[EM] YES/Number Votes Method
DEMOREP1 at aol.com
DEMOREP1 at aol.com
Sun Sep 26 19:39:49 PDT 1999
For simpler tiebreaker purposes there is the following --
Summary -
A. Vote YES/NO on each choice and rank the choices using number votes (1, 2,
etc.) or use a scale vote such as 0 to 100 with 51 or more being deemed a YES
vote). Using scale votes would show that many winners may not be especially
popular.
B. Executive and Judicial offices (1 or more positions) - Two or more choices
getting YES majorities go head to head using the number votes.
Legislative bodies - Both parties and individual candidates would go head to
head.
Head to head involves doing all of the combinations of Test Winner(s) versus
Test Loser (with other choices being deemed Other Test Losers). It would
generally require a computer to do all of the math in a large election.
C. If head to head fails to fill all positions, then drop the choice with the
lowest number of YES votes and recheck the head to head math for the
remaining position(s).
D. Legislative body winners would have a voting power equal to the number of
votes finally received (first choice votes plus transferred votes from
losers) (i.e. No STV type fractional transfer votes math to worry about.)
---
For a single winner office I note again that simple Approval Voting (vote 1,
X or YES for a choice) is defective since a first choice majority may lose.
51 A B
49 B
B wins using simple AV
That is, simple AV is defective because it fails to rank choices and have
them go head to head.
--
Even the simple 3 choice case is rather complex ---
YES votes
A
AB
AC
ABC
ACB
B
BA
BC
BAC
BCA
C
CA
CB
CAB
CBA
That is, there are 15 possible types of YES votes with the simple summary
matrix --
Place Vote
1 2 3 Total
A1 A2 A3 AT
B1 B2 B3 BT
C1 C2 C3 CT
For legislative bodies - If a party gets N Droop Quotas, then it would get N
seats.
Using the lowest YES votes as a tiebreaker would replace having to look at
worst head to head defeats.
More information about the Election-Methods
mailing list