come down on. And I kind of

(19:08):
think it's evidence that of a failure and strategy in
Trump's part, because I think he probably could do better
if he were to focus on allaying those fears that
he wants to become a dictator as opposed to harping
on like one of the things that's interesting to me.
He's campaigning very heavily in Wisconsin right now. He's already
made like two visits just to southeast Wisconsin in the

(19:30):
last two months. Because Wisconsin is up for grabs right
Every poll I've seen basically is within margin of air.
It's either guy's game, and it's a critical state. And
Trump is hammering Biden on crime in Wisconsin right like,
look at how your dims have done, look at how
much more violent this city's become. And about one percent
of registered voters in Wisconsin consider crime a major concern

(19:53):
and a presidential election. And part of that's because, like,
violent crime has dropped and like massively in Wisconsin and
wide over the last year. And I do wonder the
extent to which, because Americans views on crime are not
based on how bad crime actually is. But I also
wonder if people are start like the degree to which

(20:13):
that's a voting concern for people is fading because it
has dropped so much. And I'll be curious to see
if kind of Trump's strategy of hammering the Democrats because
they're bad on crime is going to prove to be
a serious misstep.

Speaker 2 (20:27):
Well, even Fox News has had to do recent segments
talking about how there actually has been a drop in crime,
even though Americans feel like it hasn't, which is a
quite funny little tidbit. We're all looking for the guy
who did this moment. Now, I do want to get
through a few more conviction numbers. I'm going to quote
from Politicalist report on their own poll regarding the importance

(20:48):
of the conviction and people's vote. Quote, twenty two percent
of respondents said the conviction is important to how they
will vote and that it will make them less likely
to support Trump. Only six percent of respondents took the
other out of the question, saying they are more likely
to support a nearly identical net negative effect showed up
among independents, with twenty one percent saying they are less
likely to support and five percent saying they are more

(21:10):
likely unquote now. Of those who say the conviction is
important to how they will vote, seven percent of Republicans
say they are less likely to support Trump. So that's
an interesting number, and only thirteen percent say they are
more likely, and like, come on, those people were always
going to vote for Trump anyway. Forty percent of Democrats,
of course, say that they are less likely now. Twenty

(21:31):
eight percent of Republicans say that the conviction makes them
more likely to support Trump, but it won't affect their vote,
and among those who said the conviction isn't important to
how they will vote, forty percent said that it has
basically no impact on their support of Trump. Most those
people are independents now. Political also asked respondents if they
thought the prosecution was brought to help Joe Biden, and

(21:52):
most around fifty one percent disagree with the claim, but
forty three percent agreed and said that the case had
probably been been brought to help Biden, and these results
are roughly similar among Independence, so still most people don't
think so, and there's people who have, you know, suspicions,
not not super surprising.

Speaker 3 (22:10):
Now.

Speaker 2 (22:10):
Political notes that these figures might be movable, though these
are not necessarily locked down opinions as quote roughly a
third of all responders and Independence said that they still
do not understand the details of the case well unquote,
so glorious that those are not really set in stone.
And Political also notes that there's a number of upcoming

(22:33):
events and variables that could change of the public's opinion
before November, you know, including all of the ongoing efforts
by political operatives to influence people, the public perception of
both the conviction and just you know, the election in general.

Speaker 1 (22:46):
The debates obviously too.

Speaker 2 (22:48):
The debates as well as Trump's sentencing in Manhattan on
July eleventh, which could possibly you know, entail a period
of incarceration. Probably not, but if it did, would that
would certainly impact impact these numbers. And also Manhattan Dish
Attorney Alvin Bragg's testimony before Congress on July twelfth, This
could impact the numbers you know regarding you know, how

(23:10):
many people think this case is legit versus how much
people think is just purely like a political move. But
still about half of adults do approve of Trump's conviction.
The AP did a poll with the NORC a week
after Trump's conviction, but before Hunter Biden was convicted on
that federal gun case, and US also seems more likely
to support Trump's conviction than they are to disapprove, with

