But to respond to your post, I don't actually think that based on the premise of the article and the accusation that exceptional foresight is really necessary on the part of potential hackers/manipulators like you're suggesting. There is ample historical data for virtually every county in the United States as to which way the vote will go, and once the dataset is loaded, it would be pretty trivial to figure out which districts need someone to press on the scale a little in favor of one side or another. State-wide, certain states have swung back and forth from red to blue, but county to county, the changes are much more predictable. Combine this that you're looking at a handful of states and not nation wide, and it seems a bit more possible.
I want to reiterate I'm not making an accusation, but assuming there was a mechanism that "hackers" could use to influence the results, the scope and foresight necessary is easily attainable in common datasets online, with historical data going back pretty far.
Any other way, it's kids crying about the rules after losing Monopoly.
I think votes should be recorded in a way that any set of adults could easily check for themselves with minimal tools.
https://medium.com/@jhalderm/want-to-know-if-the-election-wa...
Apparently WI does use paper ballots, but they are not checked at all unless a recount is called. Which is ridiculous.
But if they could manipulate the results in some areas, and make it obvious they were manipulated, then the whole thing becomes suspect.
Hypothetically, let's say that you're a foreign state and you want to work together with candidate A to pull off a win. You might:
1. Utilize your existing intelligence and hacking resources to expose one or more secrets about candidate B to take advantage of their secretive nature and undermine what would seemingly be their greatest strength, their experience, by leaking info about mishandling classified information.
2. Manipulate* the public in the weeks prior to the election when the undecided are making up their minds by having the FBI director indicate that they had found "more" emails they would have to go through about candidate B that could be a smoking gun of mishandling classified information. Then afterwards have the director state that the emails were the same ones they had already analyzed, such that this would seem more legitimate.
I would think in such a situation, you could reasonably ensure you have a close lead, even with two unfavorable candidates.
In such as case, possibly a 7% tweak could be enough, because you've already done much more than 7% via other measures. Maybe you don't even need to do anything more.
* - I wouldn't have thought that a late game stunt by the FBI director would have been that big of a factor, but anecdotally know of some that were planning to vote for Hillary, but changed their minds and voted for Trump in the end because Hillary came off as too shady.
The point here is not that this is what actually happened, but that this could happen. Foreign states can influence elections in a lot of ways.
And, I think we should have statisticians get involved in trying to restore faith in democracy via giving more credence to the election. It's a good exercise.
Today, we know the vote totals, so we can look back and say "wow, it's a small effect, but it would have been enough to tip things the other way."
But hackers working on Nov 8 would have had no idea how the vote was going in real time. No one does, especially in a state that uses paper ballots for some counties. The vote count only becomes apparent much later.
So how would the hackers, who were obviously highly motivated to achieve their desired result, decide that a 7% tweak in a few counties would be enough to get what they wanted? When they wanted it so badly that they're willing to commit felonies to get it?