Re: [OVC-discuss] Ranked Choice Math

From: Dylan Hirsch-Shell <dylanhs_at_gmail_dot_com>
Date: Fri Aug 08 2008 - 23:05:51 CDT

It seems to me that the biggest problem with 0-9 is that it increases the
chances that in any given election the range of the voting scale will be
less than the number of candidates, which in turn prevents people from
giving a unique rating to each candidate (i.e. ranking), thereby potentially
forcing the voter to inaccurately indicate that they regard two candidates
as equally un/desirable.

In elections with less than 10 candidates, that's obviously not a problem.
However, it still limits the ability of a voter to express something like:
Candidate A is a bit more desirable than B, who is in turn a bit more
desirable than C, who is much more desirable than D, E, F, G, H, I and J,
each of whom is only slightly better than the next.

put another way: A>B>C >>> D>E>F>G>H>I>J
or in ratings: 100, 97, 95, 15, 13, 11, 7, 5, 3, 0

Compare to the 0-9 ratings: 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0

In such a situation, voters would be much more likely to bias their ratings
toward 0 for candidates D-J, rather than having to artificially inflate
their ratings.

Another disadvantage with a 0-9 scale would be an increased chance of ties
if there are few voters and most of them strategically degenerate their
ratings to 0's and 9's. This would be equivalent to approval voting, but it
would only really be an issue in small districts. I don't think the
increase likelihood of ties would be significant in the vast majority of
practical situations.

On Fri, Aug 8, 2008 at 7:59 PM, Rick Carback <carback1@umbc.edu> wrote:

> interesting, hadn't seen that one before! I've even asked Warren Smith
> about it and at the time he had no answer.
>
> "mathematically it does not really matter very much which one you choose"
>
> I don't think his justification for 0-9 being bad is very strong, and
> much of the 2 digit justifications aren't very convincing. He doesn't
> really consider elections where range == number of candidates. I'll
> have to send him another note about it.
>
> -R
>
> On Fri, Aug 8, 2008 at 9:25 PM, Dylan Hirsch-Shell <dylanhs@gmail.com>
> wrote:
> > On Fri, Aug 8, 2008 at 5:27 AM, Rick Carback <carback1@umbc.edu> wrote:
> >>
> >> I'd say 0-9 or 0-10 is probably the best range, because it is what
> >> most people are familiar with. Others say range should be # of
> >> candidates, so that people have the ability to rank if they so choose.
> >> CRV really hasn't asked or answered the question of what range is the
> >> best range. I'm not sure anyone over there cares.
> >
> > http://rangevoting.org/Why99.html
>

_______________________________________________
OVC-discuss mailing list
OVC-discuss@listman.sonic.net
http://lists.sonic.net/mailman/listinfo/ovc-discuss
By sending email to the OVC-discuss list, you thereby agree to release the content of your posts to the Public Domain--with the exception of copyrighted material quoted according to fair use, including publicly archiving at http://gnosis.python-hosting.com/voting-project/
==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Sun Aug 31 23:17:04 2008

This archive was generated by hypermail 2.1.8 : Sun Aug 31 2008 - 23:17:18 CDT