# Real Cambridge Ballots - Letter Three

New Democracy donald at mich.com
Tue Feb 25 03:10:55 PST 1997

```Dear List members,

H O W   M A N Y   P R E F E R E N C E S   S H O U L D   B E   M A D E

This is Letter Three dealing with ballots from one ward of a Cambridge
city council election - a real election with real candidates and real
voters that made real preference selections.
515 ballots   19 candidates   9 seats   57 share(Hare quota)

Ten ballots only had one selection. Eight of these needed a second selection.
20.0 percent did not need a second selection. (1-8/10)

505 ballots had two or more selections - 167 needed a third selection.
66.9 percent did not need a third selection. (1-167/505)

487 ballots had three or more selections - 62 needed a fourth selection.
87.3 percent did not need a fourth selection. (1-62/487)

447 ballots had four or more selections - 32 needed a fifth selection.
92.8 percent did not need a fifth selection. (1-32/447)

400 ballots had five or more selections - 14 needed a sixth selection.
96.5 percent did not need a sixth selection. (1-14/400)

315 ballots had six or more selections - 3 needed a seventh selection.
99.0 percent did not need a seventh selection. (1-3/315)

In this election 93.1 percent of the votes survived to the final count.
35.4 votes out of 515 failed to make it. This election had 93+ percent
voter representation - I would say that this is a good representation.

In this election of nine seats with nineteen candidates running in the
race, we should be able to say that if all the voters would have made at
least six selections then 99 percent of the votes would have survived to
the final count.

From the above informantion I tried to come up with a rule of thumb
that can be applied to all elections. A rule that will tell us the number
of preferences a voter should make. I do not feel that we can make a
mathematical equation that gives us the number of preferences a voter
should make. The following is what I have come up with at this time:

Each voter is to make a list of six names. The first three names are
to be selected without regard to any polls. The voter picks as his first
name the candidate that he thinks is best for the job. Then the voters
picks the second best and third best as his second and third names.
The next three names are to be picked using the polls. The voter takes
a look at the top four leading candidates in the polls - and the voter
picks the one that he feels is the best of the four. If this name is not
already on his list he puts the name down as number four - if he already
has this name on his list he does not put it down again. Next the voter
looks at the top three leading candidates in the polls and he picks the
best of the three. If this name is not a repeat from his list he puts the
name down on his list. And last - the voter considers the top two leading
candidates and he picks the name of the best of the two. This name is also
put on the list if it is not already there.
The voter now has a list of three to six names. This list of names
should reflect the wish of this voter. This method of making a list applies
to any preference election - single seat - multi seat - large number of
candidates.

I now feel that the polls must play an important part in preference
elections and it follows that the polls must be honest.

Yours,

Donald Eric Davison of New Democracy at http://www.mich.com/~donald

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

```