muon2
Moderator
Atlas Icon
![*](https://talkelections.org/FORUM/IMG/starmod.gif) ![*](https://talkelections.org/FORUM/IMG/starmod.gif) ![*](https://talkelections.org/FORUM/IMG/starmod.gif) ![*](https://talkelections.org/FORUM/IMG/starmod.gif) ![*](https://talkelections.org/FORUM/IMG/starmod.gif)
Posts: 16,827
![](./avatars/Republican/R_IL.gif)
|
![](https://talkelections.org/FORUM/IMG/post/xx.gif) |
« on: December 02, 2017, 07:55:06 AM » |
|
If a vote is a tie, unless there only a handful of voters, a recount is statistically unlikely to also be a tie. Voting is a measurement of the popular will, and like any measurement there is uncertainty in the measurement. In science you would provide error bars to say that the actual measurement is within some range centered on the reported result.
We have this idea that there is a single correct value of the vote, but when there are millions of voters at hundreds of polling places, there will be measurable deviation in the result based on who and what counts it. No one counter is guaranteed to be correct so the result will change. We just use the law to determine which count shall be deemed final, though it it may not be closest to reality.
|