Newsgroups: rec.games.int-fiction
Path: news.duke.edu!newsgate.duke.edu!nntp-out.monmouth.com!newspeer.monmouth.com!news.maxwell.syr.edu!news-out.uswest.net!uunet!chi.uu.net!nyc.uu.net!world!buzzard
From: buzzard@world.std.com (Sean T Barrett)
Subject: Re: maximum # of games (was masochists)
Message-ID: <G4p0EG.Cw5@world.std.com>
Date: Mon, 27 Nov 2000 16:56:39 GMT
References: <G45vtz.791@world.std.com> <8vs3kl$4pe$1@news.lth.se> <vkmU5.14792$2A2.824092@news20.bellglobal.com> <8vtjv6$7b$1@nntp.Stanford.EDU>
Organization: The World Public Access UNIX, Brookline, MA
Lines: 59
Xref: news.duke.edu rec.games.int-fiction:59230

Tina <tina@eniac.stanford.edu> wrote:
>Kaia Vintr <kaia@xoe.com> wrote:
>>hope no one really thinks that, but I imagine some people might.  And it
>>probably matters to anyone who puts a lot of care and effort into rating 10
>>games and realises that his/her vote will only count 1/5th as much as that
>>of someone who whipped through all of them.
>
>You keep saying this, but it keeps being wrong. You are counting as
>"person's vote" a total number of votes. Yet... it's NOT. Each vote is
>worth precisely the same as every other vote.

This is only true if you take the argument out of context.
Here's the full-fledged argument:

  Let's say they population of voters consists of those who have
  time to play and vote on a lot of comp games, and those who only
  have time to play and vote on a few.

  To make the math come out more obviously, let's say that "a lot of
  games" means "every game", and "a few games" means "one game".

  Suppose the population of voters consists of 100 people of each type.
  Suppose the choice of games is randomly and evenly distributed.
  Then 100 people will play all 50 (say) games, and 100 people will
  play one game.

  Each game will receive 102 votes; 100 from the "play every game"
  voters and 2 from the "play one game" voters.

  Thus, while every vote is worth the same, the votes of the "few games"
  population underrepresent that population's opinion in the final tally.

So the point is not to compare individuals, but to compare equal-sized
samples or populations. (At which point the individual is when the sample
size goes to 1 and the argument "my vote is worth less" looks valid.)

Imagine if when women's suffrage was granted, men were allowed to vote
for every member of the house of representatives for their state, but
women were only allowed to vote for the one representing their geographic
area.  Each vote by a woman is worth as much as a vote by a man; but
the men's votes will outweight the women's votes, because the men get
more votes.

If you want every person to have an equal "voice" in the vote, people
should just be picking their top game or top 3 games and we shouldn't
use average votes; but this would discriminate against less-portable games
and in favor of games which people pick first instead of using a
random generation scheme (or, indeed, might simply get biased due
to insufficient votes to balance out the randomness).  [Note that
picking the top game or the top 3 games is not identical to saying
'only vote on 1/3 games', but related.]

I don't actually think it makes any sense to change it. But I do
sympathize with a feeling of disenfranchisement nonetheless.  But
then again, I put a lot of care and effort into 49 votes this year,
even though I knew that changing a game's rating by +-1 would only
have a +-0.01 effect on the final score--just like everybody else.

SeanB
