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.
Lol sometimes it seems like they call states when 1% of the votes are counted.
That's because, in such a large sample, if results are truly counted randomly it's possible to accurately predict a result to a less than 1% margain of error.
For example: if 52,000 people out of 100,000 (lets say this is 1% of the voting population) vote for candidate A it is statistically almost certain that between 51.5% and 52.5% of the the total vote will be for candidate A makjng it possible to accurately "call" a state at such an early stage.
That's because, in such a large sample, if results are truly counted randomly it's possible to accurately predict a result to a less than 1% margain of error.
For example: if 52,000 people out of 100,000 (lets say this is 1% of the voting population) vote for candidate A it is statistically almost certain that between 51.5% and 52.5% of the the total vote will be for candidate A makjng it possible to accurately "call" a state at such an early stage.
I'm not sure how the vote counts go in particular, but do they count the votes by district? In that case, if first they count all the votes from a district that is mostly all-white Appalachian towns, it's going to give a prediction that is quite different than the rest of New York State
That's because, in such a large sample, if results are truly counted randomly it's possible to accurately predict a result to a less than 1% margain of error.
For example: if 52,000 people out of 100,000 (lets say this is 1% of the voting population) vote for candidate A it is statistically almost certain that between 51.5% and 52.5% of the the total vote will be for candidate A makjng it possible to accurately "call" a state at such an early stage.
But does that factor in where the votes are coming from? For example, votes coming in from a rural counties vs. urban counties can be misleading.
That's why they should get rid of the whole electoral college stupidity. It makes people think their vote doesn't matter when it should....
Yes. It's an ancient, dumbass system where people vote for one candidate and the other candidate wins. It's hard to call the us a democracy when the candidate that most people vote for loses the election.
I'm not sure how the vote counts go in particular, but do they count the votes by district? In that case, if first they count all the votes from a district that is mostly all-white Appalachian towns, it's going to give a prediction that is quite different than the rest of New York State
Sometimes some districts (usually the smaller rural ones) get counted faster but the algorithems they use to calculate the result of the election take these factors into account.
Here's how it works:
District A is a rural, republican district with a population of 100'000
So far 2% of district A has voted resulting in 1424 (71%) votes for Candidate R and 576 (29%) votes for candidate D therefore the computer predicts that this county will net Candidate R 71,200 votes and Candidate D 28,800 votes.
District B is a Urban, democratic district with a population of 10,000,000
So far 2% of district A has voted resulting in 120,000 (60%) votes for Candidate D and 80,000 (40%) votes for candidate R therefore the computer predicts that this county will net Candidate D 120,000*50= 6,000,000 votes and Candidate R 4,000,000 votes.
Of course the this is just the basic idea. Projection algorithms used by major networks factor in many more complex nuanced factors as well (standard deviation, voting trends, confidence of margain of error, etc...)
I'm not sure how the vote counts go in particular, but do they count the votes by district? In that case, if first they count all the votes from a district that is mostly all-white Appalachian towns, it's going to give a prediction that is quite different than the rest of New York State
if you assume shfits are mostly correlated (most places swing similarly) you can get a good hint. Say, in previous election Republicans averaged about 55% in Appalachian towns, and the counted result show 65%. That suggests the rest of the state will be about 10 points more Republican. Once, you have a greater diversity of towns, you can get more accurately. Like, most Appalachian towns might shift similarly; but white Long Island ones might do some different, urban black ones something else. So a few each could be enough to have a decent prediction on the entire state.
I remember on the TV when they were counting Michigan and only a few rural ones were counted but they were much redder than normal it was obvious this wasn't going well for Clinton.
Quote:
Originally Posted by nei
Russian immigrant neighborhood in Brooklyn loves Trump, dislike Putin. I assumed they'd like both.
on the other hand, if 1% of the votes counted were just from Brighton Beach and similar; probably won't be informative enough as they're kinda unusual.
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.
