Re: Re: Owls and Voles --- Rankings

Mark -

I agree with you that someone who resigns with press sent to the GM
explaining why shouldn't be penalised for it unless it's something they do
repeatedly such that it becomes overly disruptive or it seems like they're
abusing the system.

As far as a player repeatedly NMRing, I'm inclined to say that if a player
doesn't log in at all for the space of a (game) year, they'll be resigned
and penalised. If they do that in more than one game in a given period of
time, they'll be blacklisted unless they get in contact to explain what
happened and provide some assurance that it won't happen again.

Those rules are, of course, not hard and fast since's it's ultimately the
GMs discretion that will be the guiding force.

jhwoll -

I don't see the need for any form of explicit dependency ranking. If you're
consistently missing deadlines and spoiling games for other players then
you're likely to get yourself blacklisted from games that I run. I see
little need to name and shame people, I'm just not interested in having
them in my games.

As far as a ranking system based on total points (no matter how you wish to
award them) goes, I am very much of the belief that such a system alters
people's incentives to a great extent. In such a system, the incentive is
to play as many games as possible, against the weakest opponents you can,
since that will maximise the points awarded to you.

A rating system (which it should be noted is not the same as a ranking
system) incentivises playing against people that are of a similar or higher
level to you, as that's the best way to improve your rating - winning a
game against weak players will have little to no effect on your rating.
Unlike a ranking system as you've described, your rating is not a strict
measure of how many wins or losses you've had, but a measure of your skill
compared to all the other players. The difference is admittedly subtle, but
important.

As far as Bounced is concerned, while I'm not very familiar with it, it
certainly doesn't give the impression of being easier to use than the
DPJudge, especially given that almost everything can be done though the
web-interface with the exception of setting absences and joining games -
the syntax for which is pretty trivial.

Chris -

Thanks for the pointer to JDPR - I can't however find any more information
about it in Google since all the pages that discuss it seem to be dead. I'm
planning on using TrueSkill which is Elo inspired as well and has been well
tested as the ranking and match making algorithm used for XBox Live, so I'd
expect reasonably similar outcomes. If you've got anything more about JDPR
I'd be interested in seeing it though.

Ed


On Thu, Jan 5, 2012 at 18:23, <[email protected]> wrote:

> **
>
>
>      I have been following the discussion regarding ratings in the group.
> I have only ever played in two Owls games (Owls_33 and Owls_Kafka), and was
> never invited back, so I suppose I am as disinterested an observer as
> anyone.  There do need to be to sets of rankings, one for dependability,
> and one for results.  The main problem with the 'results' scoring systems I
> have seen is that they inevitably affect the style of play.  My suggestion
> for a won-lost scoring system is that a solo gets 2 points, and a two-way
> draw gets 1 point for each player.  That's it.  Any other result is a loss
> for everyone, with no points awarded.  That gives you a Won-Lost-Draw
> record.  The ranking system should sort players by total points earned,
> with most wins as a tie-breaker, and least losses as a second-tier
> tie-breaker.
>
>     All this, of course, is presented on an 'IMHO' basis.  The syntax and
> details of the DPJudge system are needlessly primitive, and I much prefer
> the 'Bounced' interface.  I have seen comments to the effect of 'we don't
> just want a bounced-type system'.  My question is, 'Why not?'  It is a
> user-friendly interface.  I think that a similar system, with a simplified
> ranking system, would be a great improvement.
>
>
>
> "Democracy is two wolves and a lamb voting on what to have for lunch.
> Liberty is a well-armed lamb contesting the vote."
> - Benjamin Franklin, 1759
>
>
>
> ------------------------------
>
>  I also agree with Eric's assessment and think OWLS to VOLES scoring will
> work well.  Items like communication and roleplaying are too subjective and
> belong in a subsequent class of games.
>
> I believe a resignation should not be punished because a player may have a
> very good reason to resign a game due to personal commitments, protracted
> or stalled game, moving cities, changing jobs, etc.  A player that contacts
> the GM and explains his need to resign should not be punished.  I have had
> to do it on rare occasion in protracted games due to military moves and the
> like.  Abandoned powers are another matter.  Abandonments are rude and
> childish, unless of course the person was hit by a bus or eaten by a shark;
> then it is unfortunate.
>
> Thanks again Ed.
>
> -Mark
>
>  ------------------------------
> *From:* kirman36 <[email protected]>
> *To:* [email protected]
> *Sent:* Thursday, January 5, 2012 6:50 AM
> *Subject:* [dpjudge] Re: Owls and Voles --- Rankings
>
>
> That sounds great. I was hoping you would stay with the old scoring system
> with only slight tweaks. I personally would be against ranking that favors
> communications. In many cases that would force peoples hands. Sometimes
> saying nothing is more and other times you know you simply will be giving
> up info, no matter what you say. I do agree that players that quit games
> should pay a price. Real life happens and people need to quit at points.
> But time and time again I have felt players quit for bad reasons. Such
> as... they are losing, people will not work with them the way they want,
> Such and such power will not talk too me. Anyway I would be in favor of
> some kind of harsh score adjustment or even a sitting out time for players
> that quit games over and over. Anyway thanks for taking over Edward, just
> throwing in my 2 cents. Sounds like you got it handled.
>
> Eric Kirwan
>
>
>
>
>  
>