(23:30):
at least forty eight percent saying they approve and just
twenty nine somewhat or strongly disapproving and twenty one you know,
not approving or disapproving, to quote from the AP, Republicans
are less united on the verdict than Democrats, Roughly six
and ten Republicans disapprove, while fifteen percent approve. The other
two and ten and neither approve nor disapprove. Overall, opinions

(23:52):
on Trump have barely budged. About six and ten US
adults have an unfavorable opinion of Trump, just in line
from our fine In a poll conducted last February, four
in ten have a favorable view of Trump when also
largely unchanged since February. The numbers are equally poor for Biden.
For in ten US adults have a favorable view of
the Democratic president, while six and ten have a negative

(24:15):
one unquote.

Speaker 1 (24:16):
Yeah, this is very much unique in races that I
can recall a race to the bottom, like who can
alienate who will alienate less of the base?

Speaker 2 (24:26):
Right, yeah, no, we polls consistently are showing that there
will be historically that there is historically low voter enthusiasm.
Both candidates have very low favorability ratings, and an NBC
poll found that sixty four percent of voters say that
they are very interested in this year's election, which is
a twenty year low, so, you know, not great numbers.

(24:50):
And a new CBS poll found that among young Americans
who did vote in twenty twenty, only three quarters to
say that they'll definitely do so again. Now this this
poll also does know that Trump's support among young voters
has been almost unchanged since twenty twenty.

Speaker 1 (25:05):
Yeah, he's got done about two percent better, which is
fairly minimal considering how much Biden's lead is among that group.

Speaker 2 (25:12):
But overall, young voters do believe generally progressive values pretty consistently,
including support for a ceasefire.

Speaker 1 (25:20):
And that's I mean, part of the reason why we
may not see which you could be catastrophic for Biden
because twenty twenty, a lot of his win came in
the fact that he did deliver so much of that,
like so many young voters came out, turnout was so high,
and they overwhelmingly supported Biden. There is also, i mean,
kind of a reason why that might not wind up mattering,

(25:41):
which is where Biden I mentioned earlier. Biden does really
well among likely voters, much better than he does among
the general electorate. And this is part of a shift
among white voters with degrees that has been We get
a lot of talk and this has been significant, especially
like Latino voters shifting to ourswards. The GOP has been

(26:01):
a really important story too, but this one does not
get talked about as much. In the four years since
Biden took office, white men with degrees have shifted twenty
four points towards Biden, and he has gained nineteen points
among white women with degrees, which is like a huge
amount of his support. And also that's one of the
groups that's likeliest to vote. Like the strength that Biden

(26:23):
has gained among kind of middle of the road leaning
conservative suburban voters is potentially going to be a de
sider in this election.

Speaker 2 (26:34):
Yeah, and according to The New York Times and Sienna,
the polls do seem slightly skewed in Trump's favor actually
this year, mostly by disenfranchised voters who may not participate
in the upcoming election. And analysis they did found that
Biden had led the last three of their polls among
twenty twenty voters, but trailed among registered voters overall, which
is basically exactly what you're.

Speaker 1 (26:54):
Saying, you know, Garrison, speaking of likely voters are I
don't know, that doesn't really lead in here's the fucking ads. Look,
you don't get you don't get a good one. Every
time we do this, folks, there's too.

Speaker 2 (27:08):
Many more are likely to listen to these ads. They're fine,
So we are back. Is polling actually useful? Is this
actually useful anymore? The answer is kinda. But you know,

(27:30):
people have gotten really really anti polling in recent years.
You know, it's the twenty sixteen elections certainly contributed to that,
although if you look at the actual twenty sixteen polls,
it's kind of it's kind of interesting. In twenty sixteen,
Clinton generally pulled much higher than Trump for the duration
of the race. Though in late July the two were
neck and neck, with the with the gap closing once
again in late September and the week of the election,

(27:53):
Trump was on average trailing by less than three point
five percent behind, which is often in the margin of
they are for these polls and posters usually consider something
under three percent being a toss up. Now this is
three point five percent, so still is it was trending
towards Clinton, and there is reasons why in terms of
their polling methodology that was flawed. But the polls were

(28:15):
actually a bit closer than I think what public perception
seems to remember of the twenty sixteen polls.

