[EM] No. Condorcet and Hare do not share the same problem with computational complexity and process transparency.

[robert bristow-johnson]

> On 03/20/2024 3:27 PM EDT Michael Garman <michael.garman at rankthevote.us> wrote:
> 
> 
> Do “general trends” explain the discrepancy in levels of representation between IRV and non-IRV cities with similar characteristics?
> 

You need to learn how to do Bayesian inference.  RCV hasn't been around for long in the U.S., except for Cambridge MA.  All over the U.S. representation of POC has increased, not just some of the few places where RCV has existed for less than a decade.  Most scholars are crediting that to single-member districts in the South, which prevented corrupt legislators from drawing lines splitting the Black vote without the districts appearing gerrymandered.

But if I roll a pair of dice once or twice and I get two ones both times, do I conclude that the dice is weighted?

> IRV is compatible with risk limiting audits.

That doesn't say shit.  So is FPTP.

> Some people will always cry fraud when they lose, but IRV can be audited rigorously. 

Not like FPTP nor like Condorcet can be audited.  In Alaska in November 2022, they didn't announce results until the day before Thanksgiving.  No one knew who won until then.

There are no redundant data paths where other interested parties, namely the media and the competing campaigns, can get the raw numbers from the polling places and simply *add* those numbers up and see who wins.

This is something we have right now with FPTP and we lose it with IRV because IRV is not precinct summable.

Again, what if Georgia Secretary of State Brad Raffensperger was corrupt like apparently Jeffrey Clark is?  What prevents someone at the single-point-of-failure from surreptitiously padding (or subtracting) tallies of the vote in a close election to swing the outcome to a different candidate?  What prevents that, what keeps honest people honest, is that these tallies are already published on the night of the election at the polling places.  If they are changed significantly, people will get suspicious.  If they are changed enough that the outcome of the election is changed, then people will focus on the very cities and polling places where numbers are disparate and sealed ballot bags are opened.

Precinct Summability is necessary to do that.  Precinct Summability is a property we make use of, right now, with FPTP.  And we lose it with IRV.  And then IRV doesn't scale well when it goes statewide.  How do they centralize the ballot data securely and quickly enough that conspiracy theorists cannot call out as suspicious?  How do we find out who wins (or is expected to win) on election night in a statewide election?  We find out by adding up numbers that we learn that are published at each polling place on the evening of the election after the polls close.

Mr. Garman, I suggest you might consider doing some work as a poll worker or even as an elected polling official.  I have done both.

--

r b-j . _ . _ . _ . _ rbj at audioimagination.com

"Imagination is more important than knowledge."

.
.
.