Who is in the lead. What Regions/Districts are what majority. I mean if you can theoretically get real time data on voting machines. There is a lot you can do with that information. Everything from 'what days of the week will democrats vote' to 'predicting who will win the election'. If names are stored with a timestamp, they could potentially know exactly who you vote for and the time you do it. I think that part is far fetched. But it is an idea. All the information could be used for skewing future elections. For that parties favor.
A) Phonebanking redistributing your limited resources to GOTV on election day.
B) Responding to incidents with legal council on the ground to contest potential fraud ASAP
So how does fraud detection work? Why do you think they sink millions and millions of dollars into polling. Long before election day both sides have a pretty accurate statistical model and break down of how each district will vote and the thresholds of deviation from that model which will set off "Fraud!" bells.
This is also why the NSA probably doesn't manipulate Federal Elections.
This is also why no one would bother to take the risk to get real time results. Its not valuable enough information. Campaign technology staffs have been working (I know, I helped) and will continue to work on ways to collect exit polls on the spot via hand helds or other means as close to real time as possible... just to get a statistically significant indication of whether what's happening on the ground deviates from the model far enough to allocate resources.
Comments
Who is in the lead. What Regions/Districts are what majority. I mean if you can theoretically get real time data on voting machines. There is a lot you can do with that information. Everything from 'what days of the week will democrats vote' to 'predicting who will win the election'. If names are stored with a timestamp, they could potentially know exactly who you vote for and the time you do it. I think that part is far fetched. But it is an idea. All the information could be used for skewing future elections. For that parties favor.
Real time data is useful for 2 reasons:
A) Phonebanking redistributing your limited resources to GOTV on election day. B) Responding to incidents with legal council on the ground to contest potential fraud ASAP
So how does fraud detection work? Why do you think they sink millions and millions of dollars into polling. Long before election day both sides have a pretty accurate statistical model and break down of how each district will vote and the thresholds of deviation from that model which will set off "Fraud!" bells.
This is also why the NSA probably doesn't manipulate Federal Elections.
This is also why no one would bother to take the risk to get real time results. Its not valuable enough information. Campaign technology staffs have been working (I know, I helped) and will continue to work on ways to collect exit polls on the spot via hand helds or other means as close to real time as possible... just to get a statistically significant indication of whether what's happening on the ground deviates from the model far enough to allocate resources.