2 comments

[ 1.8 ms ] story [ 15.4 ms ] thread
On the one hand, I'm kinda glad an entirely new group of people are learning that maybe we shouldn't trust Diebold voting machines. At least these all have a paper trail...

That said, this analysis seems... contravertable. The basic argument seems to involve multiplying percentages (truncated or rounded to three digits) by the total number and observing that the amounts attributable to each candidate isn't monotonic. A search doesn't turn up any mention of rounding or truncation or digits or significant figures, and I stopped looking deeper.