Muh Polls

785,715 Views | 5734 Replies | Last: 15 hrs ago by ts5641
Philip J Fry
How long do you want to ignore this user?
AG
Seems pretty clear to me. Jt just doesn't match the narrative they are trying to paint to get the communist vote out before it's too late
SwigAg11
How long do you want to ignore this user?
AG
Didn't see this posted earlier.

SwigAg11
How long do you want to ignore this user?
AG

If you look at what that pollster said after he dropped it, he bumped up the black voter estimate and some others (currently lagging in EV). Basically, he fudged his numbers.

YokelRidesAgain
How long do you want to ignore this user?
AG
JDUB08AG said:

On simple terms, why is he applying a 2.1% national vote factor when his own polling aggregate is .9%?
Because the national polling averages do not contribute much to the model's output.
No material on this site is intended to be a substitute for professional medical advice, diagnosis or treatment. See full Medical Disclaimer.
JDUB08AG
How long do you want to ignore this user?
AG
YokelRidesAgain said:

JDUB08AG said:

On simple terms, why is he applying a 2.1% national vote factor when his own polling aggregate is .9%?
Because the national polling averages do not contribute much to the model's output.


Then why on his website would he attribute an 85% chance of a trump victory if national polling is between 0 and .9%? It's just pure hedging and a way for him to say, welp you see? I did have something there even thought it didn't factor into my model. It's a load of crap.
YokelRidesAgain
How long do you want to ignore this user?
AG
JDUB08AG said:

Then why on his website would he attribute an 85% chance of a trump victory if national polling is between 0 and .9%?
He didn't say that; he says that if the actual margin of the popular vote is between 0 and Harris +1, Trump has an 85% chance of victory.

Again, Daugherty is just misrepresenting what Silver says.
No material on this site is intended to be a substitute for professional medical advice, diagnosis or treatment. See full Medical Disclaimer.
JDUB08AG
How long do you want to ignore this user?
AG
YokelRidesAgain said:

JDUB08AG said:

Then why on his website would he attribute an 85% chance of a trump victory if national polling is between 0 and .9%?
He didn't say that; he says that if the actual margin of the popular vote is between 0 and Harris +1, Trump has an 85% chance of victory.

Again, Daugherty is just misrepresenting what Silver says.


I really must be missing something. How does his own aggregate polling have her at a .9% margin yet he is factoring in a 2.1% margin in his model?
Tom Fox
How long do you want to ignore this user?
JDUB08AG said:

YokelRidesAgain said:

JDUB08AG said:

Then why on his website would he attribute an 85% chance of a trump victory if national polling is between 0 and .9%?
He didn't say that; he says that if the actual margin of the popular vote is between 0 and Harris +1, Trump has an 85% chance of victory.

Again, Daugherty is just misrepresenting what Silver says.


I really must be missing something. How does his own aggregate polling have her at a .9% margin yet he is factoring in a 2.1% margin in his model?
Me too. I just looked at his website and he has Harris at +0.7 nationally.

Edit: Unless this idiot is saying two different things in different places. In which case, you cannot trust what he is saying because he is politicking.
Philip J Fry
How long do you want to ignore this user?
AG
And RCP has Trump up .1.

What think I'm hearing is that Bate has KH up by .9 on his free site, but 2+ on his pay site. Isn't that convenient?
YokelRidesAgain
How long do you want to ignore this user?
AG
JDUB08AG said:


I really must be missing something. How does his own aggregate polling have her at a .9% margin yet he is factoring in a 2.1% margin in his model?
I really don't know what to say other than to keep repeating the same point: the model is primarily based on state polls, not the aggregate of national polls.

There is no "factoring in"; the model's output is what it is: the peak/50th percentile of the distribution (composed of 40,000 Monte Carlo simulations) is Harris +2.
No material on this site is intended to be a substitute for professional medical advice, diagnosis or treatment. See full Medical Disclaimer.
YokelRidesAgain
How long do you want to ignore this user?
AG
Tom Fox said:

Me too. I just looked at his website and he has Harris at +0.7 nationally.
The model is updated at least once per day; the numbers changed minimally from yesterday.
No material on this site is intended to be a substitute for professional medical advice, diagnosis or treatment. See full Medical Disclaimer.
Prosperdick
How long do you want to ignore this user?
AG
Just remember when people keep asking have they "fixed" the Trump under-polling, it's related directly to the social desirabilty bias. When people are polled over the phone they are more likely to state they are not voting for Trump in order to appease the poll interviewer. This is unique mainly to Trump (although I think you could make a case for Ted Cruz and Kari Lake as well).

There's also the fact that they aren't reaching the low propensity voters either so combine that with the above and just about every poll you're looking at will under-poll Trump.

