Forget national polls. National polls are wrong, always, because such polls are unscientific. I have explained that truth in these two articles:
- Well It is Election Time, Again, and Fake News USA Today Is Touting an Unscientific National Poll Claiming Trump is Behind, Again. When Will They Ever Learn?
- WHY EVERY PRESIDENTIAL ELECTION POLL FOR 2016 YOU SHALL SEE OR HEAR WILL BE WRONG
Yet, if you go by statewide polls, you are forced to believe that fast-slipping into senility, Joe Biden, not only is going to win the 2020 USA presidential election, but also he is going to pull off a win by one of the greatest margins of victory ever in the electoral college.
Today, I set up a clever spreadsheet that takes the average of statewide polls for any state that has polls reported to Real Clear Politics' Latest 2020 Presidential Election Polls. The spreadsheet assumes a clear winner where no polls exist, e.g., Trump takes Idaho, Biden takes Hawaii, based on well-accepted beliefs about how voters in particular states lean.
While my spreadsheet is flawless, I can not believe the state polls lack flaws. The pollsters must be over-sampling in the biggest cities like Milwaukee, Minneapolis, Detroit and Philadelphia while failing to sample rural and suburban voters in the states of Wisconsin, Minnesota, Michigan and Pennsylvania, respectively.
Statewide polls suffer from nearly the same flaws national polls. Pollsters who fail at nationwide polls do so because they fail to account for the electoral college. The US presidential election is not one election but many.
To do statewide polls properly requires dividing up a poll for any state into two polls, one sampling the urban people, the other sampling everyone else, i.e., the suburban and rural folk. Thus for any state, at least 752 people need to be polled from two different sample sets, taking 376 from each, to get a margin of error of plus or minus 5% to trust that 95 times out of 100 working from the same questionnaire will give alike results.
That no polling firm does this tells you their polls are merely a non-statistical collection of opinions of respondents that can not be projected to the universe it is supposed to represent.
President Trump eked out wins in Wisconsin, Pennsylvania and Michigan in 2016. Had pollsters completed proper, scientific surveys, they would have known this before November 8, 2016.
This picture says it all:
The pollsters failed to sample randomly and in sufficient amount from all of those rural and suburban counties from where 1.4 million voters came and voted for President Trump. They failed because they do not how to do scientific polling in spite of that being upon what they stake their firms' businesses and reputations.
Gab It
