User avatar
Dixicrat
General
Posts: 523
Joined: Wed Aug 13, 2008 8:55 pm
Location: East Tennessee
Contact: ICQ

"Playing Strength" ratings for Gamers

Wed Sep 10, 2008 9:12 pm

Ever wonder how your playing strength compares with other gamers? Aphrodite Mae and I have been discussing the topic, and we'd like other members and guests on the forums to join this discussion.

Our idea is simple: A numeric rating, much like with FIDE (the international chess federation) or the USCF. The rating can be used to get an idea of the relative playing strength of potential opponents.

I'm not familiar with FIDE, but with the United States Chess Federation, I seem to recall that a difference in rating of 200 points indicates that the higher ranked player will win about 3 out of 4 games; and with a ratings difference of 400 points, the higher ranked player will win almost every game.

Until players have played 20 rated games, they have a provisional rating which is designated with an asterisk, e.g. 1785*. There are various classes of ratings: in the USCF, each class spans about 200 points, I seem to recall.

My wife (Aphrodite Mae) pointed out to me that this might be particularly useful in evaluating which AARs bear closer scrutiny. She also recommends that "officially rated" games involve a third participant as the host/referee, and that this person be "credentialed", much like tournament directors in the USCF. The credentialing might require a certain level of reputation on the forums, for example.

Changes in chess ratings involve a fairly complex mathematical formula, but it is easy enough to understand with a rudimentary understanding of algebra. One of the characteristics of the ratings which proceed from this formula is that the greater the difference in rating is, the greater the impact on each player's rating. Thus, if for example Mae somehow defeated Jabberwock in a "rated" AACW scenario, her rating would skyrocket.

Regards,
Both of us
[SIZE="3"]Regards,[/size]
Dixicrat

[SIGPIC][/SIGPIC]

Basic Training for AACW newcomers

User avatar
Spharv2
Posts: 1540
Joined: Sat Feb 25, 2006 5:39 am
Location: Tallahassee, FL

Wed Sep 10, 2008 10:24 pm

Interesting idea, but it could be tough to implement with unbalanced sides, Chess is evenly balanced after all.
Official Queen's Ambassador to the South
[SIGPIC][/SIGPIC]

User avatar
Dixicrat
General
Posts: 523
Joined: Wed Aug 13, 2008 8:55 pm
Location: East Tennessee
Contact: ICQ

Thu Sep 11, 2008 4:21 pm

Spharv2 wrote:Interesting idea, but it could be tough to implement with unbalanced sides, Chess is evenly balanced after all.


Avalon Hill used to have a PBM system which involved ratings, didn't they? I never participated in it myself, but I can certainly recall unbalanced scenarios in many AH games. Maybe someone who remembers more can comment...?

User avatar
Spharv2
Posts: 1540
Joined: Sat Feb 25, 2006 5:39 am
Location: Tallahassee, FL

Thu Sep 11, 2008 4:58 pm

Oh, it could be done, but it would be tough. You'd need a different point value for each side in each scenario. Having seen how impossible it was to get anything near agreement on general's attributes, good luck getting the balancing done. :) If you wanted to do it, I'd suggest keeping the team doing the evaluation as small as possible. You can adjust later if you want to. It would be a nice thing to have going into a PBEM though.
Official Queen's Ambassador to the South

[SIGPIC][/SIGPIC]

Bertram
Posts: 454
Joined: Thu Jan 31, 2008 8:22 pm

Thu Sep 11, 2008 6:17 pm

In general there are two ways to do it with an unbalanced game
- just ignore the unbalance. Assume each player plays enough scenario's on either side to even out the scores. Of course there will be players that "game the system" by only playing easy sides. As long as it isnt a competition, but just a way of assigning equally skilled players this isnt a real problem. Those players usually opt to play only the easy side of the scenario anyway, and eventually it will even out.
- Grade each scenario, both on average score for each side, and on weight of the scenario, and use that to give points. Problem is that you need a good number of finished battles to do that, and with games like these the average tends to drift, as players discover new ways of fighting the battles. Especially with Fog of War a scenario tends to play very different the twentieth time compared to the first time.

Of course for a meaningfull result you need a reasonable number of games played, and for the scenario's and the rusle to stay more or less the same. Both are, I think, rather problematic with the Ageod games...

But having said that, even a database with which side won, at what date and with what kind of NM and VP scores might be interesting, from a balancing and game design point of view.

Return to “General discussions”

Who is online

Users browsing this forum: No registered users and 23 guests