Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
I think one could argue that the enthusiasm against Hillary was a decent part of what propelled Trump to victory in 2016.
As for Silver, his site is based on a mathematical model that is simply taking inputs and spitting out outputs. He could die tomorrow and the model would continue to run. He’s not editorializing inside the model.
If that were the case we wouldn't be seeing thousands attending Trump rallies. His base is stronger today than in 2016. Trump is more popular than ever.
To be sure H. Clinton wasn't an exciting candidate but is Biden no better. Why would someone who doesn't like either candidate bother to vote? Because they hate Trump more than Biden? No. They will sit 2020 out or leave the top spot unmarked in protest like in 2016.
Silver? Nate analyzes the data. My point is he knows this data esp. in the battleground states isn't necessarily anymore reliable in 2020 than it was in 2016. But his readers don't want to hear this.
I find it incredibly sad that you just read an article about how much work pollsters are doing to be accurate, then somehow used it to dismiss polls as propaganda. Talk about confirmation bias. I mean if they are just propaganda tools, why do all that work? This is quite the vast conspiracy you’re floating. Do you have any idea how hard it would be to keep something like that secret?
Are polls perfect? Of course not. There are countless people putting in countless hours in the name of trying to be more accurate. They will never be perfect, but calling them out as propagandists is so incredibly cynical.
Let’s keep in mind this bit from the reading:
Your choice to be an ostrich.
You're leaving out the fact that the polls didn't narrow until right before the election. The pollsters were covering their asses.
Prior to that, they were frequently showing Hillary up by double digits.
If that were the case we wouldn't be seeing thousands attending Trump rallies. His base is stronger today than in 2016. Trump is more popular than ever.
To be sure H. Clinton wasn't an exciting candidate but is Biden no better. Why would someone who doesn't like either candidate bother to vote? Because they hate Trump more than Biden? No. They will sit 2020 out or leave the top spot unmarked in protest like in 2016.
Silver? Nate analyzes the data. My point is he knows this data esp. in the battleground states isn't necessarily anymore reliable in 2020 than it was in 2016. But his readers don't want to hear this.
Biden is not better in your very subjective opinion. It’s not a fact. Everything you posted above is your opinion and not supported by data. The data, in fact, seem to show that Biden has lower negatives than Hillary. He’s hardly as polarizing a figure. We will see how that plays out.
Silver and his team built a model that provides a probability based on the inputs of multiple polls. They are not tweaking the model during the election. That wouldn’t be a reasonable thing to do. Data from polls are input and the model spits a probability out. Like I said, Silver and his entire team could die tomorrow and nothing would change in the way his model works. Poll data would continue to go in and a probability would continue to be generated. It seems like you don’t understand how statistical modeling works.
Quote:
Originally Posted by RhodyRepub
You're leaving out the fact that the polls didn't narrow until right before the election. The pollsters were covering their asses.
Prior to that, they were frequently showing Hillary up by double digits.
Are you speaking through the lens of a single election in 2016? That’s just not how it works.
Pollsters don’t start fudging their results as election gets close. Did you look at the volatility during the leading up to Election Day in 2016? It was all over the place.
Like I said earlier, the polling error in 2008 and 2012 showed the Republican candidates performing better than the actual results. This isn’t a vast left wing conspiracy. It’s just sampling error.
Biden is not better in your very subjective opinion. It’s not a fact. Everything you posted above is your opinion and not supported by data. The data, in fact, seem to show that Biden has lower negatives than Hillary. He’s hardly as polarizing a figure. We will see how that plays out.
Silver and his team built a model that provides a probability based on the inputs of multiple polls. They are not tweaking the model during the election. That wouldn’t be a reasonable thing to do. Data from polls are input and the model spits a probability out. Like I said, Silver and his entire team could die tomorrow and nothing would change in the way his model works. Poll data would continue to go in and a probability would continue to be generated. It seems like you don’t understand how statistical modeling works.
Are you speaking through the lens of a single election in 2016? That’s just not how it works.
Pollsters don’t start fudging their results as election gets close. Did you look at the volatility during the leading up to Election Day in 2016? It was all over the place.
Like I said earlier, the polling error in 2008 and 2012 showed the Republican candidates performing better than the actual results. This isn’t a vast left wing conspiracy. It’s just sampling error.
