[EM] D2MAC can be much more efficient than Range Voting (corrected)

[Abd ul-Rahman Lomax]

At 04:55 AM 3/8/2007, Scott Ritchie wrote:
>But let us not forget that even random candidate can get better results
>than any method some of the time.

This would depend on the goal of the election. As noted previously, 
if, for example, the goal of the election, or more specifically of a 
series of elections, is to spread out "victory" such that all groups 
have, over time, a "turn at bat," then random candidate could do 
better than methods not designed to do this. But the process might be 
even better if it were deterministic.

For example, it was proposed that history be used to redistribute 
victories. If the first victory were awarded by random ballot, then 
subsequent victories would be awarded, it would seem, less and less 
randomly until the last victory was forced by history.

However, it might be more fair, in fact, to award the first victory 
by plurality, then the next by plurality among the remaining, etc. 
This gives the first victory to the candidate most widely supported. 
Why would this be fairer than the initial random distribution?

Well, conditions change, people, both voters and candidates, die, and 
parties go out of existence. Being first is more likely to be of 
actual use, we can't really be sure that "later" will come.

It's debatable. However, I don't think it would be easy to show that 
the deterministic method I described would be *worse* than random ballot.

I think you could show that the deterministic method, over time, 
satisfied more voters, gave them the victory that they desired, than 
random ballot. Because of voters dying, specifically, before their 
turn comes. But I haven't done the math.

It would be important to determine conditions where Mr. Richie's 
statement is true. Under what conditions could random ballot be the 
ideal method? What goal of an election would be best served by random 
ballot, and not by another method designed specifically for that goal?