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 can explain to you at length why flipping two heads in a row is only 25% likely to happen. Then you might do it. Cool! Then I could explain it to you again and maybe you'll do it again. That doesn't make the statistics wrong.
Right now, 538 says Trump has a 13% chance of winning. That's not zero. It means it's unlikely. (At this moment, still over a week from election day.) It might change as the day gets closer, or it might not. All of his swing state forecasts are pretty close, ranging from 50-50 to around 70-30 in either direction. They might tighten up or they might not. But all they are right now are snapshots in time. We'll have to see what happens on election day itself.
So what? Garbage in/garbage out. If the data is skewed or there are variables the 538 model can't account for...what's the point.
I have followed Rasmussen.
*They poll likely voters all during the cycle. This close to the election everyone claims to be talking with likely voters; Rasmussen never varies.
*They poll Presidential approval every day. They are the only one doing that and average each day's results with the previous 2. Then at the end of the month, they report the full 15,000 interviews and compare to previous months.
*They got 2016 right. They don't actually predict electoral votes, but they got the popular vote right.
*They are excluded from 538 and RCP, even though they were accurate in 2016. Since RCP and 538 excluded Rasmussen and include polls who interview adults and registered voters, I am suspicious of RCP and 538.
They show Biden up by three points, nationwide. Two weeks ago Biden was up by 12. They show Trump with 47% of the Black vote. That is nearly unbelievable, but if true or nearly true Trump will take enough votes from large metro areas in PA, MI, MN, and WI to easily retain the Oval Office.
Seriously doubt that 47% but Trump will do much better with the Black (and Hispanic) vote in 2020.
Equally bad for Biden is the Black turnout will likely mirror Clinton.
Not a chance he gets anywhere near the turnout Obama saw in 2008/2012.
What I am saying is the polling data used in 2020 like the data in 2016 isn't accurate. In fact I'd argue it's *off even more today.
But you have no evidence for that. Absolutely none. The only polls that were significantly wrong in 2016 were swing state polls in midwestern states that weren't expected to be real swing states. Because they were not thought of as swing states, the only polling attention they got was from small, local pollsters like the local news station or newspaper. This time around, they are all being polled by big pollsters with good track records.
I have followed Rasmussen.
*They poll likely voters all during the cycle. This close to the election everyone claims to be talking with likely voters; Rasmussen never varies.
*They poll Presidential approval every day. They are the only one doing that and average each day's results with the previous 2. Then at the end of the month, they report the full 15,000 interviews and compare to previous months.
*They got 2016 right. They don't actually predict electoral votes, but they got the popular vote right.
*They are excluded from 538 and RCP, even though they were accurate in 2016. Since RCP and 538 excluded Rasmussen and include polls who interview adults and registered voters, I am suspicious of RCP and 538.
They show Biden up by three points, nationwide. Two weeks ago Biden was up by 12.
They show Trump with 47% of the Black vote. That is nearly unbelievable, but if true or nearly true Trump will take enough votes from large metro areas in PA, MI, MN, and WI to easily retain the Oval Office.
Rasmussen is a garbage poll, and they are widely regarded as such by those in the industry. The very fact that they have moved from Biden +12 to Biden +3 in two weeks is proof of that. There has certainly not been that much actual movement in the race.
The same thing happened with them in 2016. They had Trump +7 I think in their third-to-last poll. The last poll has Hillary +1 or so.
If you are interpreting Trump as "having NO chance at all of being re-elected," you are misunderstanding basic statistics.
This is how so many people walked away from 2016 thinking there was some fundamental failure. FiveThirtyEight gave Trump a 30% chance of winning. Is the fact that he won proof that they were wrong? Of course not. Things that have a 30% chance of happening happen all the time.
It depends on what polls were wrong, of course. In 2016, the national RCP average was within 1.1% of the final outcome, so there was no major polling error there. Rather, the polling error happened in a few specific swing states like MI, PA and WI that weren't expected to be close. But the pollsters working those states were small, local pollsters. This time around, major outfits like CBS News and Reuters are working those states.
If FiveThirtyEight gives Trump a 30% chance of winning again this year, and he wins, they are ineffective.
I'll give them the benefit of the doubt in 2016, but not twice in a row.
I hope the polls are warong again, Trump wins, and most of them go out of business because nobody believes them anymore. Frank Luntz agrees with me...the industry will implode...massive downsizing and dramatic changes in how they gather opinions. Polling by phone and getting <5% resposse rates is scewing results way too much.
Almost nobody answers calls from unknown numbers anymore due to the Billions of spam/robocalls we've all received. Think about what kind of person DOES answer a call from an unknown number on their cell phone.
If FiveThirtyEight gives Trump a 30% chance of winning again this year, and he wins, they are ineffective.
I'll give them the benefit of the doubt in 2016, but not twice in a row.
I hope the polls are warong again, Trump wins, and most of them go out of business because nobody believes them anymore. Frank Luntz agrees with me...the industry will implode...massive downsizing and dramatic changes in how they gather opinions. Polling by phone and getting <5% resposse rates is scewing results way too much.
Almost nobody answers calls from unknown numbers anymore due to the Billions of spam/robocalls we've all received. Think about what kind of person DOES answer a call from an unknown number on their cell phone.
What type person is that? How do they vote?
If you general hypothesis were correct, there would have been major polling errors from quality polling outfits in 2016. There weren't. The RCP final average was only off by 1.1%.
If FiveThirtyEight gives Trump a 30% chance of winning again this year, and he wins, they are ineffective.
I'll give them the benefit of the doubt in 2016, but not twice in a row.
I hope the polls are warong again, Trump wins, and most of them go out of business because nobody believes them anymore. Frank Luntz agrees with me...the industry will implode...massive downsizing and dramatic changes in how they gather opinions. Polling by phone and getting <5% resposse rates is scewing results way too much.
Almost nobody answers calls from unknown numbers anymore due to the Billions of spam/robocalls we've all received. Think about what kind of person DOES answer a call from an unknown number on their cell phone.
Every poll on Earth has Biden crushing Trump like a bug.
Trump has NO chance at all of being re-elected, according to the polls.
So, if Trump wins, are you willing to finally toss the pollsters in the trash?
You asked the wrong question.
Here's the right question: Has any of the polls declared Trump (or Biden) to be the winner?
The answer is no.
And no legitimate pollster would do such a stupid thing. Until an election is decided, all polls will only give some statistical information as to how a race is going. It's up to those who real the poll to guess who is a ahead or is behind.
Now people who make a living giving their opinions to others can make the claim Trump is losing, they can, because they're selling their thoughts, not facts.
So if they're wrong, that's on them, not the pollsters. Put the blame on where it belongs.
Throwing the pollsters in the trash is like throwing your calculator in the trash because you wrote down the numbers wrong.
You asked the wrong question.
Here's the right question: Has any of the polls declared Trump (or Biden) to be the winner?
The answer is no.
And no legitimate pollster would do such a stupid thing. Until an election is decided, all polls will only give some statistical information as to how a race is going. It's up to those who real the poll to guess who is a ahead or is behind.
Now people who make a living giving their opinions to others can make the claim Trump is losing, they can, because they're selling their thoughts, not facts.
So if they're wrong, that's on them, not the pollsters. Put the blame on where it belongs.
Throwing the pollsters in the trash is like throwing your calculator in the trash because you wrote down the numbers wrong.
Or maybe the calculator (538 model) is working but the batteries (data) are bad and you can't find the right batteries that fit.
If that's the case you might need in a new calculator.
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.