Speaker 1 (28:21):
Yes, and this is a part of why the public
memory of twenty sixteen and to an extent, twenty twenty
and to an extent every election is so bad. Is
you can't emphasize this enough. People are dogshit and understanding
what poles say, right. They are really bad at understanding uncertainty.

(28:41):
One of the things that I hate to keep going
back to the Nate silver Well, but I think he's
a fascinating case study. And one of the things he
pointed out after twenty sixteen the minute you have a
forecast where there's less certainty. People don't like that. The
minute you have a forecast that doesn't have a Democrat winning,
they don't like that very much. And it's to point out,
like his he kind of started to become a heel

(29:03):
as soon as he started showing that like Trump had
a real shot at winning, and as his forecast continue
to show kind of weakness among Democrats, it got people
angrier and angrier. And that's most of what makes people
determine whether or not something is a credible source on
the election. And that's kind of why a lot of
this is like a doomed effort, is because people consider,

(29:27):
you know, an expert credible if they are saying something
they want to hear, because most of what people want
in terms of election polling is to feel reassured that
things are going to be okay, right, and that's that's
your kind of It's always like a confirmation bias game.
And it's also one of those things where like the
instant you do well, if you are legitimately a rigorous,

(29:49):
you know, expert, and you predict things correctly, you're going
to suddenly be this focus of so much media attention
and have so much money and job offers thrown your
way that it will inherently drive you mad, which is
part of why again, I am predicting a twenty nine
percent chance that things are basically the way they were
in twenty twenty, so I can get all that sweet,
sweet CNN money, you know, if I wind up being right.

(30:11):
I am curious Garrison kind of in that line, because
as our official poll expert, you kind of came into this,
I don't think with a strong set of biases about
what would happen when you actually started drilling down into
the numbers. Did that change it all your impression of
what was going on this election?

Speaker 2 (30:29):
I think I thought that the numbers for Biden would
be slightly worse. I think that's kind of the general feeling,
and that has been you know, what the numbers have
kind of looked like in my cursory glances the past
few months. But looking more into kind of polling science,
what these pollsters are are saying the gap is usually
within this three percent, that it feels like it's going

(30:50):
to be a very close election. It'll be much closer
than it was in twenty twenty. Poles thought that twenty
twenty would be a much, much more obvious win for Biden.
It was, it was it was a closer election than
what people thought. But this I think will be even
even closer. So gonna it's gonna be a tricky one.
We're going to be kind of on the edge of
our seat come from election night, which is what no

(31:12):
one wants to hear, right, No, especially since you have this,
You have a lot of people who want to hear
Biden is doomed because they have, for generally good reasons,
come to despise Biden over the last totally years, and
they just want to know that like the things they're
angry about matter.

Speaker 1 (31:29):
And the thing that I all I can say to
those people is like, I don't know that anything matters,
and I do think there's a really good chance. I
think this is basically a coin flip.

Speaker 2 (31:40):
Yeah, And I think you know, polling is is going
to look very different this year because Trump is not
the incumbent. I think there's a lot of other factors
that are contributing to the polls, and pollsters have adjusted
a lot since twenty sixteen to make sure that more
Trump support is accounted for both in twenty twenty and
in twenty sixteen. The error did not come from overestimating

(32:03):
the support of Clinton and Biden. It came from underestimating
Trump's support. And this has been fixed fixed for via
a number a number of methods. You know, there's certain
theories people have had, like the quote unquote shy Trump
voter theory, which is kind of largely disputed, saying that
you know.

Speaker 1 (32:21):
People certainly by this fucking point.

