Ties: the only time your vote really matters (but you should still vote)

[parker friedland]

Your vote always matters in the sense that your vote is still valuable (politicians will still try to "earn" it), but this is the one time that your vote matters matters:
https://www.youtube.com/watch?v=VM4A6Fxw0Io

[parker friedland]

Correction:
Ties: the only time your vote really matters (see 1st and 5th footnotes) (but you should still vote)

1st Footnote: except for when there are an odd number of voters, in which case your vote has a 50% chance of counting when ever there is a semi tie (see 2nd, 3rd and 4rd footnotes)
2nd Footnote: by semi tie, I mean an election that could become a tie if just one voter forgot to show up or remembered to show up.
3rd Footnote: or an odd number of approvals between the top two front runners if approval voting is used (and don't even get me started on N way ties or other alternative voting methods)
4th Footnote: your vote would not have a 50% chance of counting but a 100% chance or 0% chance of counting depending on who you voted for.

5th Footnote: your vote always matters in the sense that your vote is still valuable (politicians will still try to "earn" it), but this is just the time when it matters matters.

[Warren D Smith]

I am somewhat befuddled by the 5 corrections in a row...

The only time your vote directly matters is when it creates or breaks a tie.
Elections that are tied, or 1-off being tied, are very rare.
However (!), one allegedly just happened:

In the 94th Virginia delegate race of 2016,
Democrat Shelly Simonds defeated Republican David Yancey by a single vote
11608 to 11607.
But then a 3 judge court ruled it actually was a tie: 11608-11608.
This was because Yancey's team
found a single extra ballot which had not been counted
because it had been judged "ambiguous" due to voter filling in both
bubbles. But the voter also had slashed a line thru the Simonds bubble.
The 3 judges believed this meant the voter had intended to vote Yancey.

To break the tie, a random choice (e.g. coin toss) will be made!
Even more amazingly, if Simonds wins it, that will create a party tie
in the legislature, while if Yancey wins it the Republicans
will have majority control of the legislature.

Perhaps it should have been argued that 1 voter meant Yancey with
probability 2/3 and meant Simonds with probability 1/3,
in which case Simonds should have won by 2/3 of a vote!?
I mean, I do not think it was 100% clear the voter meant Yancey, and
any percentage below 100%, such as 99%, "should"
have been regarded as a win for Simonds... right?

Anyhow, usually whenever it is a tie or near-tie, the amount of
voter errors and other screwups is large. I do not believe in
claimed accuracies better than 0.1%.

https://slate.com/news-and-politics/2017/12/virginia-recount-ends-in-tie-to-be-determined-by-drawing-lots.html
http://www.cnn.com/2017/12/19/politics/virginia-house-of-delegates/index.html


--
Warren D. Smith
http://RangeVoting.org <-- add your endorsement (by clicking
"endorse" as 1st step)

[Warren D Smith]

here is a photo of the disputed ballot -
judge for yourself!

https://twitter.com/virginianpilot/status/943589969603555328/photo/1