Newsgroups: rec.games.int-fiction
Path: news.duke.edu!newsgate.duke.edu!nntp-out.monmouth.com!newspeer.monmouth.com!news-peer.gip.net!news.gsl.net!gip.net!cpk-news-hub1.bbnplanet.com!denver-snf1.gtei.net!news.gtei.net!coop.net!world!buzzard
From: buzzard@world.std.com (Sean T Barrett)
Subject: Re: [Comp2k] Results
Message-ID: <G4AL1w.1n7@world.std.com>
Date: Sun, 19 Nov 2000 21:58:44 GMT
References: <jdu297hp2o.fsf@login2.phy.duke.edu> <3tHR5.97276$1C6.3855251@news20.bellglobal.com> <3a180897$1$zeoyber$mr2ice@news.supernews.com> <SpVR5.98732$1C6.4002409@news20.bellglobal.com>
Organization: The World Public Access UNIX, Brookline, MA
Lines: 36
Xref: news.duke.edu rec.games.int-fiction:58757

Kaia Vintr <kaia@xoe.com> wrote:
>There were probably many different interpretations of the scale.
>But whatever your interpretation, if you wanted your vote to count
>you needed to use both ends. If I had understood 10 to mean
>perfect and 1 extremely negative I would have given mostly 5s
>and had little impact on the results.

If you want your vote to count as much as possible, you should
*only* vote 1's and 10's.  That will likely have the biggest
impact.

Not necessarily the impact you want, though; if you think a game
deserves a 7, and the average in the comp is otherwise a 7.5, then
a '1' is the best thing to vote; but if the average in the comp
is otherwise a 6.5, then a '10' is the best thing to vote;
and since you don't know which way it's going to go, the best
way to move it as close to 7 as possible is vote it a 7; the
further the average (not counting you) is from 7, the further
it will move to 7.

On the other hand, on finding that all my votes ranged from 1-8,
I did in fact renormalize them to 1-9, and I'm thinking that
next year I may try to simply spread my votes out evenly from
1-10 after the fact.

It's also worth noting that bumping up my vote by 1 (when I
renormalized) boosted some games' comp scores by roughly 0.01,
which was the difference between adjacent games in a few places;
i.e. that bumping a vote by one all by itself ("one vote"?)
could have made a difference.  Bumping to all 1s and 10s would
no doubt make a bigger difference.

But if everybody did it, it would be equivalent to everyone voting
"good" or "bad", which is kind of pointless.

SeanB