Speaker 2 (32:24):
Yes, no, saying that people who like support Trump are
too scared to tell polsters that they support Trump. Quite
quite silly. It's essentially it's essentially blaming, blaming, like pull
errors on people just lying to pollsters because they're too nervous. So,
I don't know. There's a lot of other stuff we have.
We have adjusted for white, non college educated voters, you know,

(32:46):
because people who have a college degree are more likely
to respond to polls. So all this does get adjusted for,
especially since twenty sixteen, because that was the main cause
of the polls kind of being fucked up that year.
So what exactly happened in twenty twenty Then if these
things like the non college vote and the shy Trump
voter theory were sort of adjusted for, well, a few

(33:07):
things happened. The pandemic one, you know, made certain pulling
figures a little bit unique. The election also featured the
highest number of voter turnout in decades, something that we're
probably not expected to see in twenty twenty four. In
twenty twenty, the national polls were too favorable to Biden
by three point nine points, state polls by four point
three I'm going to read a report from the American

(33:30):
Association for Public Opinion Research analyzing twenty twenty election poll errors. Quote.
If the voter's most supportive of Trump were least likely
to participate in polls, then the polling error may be
explained as follows. Self identified Republicans who choose to respond
to polls are more likely to support Democrats, and those
who choose not to respond to polls are more likely

(33:53):
to support Republicans. Even if the correct percentage of self
identified Republicans were pulled, differences in their publicans who did
and did not respond could produce the observed polling error unquote.
If this was indeed the issue, it was probably made
worse by Trump and twenty twenty by being very disparaging
to polls, making his base probably less likely to honestly

(34:16):
engage with polling metrics, and both in twenty sixteen and
twenty twenty there was large, large post mortems among the
polling community trying to figure out how to improve, and
twenty twenty two's polls were more accurate than any election
since nineteen ninety eight, with almost no bias towards either party.

(34:36):
So that is a good side in terms of the
accuracy of polls.

Speaker 1 (34:42):
This not being nonsense.

Speaker 2 (34:43):
Yeah, correct, So a poster named Nathaniel Rackitch said, quote,
pollsterry utility isn't telling us who will win, but rather
in roughly how close a race is and therefore how
confident we should be in the outcome. Historically, candidate's leading
polls by at least twenty points have won in ninety
nine percent of the time, but candidate's leading polls made
less than three points have won just fifty five percent

(35:03):
of the time unquote, And that kind of lines up
with our current situation, right. Biden was even though the
polls were slightly skewed towards Biden in twenty twenty, he
was so far ahead that most of the polls in
terms of saying who would win, we're still correct because
Biden was just so far ahead this time. That will
not be the case. That's not that's not what the
poles are going to say. The polls are going to
show this being a much closer race, and that I

(35:25):
think that is what it's going to be come come November.
So yeah, that's kind of that's kind of the low
down of the current the current polling situation. I'll be
curious to see, you know, what the numbers are post
debate and especially after the sentencing in July.

Speaker 1 (35:40):
Yeah, we'll see. And I should note that Nate Silver
just released his official forecast today and it's it's almost
the opposite of that weird German man who gave Biden
a seventy five percent chance of winning. Nate gives Trump
a sixty five percent chance of winning. So we are
going to see which of the election pundits who make

(36:00):
their entire living off of gambling on elections winds up
getting to be feted on all of the talk shows
in like January of twenty twenty five.

Speaker 3 (36:08):
That'll be.

Speaker 2 (36:09):
That's so you'll toss up.

Speaker 1 (36:10):
Honestly, that's thereal'll toss up. It's Helmet versus Nate, Baby,
who's gonna win? I kind of think they both might
be common.

Speaker 3 (36:25):
It could Happen here as a production of cool Zone Media.
For more podcasts from cool Zone Media, visit our website
coolzonemedia dot com or check us out on the iHeartRadio app,
Apple Podcasts, or wherever you listen to podcasts. You can
find sources for It could Happen here, updated monthly at
coolzonemedia dot com slash sources. Thanks for listening.