Quote:
Originally Posted by scuba steve
adraco, can you point out two instances where the states have gone against the popular vote of that state? i understand at the macro level, i'm trying to find instances at the state level
|
Illoyal electors have never changed the election. Normally it's only one or at most handful of electors who, for some reason, swing their vote. Once have all electors of Pennsylvania gone against the outcome of their state election, but that was more than 100 years ago. The intersting part is that it has happened nine times since 1900. It's often used as some kind of statement, to put focus on something in particular.
Quote:
Originally Posted by sam_mrskincash
While you are at it do you understand how the news outlets are able to project winners while sometimes also saying that less than 1% of the vote has been reported? That boggles my mind.
|
Some states have a long history of voting for either party. They might even have a remarkably higher number of party members as registered voters. Then when they get the number of total voters, it's fairly easy to run a statistical model with a 90-95 % confidence interval and see that with this amount of voters, this amount of registered voters on each party side, combined with the voting history of this state, add in factors like "white males in rural areas tend to vote republican" and "women in urban areas tend to vote democratic" and then run that model a few times through your regular desktop computer, with the statistical program SPSS and you too would be able to predict the outcome with say some 90% certainty.
I believe that CNN at the closing of the voting booths in Ohio, had nine states marked on their map as "battleground states". Many other states were fairly easy to call either way, due to their voting history, demographic, and the surveys done leading up to election day as well as some surveys done directly with voters during the election day after people have cast their vote. The number you need, is often the total amount of voters who showed up, then you can apply a statistical curve of how they likely voted. With a high enough sample (million of voters) the mass will become pretty predictable.
Compare to Mr Skincash
I have no idea the size of your program, but let's assume you do a decent enough amount of sales every day and that your program have been running for a good number of years. That would give you enough data to be able to predict the average number of sales for a regular, normal Wednesday like today to a certain degree. With a big enough sample to draw conclusions from, one can be pretty certain about the behavior of the mass, people are not as spontaneous as they like to think.
