
16:06
QUESTION: How are you doing?

16:37
Well

30:16
QUESTION: Did he say whether all states are assumed to have the same number of voters?

30:28
thanks

37:38
QUESTION: Can you discuss again, briefly, the considerations for setting a threshold T for the states vs the threshold for the national vote?

40:38
QUESTION: what happens if you assume that the distribution in states is modelled as a Gaussian around the national distribution between the parties?

41:01
Q

41:20
QUESTION: Can we have the precise references to the main other papers mentioned in the talk given in the chat? Thanks!

43:34
fraud con

43:43
fraud conditional on polls?

45:40
Yes

47:08
https://mikegeruso.com/

47:18
thanks

47:47
Thank you - the spelling of the first author I first tried with "Spears" led to unrelated celebrity websites ...

01:05:15
What is this green thing? :)

01:19:23
QUESTION: If I was using a variable threshold (e.g., vote for all candidates with utility above the mean utility), would you detect that? Or classify that as something else?

01:19:53
If I *were* using . . .

01:28:00
QUESTION: Connection to real-life (usually municipal) elections that do use a form of AV, but where Vote-for-k is the maximum you can do.

01:29:00
QUESTION: does the missing vote model entail that the voters have information about nearly all of the votes by others?

01:32:15
Peter, we have polls, at least here. We have info on how people say they are going to vote.

01:34:57
THANKS

01:35:01
Thanks !

01:35:03
Thank you!!

01:35:06
thanks!!!