Not true. I provided a link earlier which points out the polling data Silver receives with regards to the battleground states is still consider by some in the industry to be off. If the data isn't accurate his analysis isn't either. Again my point is Silver knows the battleground numbers are *skewed but his readers come for a reassurance of a Biden win so that's what he gives them. https://www.politico.com/news/2020/0...s-biden-324210
*Quinnipiac has Biden up by 13% in Penn. Not a chance that's even remotely close. Yet, it's still in the mix which help throw off the averages.
Not true. I provided a link earlier which points out the polling data Silver receives with regards to the battleground states is still consider by some in the industry to be off. If the data isn't accurate his analysis isn't either. Again my point is Silver knows the battleground numbers are skewed (for example not a chance Biden is up 13% in Penn.) but his readers come for a reassurance of a Biden win so that's what he gives them. https://www.politico.com/news/2020/0...s-biden-324210
This is simply a lack of understanding about how his model works. His model does not show Biden up 13% in Penn. It has Biden up 6% in Penn. His model will take the outlier that you are referencing and weight it based on several factors. I have no idea what you’re talking about when you say he is trying to provide people with reassurance that Biden will win. He calls out the risks to the numbers his model is providing all the time. It almost seems like he is going out of his way to tell people that despite what the model is showing, there is always unknown risk. The article I posted to start this thread goes into detail on where some of that risk could come from.
His model is based on data inputs. He doesn’t get to change it based on a narrative. The only thing that can change the outputs of his model is if the data the pollsters are delivering changes. He weights pollsters based on their historical accuracy, not their propensity to deliver positive news to democrats.
When was the last time you took a statistics class?
I know the whole “the world is a vast conspiracy against Trump” narrative is fun for you guys to play, but at a certain point critical thinking needs to be part of the equation.
This is simply a lack of understanding about how his model works. His model does not show Biden up 13% in Penn. It has Biden up 6% in Penn. His model will take the outlier that you are referencing and weight it based on several factors. I have no idea what you’re talking about when you say he is trying to provide people with reassurance that Biden will win. He calls out the risks to the numbers his model is providing all the time. It almost seems like he is going out of his way to tell people that despite what the model is showing, there is always unknown risk. The article I posted to start this thread goes into detail on where some of that risk could come from.
His model is based on data inputs. He doesn’t get to change it based on a narrative. The only thing that can change the outputs of his model is if the data the pollsters are delivering changes. He weights pollsters based on their historical accuracy, not their propensity to deliver positive news to democrats.
When was the last time you took a statistics class?
I know the whole “the world is a vast conspiracy against Trump” narrative is fun for you guys to play, but at a certain point critical thinking needs to be part of the equation.
Silver's model factors in outliners such as the 13% which skew the overall averages. As far as reassurance for his readers. I'm talking about his analysis of the data. For the entire 2020 election his analysis has predicted a Biden victory. Just last week he had Trump with what...a 15% chance of winning?
Yet, here we are just two weeks out and the polls are tightening Why? What changed? Nothing changed. The fact is Biden has never been up by as much as the battleground state polls suggested. And my feeling is Silver has always known this.
So... what I predict is we're going to see a repeat of 2016 where his readers scream after the election that 538 screwed up again. Silver will say the Trump win was within the margin of error which will be correct.
However, that won't change the fact his analysis of the data regarding Trump's chances from the start of the Rep. primaries in 2016 through the 2020 election have been completely wrong.
This is simply a lack of understanding about how his model works. His model does not show Biden up 13% in Penn. It has Biden up 6% in Penn. His model will take the outlier that you are referencing and weight it based on several factors. I have no idea what you’re talking about when you say he is trying to provide people with reassurance that Biden will win. He calls out the risks to the numbers his model is providing all the time. It almost seems like he is going out of his way to tell people that despite what the model is showing, there is always unknown risk. The article I posted to start this thread goes into detail on where some of that risk could come from.
His model is based on data inputs. He doesn’t get to change it based on a narrative. The only thing that can change the outputs of his model is if the data the pollsters are delivering changes. He weights pollsters based on their historical accuracy, not their propensity to deliver positive news to democrats.
When was the last time you took a statistics class?
I know the whole “the world is a vast conspiracy against Trump” narrative is fun for you guys to play, but at a certain point critical thinking needs to be part of the equation.
Right which is why the polls have suddenly tightened. The election is right around the corner. Doesn't matter if they've been predicting a Biden win since Jan 1 of 2020.
I read that article the other day when it appeared on my firefox home page but didn't post it considering the source, and polling evaluation is a waste of time. Voting day counts. Nothing else.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
