Still think it's too early...
-
Do you buy groceries? Pay house and car insurance? Have some recent medical bills? Have many illegal aliens where you live? Does crime seem like it has increased? Think DEI is ruining the country?
Those are advantage Trump.
Abortion?
If you are in favor, advantage Harris.
-
@Horace said in Still think it's too early...:
seems difficult to get a bead on the perspectives of whomever is changing their minds about preferred candidates to lead to these polling results changes. I don't think those sorts of people ever post online or discuss politics in general. Everybody who discusses politics will never change their mind about anything, at least in the short term.
First, the Silver forecast is not a poll, it's a model that does take into effect poll changes, but also other criteria and trends.
Second, I think that polling changes a lot based on enthusiasm. The enthusiasm for the democrat candidate is waning a little.
@LuFins-Dad said in Still think it's too early...:
@Horace said in Still think it's too early...:
seems difficult to get a bead on the perspectives of whomever is changing their minds about preferred candidates to lead to these polling results changes. I don't think those sorts of people ever post online or discuss politics in general. Everybody who discusses politics will never change their mind about anything, at least in the short term.
First, the Silver forecast is not a poll, it's a model that does take into effect poll changes, but also other criteria and trends.
The model is only meaningful insofar as it maps to people's voting preferences changes. So to whatever extent this is a meaningful point, it's a point that would lead one to use the model results as toilet paper rather than a predictor of elections.
Second, I think that polling changes a lot based on enthusiasm. The enthusiasm for the democrat candidate is waning a little.
So if the model is accurate, this would give some insight into how people might change their minds. What, they read a bunch of enthusiastic media about a candidate and would be inclined to vote where they may not have voted before? Sure, possible, but again, these people don't discuss politics online or probably in person.
-
@LuFins-Dad said in Still think it's too early...:
@Horace said in Still think it's too early...:
seems difficult to get a bead on the perspectives of whomever is changing their minds about preferred candidates to lead to these polling results changes. I don't think those sorts of people ever post online or discuss politics in general. Everybody who discusses politics will never change their mind about anything, at least in the short term.
First, the Silver forecast is not a poll, it's a model that does take into effect poll changes, but also other criteria and trends.
The model is only meaningful insofar as it maps to people's voting preferences changes. So to whatever extent this is a meaningful point, it's a point that would lead one to use the model results as toilet paper rather than a predictor of elections.
Second, I think that polling changes a lot based on enthusiasm. The enthusiasm for the democrat candidate is waning a little.
So if the model is accurate, this would give some insight into how people might change their minds. What, they read a bunch of enthusiastic media about a candidate and would be inclined to vote where they may not have voted before? Sure, possible, but again, these people don't discuss politics online or probably in person.
@Horace said in Still think it's too early...:
@LuFins-Dad said in Still think it's too early...:
@Horace said in Still think it's too early...:
seems difficult to get a bead on the perspectives of whomever is changing their minds about preferred candidates to lead to these polling results changes. I don't think those sorts of people ever post online or discuss politics in general. Everybody who discusses politics will never change their mind about anything, at least in the short term.
First, the Silver forecast is not a poll, it's a model that does take into effect poll changes, but also other criteria and trends.
The model is only meaningful insofar as it maps to people's voting preferences changes. So to whatever extent this is a meaningful point, it's a point that would lead one to use the model results as toilet paper rather than a predictor of elections.
Second, I think that polling changes a lot based on enthusiasm. The enthusiasm for the democrat candidate is waning a little.
So if the model is accurate, this would give some insight into how people might change their minds. What, they read a bunch of enthusiastic media about a candidate and would be inclined to vote where they may not have voted before? Sure, possible, but again, these people don't discuss politics online or probably in person.
I think you're missing the single biggest challenges with the polls, and it has nothing to do with changing minds, it has to do with responses. Enthusiastic voters are more likely to respond to poll requests. Unenthusiastic voters are more likely to not respond. We also know that Republicans voters are generally less likely to respond. In all likelihood, the voter isn't changing their minds, they just aren't responding to the poll surveys.
The Brad
-
...which would lead one to use the model results as toilet paper rather than election predictors.
Edit: I see you are only talking about polls rather than Silver's model. I meant to refer to Silver's model rather than polls in my original post, and if Silver is able to model away these flaws in polls, then his models would be a better predictor of voting behavior. Which gets back to my original musing about who these people are who change their minds to the extent this model indicates, and what is their perspective. I am not talking about the details of how models and polls differ, I am talking about why people change their minds to cause this difference in election results predictions.
