From: Kathy Dopp <kathy.dopp@gmail.com><br>
> <br>
<br>
Comments below<br>
<br>
> ---------- Forwarded message ----------<br>
> From: <LWVTopics@yahoogroups.com><br>
> Date: Sat, May 10, 2008 at 7:22 AM<br>
> <br>
> Pathological scenarios<br>
> can be constructed for all voting systems, including whatever ones Kathy<br>
> may prefer to IRV.<br>
<br>
Diagrams from Ka-Ping Yee show that IRV is very weird <font face="Arial, Helvetica, sans-serif">when</font> compared to <br>
other<font face="Arial, Helvetica, sans-serif"> </font>methods.<br>
<br>
They <font face="Arial, Helvetica, sans-serif">are a bit complex, but are very informative once you get them. They<br>
allow voting methods to be compared visually.<br>
</font><br>
http://zesty.ca/voting/sim/<font face="Arial, Helvetica, sans-serif"><br>
<br>
Also, his results were replicated elsewhere:<br>
<br>
http://bolson.org/voting/sim_one_seat/www/<br>
http://rangevoting.org/IEVS/Pictures.html<br>
http://ivnryan.com/ping_yee/results.html<br>
</font><br>
Also, IRV suffers from centre squeeze. This means that a compromise<br>
candidate who would be mid-way between the 2 main contenders cannot<br>
win.<br>
<br>
> Yes. I prefer IRV to all other systems for electing a single winner to<br>
> an executive office because it elects a majority winner in a single<br>
> election without the need for a separate expensive runoff election. <br>
<br>
Maybe. It depends on how you define majority.<br>
<br>
For example<br>
<br>
45: A>B>C<br>
40: B>C>A<br>
15: C<br>
<br>
C gets eliminated in the first round and A wins with 45 of the remaining<br>
votes (85), so he has 53% of the remaining vote but only 45% of the votes<br>
cast. This is a majority of the remaining votes but not a majority of <br>
the cast votes.<br>
<br>
> I'm not sure how Kathy defines "support", but it must be different from<br>
> how I would define it. IRV uses the same basic logic as a traditional<br>
> runoff election to find a majority winner. Any argument along the<br>
> lines Kathy presents is equally true of IRV AND traditional runoff<br>
> elections. Imagine a traditional runoff election (or IRV election)<br>
> with three candidates, A, B, and C. If A gets 48%, B gets 46% and C<br>
> gets 6%, the runoff would be between A or B. Now let's say all the C<br>
> voters prefer B over A, so B ends up with 52% to A's 48%.<br>
<br>
The situation is that:<br>
<br>
C beats A<br>
C beats B<br>
A beats B<br>
<br>
Shouldn't C win ?<br>
<br>
More importantly, if A or B were to decide not to run, then C wins <br>
with IRV.<br>
<br>
I think the above example could be described as something like<br>
<br>
48%: Love A, like C and hate B<br>
46%: Love B, like C and hate <font face="Arial, Helvetica, sans-serif">A</font><br>
6%: Love C<br>
<br>
A is loved by 48%<br>
B is loved by 46%<br>
C is at least liked by everyone<br>
<br>
It isn't obvious that electing C is a bad choice.<br>
<br>
> I don't think most people would support a system<br>
> that would declare the winner of a 3-way race the candidate who was the<br>
> first choice of only 6% of the voters.<br>
<br>
First choices aren't everything. C gets 1st or 2nd choice from everyone.<br>
<br>
Also, neither A or B are 1st choice of a majority.<br>
<br>
> But all<br>
> of them agree that our current election method, plurality elections<br>
> (complete with the spoiler problem), is the worst method.<br>
<br>
Right, plurality is useless. However, IRV might be only slightly better.<br>
<br>
It does eliminate the spoiler, but only when the 3rd party is small.<br>
<br>
In Australia where it is used, it maintains the 2 party system, so if<br>
helping 3rd parties have a chance is your goal, then IRV isn't going<br>
to help much.<br>
<br>
> The vast majority of informed opinion (not just that of the<br>
> IRV-dislikers) is that IRV is the best option for single-seat<br>
> elections. <br>
<br>
> >Also some other reasons not to promote IRV (besides the fact that it<br>
> >often ends up putting candidates into office that are not supported by<br>
> >- or even are opposed by - a majority of voters) include:<br>
> <br>
> It is theoretically possible to have an election in which NO candidate<br>
> has majority support, like a rock, paper, scissors endless loop, but IRV<br>
> is FAR more likely to elect a true majority choice than our current<br>
> plurality election method with its rampant spoiler problem.<br>
<br>