As someone mentioned above, at this point the polls are 100% about influence and not accuracy. You're much better off on the early voting thread if you want more accuracy about election results.
backintexas2013
How long do you want to ignore this user?
AG
I think the NYT said white democrats are 16% more likely to respond to polls compared to white republicans.
Captn_Ag05
How long do you want to ignore this user?
AG


will25u
How long do you want to ignore this user?
As long as Republicans get out an vote on ED, it is looking good.

texagbeliever
How long do you want to ignore this user?
YokelRidesAgain said:

JDUB08AG said:


I really must be missing something. How does his own aggregate polling have her at a .9% margin yet he is factoring in a 2.1% margin in his model?
I really don't know what to say other than to keep repeating the same point: the model is primarily based on state polls, not the aggregate of national polls.

There is no "factoring in"; the model's output is what it is: the peak/50th percentile of the distribution (composed of 40,000 Monte Carlo simulations) is Harris +2.


So you think it is perfectly reasonable for the nation to shift 5 points to the right nationally but the swing states will have less than 1 percent shift to the right. Yeah that math just doesn't add up. If someone isn't willing to reconcile those discrepancies then they shouldn't be taken seriously.
sam callahan
How long do you want to ignore this user?
Regarding Silver, I have said before that anyone calculating odds to the decimal place based on polls with +/- 4% margin of error is doing much more marketing than statistics.
outofstateaggie
How long do you want to ignore this user?
AG
kgtg79
How long do you want to ignore this user?
Trump will get 300 EC
YokelRidesAgain
How long do you want to ignore this user?
AG
texagbeliever said:


So you think it is perfectly reasonable for the nation to shift 5 points to the right nationally but the swing states will have less than 1 percent shift to the right. Yeah that math just doesn't add up. If someone isn't willing to reconcile those discrepancies then they shouldn't be taken seriously.
It is true that the national polling suggests more of a movement toward Trump than swing state polling.

There are scenarios where that could be true (for example, Trump gaining a lot of votes in big states he is going to lose), but more likely either the national polls or the state polls are off. Silver's model, by design, puts more value on the aggregate of individual state polls. That is why the model "thinks" Harris is up by slightly more than national polling suggests.
No material on this site is intended to be a substitute for professional medical advice, diagnosis or treatment. See full Medical Disclaimer.
Casual Cynic
How long do you want to ignore this user?
Trump only needs to gain .2 percent from 2020 to win. He'll do that quite easily. Without covid to terrify voters Democrats don't have much of a message.
ttha_aggie_09
How long do you want to ignore this user?
AG
SIAP

nortex97
How long do you want to ignore this user?
AG

Silver goes bearish last second on Harris.





NH is going to be razor tight, but I think it stays blue:


Amazingly, over 76 million have already voted this am.
satexas
How long do you want to ignore this user?
AG
For whatever it's worth, the Vegas market moved Trump from -130 back up to -140 this morning (Electoral College, not Population)

(It's biggest margin of late was -200 3-4 days ago).

For those that don't understand Vegas (betting lines) ... basically +/-100 or +/-110 is 50/50 ('pickem') odds.

For those that think this is off-topic... it's not. It's predictive just like the polls are - where "the smart money is" on the election based on what people think of the polls and what will happen.

I watch Vegas more than I do the polls themselves...
Quo Vadis?
How long do you want to ignore this user?
satexas said:

For whatever it's worth, the Vegas market moved Trump from -130 back up to -140 this morning (Electoral College, not Population)

(It's biggest margin of late was -200 3-4 days ago).

For those that don't understand Vegas (betting lines) ... basically +/-100 or +/-110 is 50/50 ('pickem') odds.

For those that think this is off-topic... it's not. It's predictive just like the polls are - where "the smart money is" on the election based on what people think of the polls and what will happen.

I watch Vegas more than I do the polls themselves...


For you non gamblers, -140 is equal to about 58% chance Trump wins. It means you have to bet $140 dollars to net a $100 profit.
AtticusMatlock
How long do you want to ignore this user?
NH GOP may be able to flip a House seat with those numbers.
nortex97
How long do you want to ignore this user?
AG
I like the poll, other than the Indy splits. Agreed, I opened the pdf just hoping to see if they polled the House race, disappointed they didn't. The most recent one stretched back into October and had her down over 10 (mostly pre-debate).
nortex97
How long do you want to ignore this user?
AG


ts5641
How long do you want to ignore this user?
Those battlegrounds are so freaking tight!
Barnyard96
How long do you want to ignore this user?
AG
Dude from Trafalgar said to keep an eye on New Hampshire election night. They report early and will tell where the electorate is breaking. Biden won by 7 in 2020

Philip J Fry
How long do you want to ignore this user?
AG
https://polymarket.com/event/presidential-election-winner-2024?tid=1730730826795

Trump inching back to 60% on poly market
Gaeilge
How long do you want to ignore this user?
kgtg79
How long do you want to ignore this user?
Who is Kalshi?
Gaeilge
How long do you want to ignore this user?
kgtg79 said:

Who is Kalshi?
Betting Market
nortex97
How long do you want to ignore this user?
AG


First Page Last Page
Page 156 of 164
 
×
subscribe Verify your student status
See Subscription Benefits
Trial only available to users who have never subscribed or participated in a previous trial.