[EM] 12/13/02 - Giving `crutches to weak candidates':

Craig Carey research at ijs.co.nz
Fri Dec 20 13:39:35 PST 2002

Her are some examples constructed to be similar to the example that was
recently provided.

The Alternative Vote is only discarding about 17% of the vote which is a lot less than what it can do for
4 candidates.

These 1 winner 5 candidate examples are designed to:
* Have B win
* Have an order of elimination be E, then A, then D
* Have C before B and D, and B and D before A and, both counting of the
1st preference and the 2nd preference.

AB    19
BC    20
C     37
CD     1
D     23

1st pref: A:B:C:D:E = 19:20:38:23:0
2nd pref: A:B:C:D:E =. 0:19:20: 1:0

AB    19
BC    20
C     38
D     19
ED     4

1st pref: A:B:C:D:E = 19:20:38:19:4
2nd pref: A:B:C:D:E =. 0:19:20: 4:0

Candidate A flips from being winner into a loser here:

----------------------------
A       19999      80004
B           1          5
BA      19997      19997
CB      40002      40002
DBA     20001      20001
----------------------------
Total:     100000     160009
AV Winner:      A          B

Details:
[1]. 1:elim B: 19997+19999=39996(A)
[1]. 2:elim D: 39996+20001=59997(A)
[1]. 3:test 40002(C)<59997(A)-->A wins
[2]. 1:elim D: 20001+19997=39998(BA)
[2]. 2:elim C: 40002+5+39998=80005
[2]. 3:test 80004(A)<80005(B)-->B wins

As can be spotted, a mere 4 First Past the Post votes, are entirely
enough to cause the banishments of the hopes of 60,005 other voters
that also would use First Past the Post papers in a 1 winner STV
election.

Monotonicity is the principle that prohibits that problem.

The man-cow, Russ P., probably is reluctant to copy that page onto
the black website that was translated into French. Presumably it is
the best 4 candidate 1 winner Alternative Example known.

At 2002\12\20 14:25 -0500 Friday, Adam Tarr wrote:
>I must admit I'm not the least bit surprised that Donald Davison has not
>responded to my message from 12/18.  This is standard debating style for
>Donald -- post an inflammatory criticism of Condorcet or Approval, ignore
>the reasoned responses, rinse, repeat.  Would it be too much to ask him to
>defend IRV from criticism, this time?
>
>why the centrist candidate below should lose (although he
>shouldn't).  Explain why the example below is unrealistic (although it

As more candidates are added, the clarity of the examples would reduce
because there was a considering of only the 2nd preferences and only
the 3rd preferences, ... If Donald says that the attack on IRV must be
so very severe that 100 candidates are in the example, then nothing
could be said since the winners are too hard to guess at.

>isn't).  Explain why IRV's monotonicity violations are less of a concern
>that Condorcet's cyclical ties (although they aren't).  Even better,
>provide an example that is realistic in your estimation, where Condorcet
>elects a bad candidate. (I bet you can't.)
>

>
>On 12/18/2002 I wrote:
>
>>Don Davison wrote:
>>
>>>... why are
>>>you supporting Condorcet and/or Approval Voting?  For, this is what these
>>>two method do, they give `crutches to weak candidates'.
...

----