The condorcet loop doesn't apply to IRV.<br>
<br>
The reason that the winner can sometimes not have majority support is <br>
that some ballots will 'exhaust'. If all the candidates on the ballot<br>
have already been eliminated (Say the voter only votes for 1 candidate<br>
and he is eliminated), then the total number of ballots remaining drops.<br>
This means that you can get more than half of the votes in the last<br>
round with less than half of the votes cast in the first round.<br>
<br>
> IRV has been counted easily and quickly in recent U.S. elections in<br>
> places like Burlington, Vermont, and Cary, North Carolina. Auditing<br>
> can be more complicated, simply because there is more information from<br>
> each voter to double check, but can also be more thorough. <br>
<br>
Another issue is that it requires a central office to handle the counts.<br>
With plurality (and many other systems), each polling station can <br>
announce its results and then they are added together by the central<br>
office.<br>
<br>
With IRV, they must all be sent to a central location<font face="Arial, Helvetica, sans-serif"> for counting</font>, <font face="Arial, Helvetica, sans-serif"><br>
</font>or each <font face="Arial, Helvetica, sans-serif"></font>round must be completed locally and then new instructions sent<br>
based on the result<font face="Arial, Helvetica, sans-serif"> of each round</font>.<br>
<br>
> San<br>
> Francisco and Burlington, for example, posted a record of every<br>
> ballot's set of rankings on the Internet, allowing anyone to tally<br>
> those elections themselves.<br>
<br>
That's pretty cool, and should be encouraged as long as paper ballots<br>
were used. <br>
<br>
Was it electronic voting? If so then the cure is worse that the disease.<br>
<br>
IRV may end up encouraging electronic voting due to its <font face="Arial, Helvetica, sans-serif">counting </font>complexity <font face="Arial, Helvetica, sans-serif"><br>
</font>with all the security issues that leads to.<br>
<br>
> As to IRV not achieving its "claims": There are two claims often made<br>
> for IRV that may be somewhat over-stated. Although political<br>
> scientists and courts have recognized IRV as a majority voting method,<br>
> it is possible that some voters will not rank either of the two<br>
> finalists in the runoff count. <br>
<br>
Right, it comes down to how you define majority.<br>
<br>
Also, with IRV we are only sure that the final winner is prefered to one<br>
other candidate (the one he beats in the last round). He might lose<br>
when compared to every other candiate.<br>
<br>
> The second claim of IRV that may be over-stated is that it eliminates<br>
> the "spoiler problem." While compared to our current voting method,<br>
> IRV indeed solves the spoiler problem when independents or third<br>
> parties run against two major parties. However, concerns about<br>
> "spoilers" could possibly resurface in some scenarios when there are<br>
> three or more major parties. But none of those scenarios have appeared<br>
> in practice in those countries that use IRV and have more than two<br>
> major parties.<br>
<br>
In Australia, IRV results in an effective 2 party system. However, they <br>
have more than 2 parties in the their Senate.<br>
<br>
> As Arrow's Impossibility Theorem shows, it is impossible to invent a<br>
> perfect voting method, as reasonable and desirable features that one<br>
> would want in a voting method are mutually exclusive.<br>
<font face="Arial, Helvetica, sans-serif"><br>
It doesn't mean that you can't compare methods to each other.<br>
<br>
As the Yee diagrams show, IRV favours partisan candidates. You<br>
win by getting lots of first choices. It doesn't elect compromise <br>
candidates.<br>
<br>
In effect, one block of voters gets their first choice and the other<br>
block gets a candidate they hate.<br>
<br>
Wouldn't it be better to have a compromise candidate that most<br>
of the voters like, even if nobody gets their favourite ?<br>
</font><br>
> IRV does better<br>
> than most in real world experience, and virtually all elections methods<br>
> experts agree it is superior to plurality elections we use currently.<br>
<br>
That is a pretty low bar. Plurality is a horrible system.<br>
<br>
<div class="AOLPromoFooter">
<hr style="margin-top:10px;" />
AOL's new homepage has launched. Take a <a href="http://info.aol.co.uk/homepage/" target=_blank>tour</a> now.
</div>