
Adding last choices
a,b,c,de
a,b,c,de
a,b,c,de
b,a,c,de
c,b,a,de
c,b,a,de
d,c,e,ba
e,d,c,ba
e,d,c,ba
e,d,c,ba
Head to head winners
Spok Cond
a,b 3, 3 3, 7 *
a,c 4, 2 4, 6 *
a,d 6, 0 6, 4
a,e 0, 0 6, 4 *
b,c 4, 6
b,d 6, 4
b,e 0, 4 6, 4 *
c,d 6, 4
c,e 1, 3 7, 3 *
d,e 1, 3 7, 3 *
*= Condorcet head to head winner changes compared to Spokane
Condorcet summary
a >d,e
b,c> a
b,c>a>d,e
b>d,e
c>d,e
c>b c is Condorcet winner
Comments The Spokane example is set up to produce its result using the
Spokane method 10 total votes using the 10 possible candidate combinations.
Note that only the b,c,d combinations have a total of 10 votes in the Spok
column above with the Condorcet result of them being c>b>d.
The Spokane method is approval voting in rotation for n1 choices, n2
choices, n3 choices, etc. (n= number of candidates) and eliminating the
least approved in each round. (instant run offs) or another words a form of
approval voting with instant runoffs.
The Spokane method most certainly does not conform with the plain Condorcet
method.
Truncated ballots would cause candidates to be eliminated earlier.
Also, a Condorcet winner as the first choice on 6 ballots would be eliminated
in a Spokane method first round (a direct violation of majority rule) if each
of the other four candidates were mentioned 7 or more times on the 10 ballots
(40 Spokane positions (10 ballots x 4) minus 6 = 34 positions for the other
four candidates or an average of 8.5 positions).
If anything the Spokane method shows a fatal defect in approval voting if the
votes in approval voting are not exactly equal.