-
I wish they'd call me up. I could really screw up their predictions. I'm voting for Eisenhower, and so is my 140 year old grammy.
I was only joking
-
They aren’t predictors, they are valuable tools in seeing where the race stands at this point. The momentum for Kamala has waned, particularly in the swing states. She’ll very likely win by millions in California, but a loss by 10,000 in Pennsylvania crushes her opportunities. And the model also tells us that if she is doing poorly in PA, she is likely doing worse in Michigan and Wisconsin than advertised by the polling.
-
They aren’t predictors, they are valuable tools in seeing where the race stands at this point. The momentum for Kamala has waned, particularly in the swing states. She’ll very likely win by millions in California, but a loss by 10,000 in Pennsylvania crushes her opportunities. And the model also tells us that if she is doing poorly in PA, she is likely doing worse in Michigan and Wisconsin than advertised by the polling.
@LuFins-Dad said in Still think it's too early...:
They aren’t predictors, they are valuable tools in seeing where the race stands at this point. The momentum for Kamala has waned, particularly in the swing states. She’ll very likely win by millions in California, but a loss by 10,000 in Pennsylvania crushes her opportunities. And the model also tells us that if she is doing poorly in PA, she is likely doing worse in Michigan and Wisconsin than advertised by the polling.
Ok thank you. None of this actually has anything to do with my original point about the people changing their minds, and their perspectives. And yes, to the extent the model is meaningful, it must indicate that people are changing their minds.
-
If it was the same 10000 people polled every time, then yes, it would indicate they are changing their minds. Since it’s not, it’s far more about voter enthusiasm.
-
If it was the same 10000 people polled every time, then yes, it would indicate they are changing their minds. Since it’s not, it’s far more about voter enthusiasm.
@LuFins-Dad said in Still think it's too early...:
If it was the same 10000 people polled every time, then yes, it would indicate they are changing their minds. Since it’s not, it’s far more about voter enthusiasm.
You are only giving reasons why the model is not fit for purpose. An enthusiastic voter's vote counts as much as an unenthusiastic voter's vote.
If enthusiasm causes a non-voter to vote, then that would be a change of mind I am talking about. The model has to model change in voting actions, to whatever extent it is meaningful in the way it is supposed to be meaningful.
-
@Horace said in Still think it's too early...:
@LuFins-Dad said in Still think it's too early...:
@Horace said in Still think it's too early...:
seems difficult to get a bead on the perspectives of whomever is changing their minds about preferred candidates to lead to these polling results changes. I don't think those sorts of people ever post online or discuss politics in general. Everybody who discusses politics will never change their mind about anything, at least in the short term.
First, the Silver forecast is not a poll, it's a model that does take into effect poll changes, but also other criteria and trends.
The model is only meaningful insofar as it maps to people's voting preferences changes. So to whatever extent this is a meaningful point, it's a point that would lead one to use the model results as toilet paper rather than a predictor of elections.
Second, I think that polling changes a lot based on enthusiasm. The enthusiasm for the democrat candidate is waning a little.
So if the model is accurate, this would give some insight into how people might change their minds. What, they read a bunch of enthusiastic media about a candidate and would be inclined to vote where they may not have voted before? Sure, possible, but again, these people don't discuss politics online or probably in person.
I think you're missing the single biggest challenges with the polls, and it has nothing to do with changing minds, it has to do with responses. Enthusiastic voters are more likely to respond to poll requests. Unenthusiastic voters are more likely to not respond. We also know that Republicans voters are generally less likely to respond. In all likelihood, the voter isn't changing their minds, they just aren't responding to the poll surveys.
@LuFins-Dad said in Still think it's too early...:
I think you're missing the single biggest challenges with the polls, and it has nothing to do with changing minds, it has to do with responses. Enthusiastic voters are more likely to respond to poll requests. Unenthusiastic voters are more likely to not respond. We also know that Republicans voters are generally less likely to respond. In all likelihood, the voter isn't changing their minds, they just aren't responding to the poll surveys.
Yep. I never answer them. Ever.
-
I wish they'd call me up. I could really screw up their predictions. I'm voting for Eisenhower, and so is my 140 year old grammy.
@Doctor-Phibes said in Still think it's too early...:
I'm voting for Eisenhower, and so is my 140 year old grammy.
A recently passed great American folk musician’s deceased grandfather, a carpenter, always voted for Eisenhower because Lincoln won the war.
