[Icehouse] IGDC rankings idea

[David Artman]

> From: Dale Sheldon <dales at digitalfreaks.org>
> Instincts are flawed.

Sure, I'm just saying that something in the back of my mind "likes" the
notion of point allocation over ranking. It *seems* more nuanced--in
that it permits more "relative" ranking than "better, worse, tied"--and
that nuance *seems* more powerful for getting at the best possible ranks
across several ballots.

> I did this vote's solution by hand, in between fits of work.  I can show 
> you how, if it'll improve your opinion of the method :)

Not at this time, thanks. :)

That was really just a minor point--I trust the Dark Masters. Sure, I'd
like to eventually grok the CRP method and be able to do it by hand (or
with script). But the A&T method would let me do such trivially, with a
spreadsheet.

I defer to those who can "see the numbers at work." If we are going to
compare methods, to find one that suits us, then I see these as the
priorities:
1) Stability - Is not easily manipulated by a bloc of voters.
2) Flexibility - Freedom to vote for some or all games means more
ballots, in the end.
3) Openness - Anyone can judge; no cabal or panel of judges.

Further, I am not adverse to a significant tweak to the competition, to
more-readily achieve the IGDC Goals. Require feedback to provide a rank;
weight rankings based on the (claimed) number of games played;
something.  You might notice that "fair" doesn't appear in the above
priorities for a voting method: rather than bog down in the semantics of
"fair," I'd rather unpack the IGDC Goals and look at ways they can be
better served (if not by a competition, then by what?).

Perhaps we need to begin to dictate "how to judge," rather than leaving
it up to subjective vagueness? Maybe have folks give points to games
based on a set of criteria, and use those points totals to establish the
ranks. Yes, that's still "gamable" or "biasable," but perhaps it would
dampen the tendency for a game-type preference to dominate rankings.

For instance, suppose a bunch of BGG folks decide to participate--real
die-hard grognards. I suspect that any game which is, say, a miniatures
or RPG game is basically doomed, at that point: it's the wrong game
type, and there's no judging constraints. HOWEVER, that could be
dampened, if every judge had to rank on, say, these criteria:
* Rules Clarity (10 points) - Could you play immediately, did anything
confuse?
* Conformity to Restriction (20 points) - How much did the game embrace
the design restriction (or did it give lip-service to it)?
* Fun Factor (20 points) - Was it fun?
* Creativity (20 points) - Was it unique, or was it a reskinning of an
existing game using pyramids?
* Replayability (20 points) - Is it worth playing more than a few times?
Is it a "lifetime to master" game or something you play for a bit then
forget forever?
* Looniness (10 points) - Does it encourage the players to invest in
their pyramids collection? Does it reflect well on the Looney Labs and
their social ideals?

See what I mean? Such criteria make the judges think about more than
just "I like board games, so this wins" or "I hate randomness so this
loses." Thoughts?
David