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.
Polls are often wrong, and that may very well be the case here. I think polls are better for showing a campaign where the swing states are so that they can know where to spend their resources. Why spend lots of resources in deep red Tennessee or deep blue New York when you could spend them in Pennsylvania?
Also, as I've said elsewhere, I think using the enthusiasm of Biden's supporters as a barometer for who will win is missing why many of those voting for him are doing it. Many will be voting for him not because they are enthusiastic supporters of him but because he has the best realistic chance to beat Trump. Thus, these folks aren't going to rallies and waving Biden flags, they just want Trump out. But their vote counts just the same.
This is exactly what I just said in another thread.
Nobody is denying Trump's base is deeply enthused by him, given the fact that they don't care about him catching COVID or cheating on his taxes.
Joe Biden was certainly not my first choice. I haven't donated a cent to his campaign, nor have I attended any of his virtual rallies, but that doesn't mean I voted for Trump. LOL
Ah yes, Dinesh...the guy who graduated with a useless English degree because he struggled mightily with college math is now an expert on statistical analysis.
Quote:
Originally Posted by bu2
Anonymous internet poster who doesn't understand statistics, sampling or test design? I'd go with the felon. But I didn't listen to his video. I don't know why he claims they are off.
I took undergraduate classes in polling statistics and graduate classes in regression analysis at an Ivy League business school. Go ahead and listen to the felon with an English degree.
You just cannot trust any poll. They don't fully advertise or wave their methods of conducting the polls.
Their "nationwide" polling could mean a different "nationwide" than most common people are thinking. Their nationwide definition could include polling in California, Oregon, Maryland, New Jersey and add in Pennsylvania as the swing state, to get some numbers up a bit for Trump.
It's all results for the audience (us) and the process is anyone's guess. Some divulge the "process" but I would take it with a grain of salt.
All I remember was "hillary has a 98% chance of winning"
if the polls are wrong again this time, I think it's the end of polling as we know it
The 98% of winning was in relation to the popular vote which she DID win. So the polls were correct.
What the polls failed to do in 2016 was account for voting blocks that weren't traditionally included in polls, specifically voters without a college degree. In 2018 and currently, pollsters are making a better a better effort to include them to make sure that their polls are more accurate. This is why the 2018 were accurate.
In addition, the polls in 2016 did not have enough time to take into account a lot of the last minute surprises in October in many of the states that flipped. If you disagree, please show me a state poll that was taken within two weeks of the election, where Hillary was up by more that the Margin of Error and lost.
How could Biden be so far ahead, when no one shows up to his events or is at all excited by his candidacy? I see some signs on lawns, but that's about it. No one can even articulate well what they are voting for in Biden.
Dinesh d'souza indicated that the polls are inaccurate and also do not account for Trump voters who are less vocal in their support. Does this alone explain a 10 point lead?
It might just be that one has nothing to do with the other. The same way that magic rock I keep in my pocket might keep the lions away. Or maybe it is just coincidence. The same way that is seems like my favorite sports team only loses when I watch them. However that is clearly superstition. If I had that sort of power, I'm sure the owners would pay me to not watch them.
While it might seem to make sense that more rallies and more signs equals more support, it could just be that a subset of Trump supporters are more vocal in their support than Biden supporters. But, there are millions of Trump and Biden supporters that have no interest in going to rallies. Frankly, I'm not interested in what Biden has to say on the campaign trail. I'd rather see his interviews and debates where he doesn't get to pick the questions himself. And the same goes for Trump. Campaign rallies on both sides are merely propaganda. There's nothing to learn from that.
You just cannot trust any poll. They don't fully advertise or wave their methods of conducting the polls.
Their "nationwide" polling could mean a different "nationwide" than most common people are thinking. Their nationwide definition could include polling in California, Oregon, Maryland, New Jersey and add in Pennsylvania as the swing state, to get some numbers up a bit for Trump.
It's all results for the audience (us) and the process is anyone's guess. Some divulge the "process" but I would take it with a grain of salt.
It's just more than having them reveal their methods but in understanding why their methods could be skewing their results. For example, there is a poll that only does calling during the evenings which can be more biased against those who work in the evening or overnight. Another asks for the youngest voter in the household which may miss older voters who are more likely to actually vote. Then there is the way that they define likely voter can differ, some may do it based on a self report and others can do it based on voting records. Doing a survey by a live interviewer vs a interactive automatic voice can produce differences. RDD vs cell samples. Web responses vs phone. There are just so many variables. That is why looking at one poll is dangerous but looking at the average of all the polls is a better sign of what the reality is.
Because 1. 99% of pollsters have a leftist bias, 2. 99% of polls are underpolling men and republicans and are polling urban cities not the entire state evenly. So its of course going to skew the results. They will once again be proven to be liars on November 3rd.
In 2 of the past 3 elections, polls had a right bias. But don't let the facts get in the way of BS.
Also, it would be one thing if the polls were mixed and moving erratically (they sure seemed to during the 2016 election, and I was worried the closer it got to the election), but they aren't. They've been consistent and the lead Biden has over Trump has continued to expand, not shrink.
Exactly, they use it to determine where to spend ad money as well. Every campaign relies on polling data, it's not exact, but it's usually pretty close.
This idea that polls are totally fake and used to form some kind of narrative that doesn't exist is completely absurd. It's just math.
If all the polls had Trump up 10 points, every single Trump supporter would be singing their praises. This is not about the polls, it's about their inability to view anything objectively regarding Trump.
Also, it would be one thing if the polls were mixed and moving erratically (they sure seemed to during the 2016 election, and I was worried the closer it got to the election), but they aren't. They've been consistent and the lead Biden has over Trump has continued to expand, not shrink.
They are just trying to suppress Trump voters from going to the polls, I guess they really think that's going to work, its quite obvious the media and pollsters don't know Trump voters very well. NOTHING will stop us from voting for Trump on election day. I don't care if a poll showed Biden up 90% to 10%!
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.
