Page 1 of 1

Ranking

Posted: Mon Oct 05, 2015 3:39 pm
by Mauro
During tournaments, how does ranking work? I read somewhere, not sure if on Osamuko or where, that usually position average is used as indicator, but on Arcturus I read World series used standard cumulative score, with points to the first decimal place (so I'm not sure how they are counted).

I was wondering: which is the most used method during official tournaments? How does it works at worlds?

Re: Ranking

Posted: Mon Oct 05, 2015 5:56 pm
by Gnom
I don't remember seeing a riichi tournament that uses position as the winning factor (although it is common in MCR and possibly other mahjong rules through the system of table points). While it may be possible, the standard is to sum up the points of each game.

Re: Ranking

Posted: Mon Oct 05, 2015 6:01 pm
by Shirluban
In EMA and WRC tournaments, you sum your points from every hanchan. More points the better.

MCR uses table points, but it's essentially the same: you sum table points, ties are resolved by looking at points' sum.

Re: Ranking

Posted: Mon Oct 05, 2015 6:37 pm
by Mauro
Thanks; points I guess are those with uma; what are table points?

About world series points, they have decimal because to calculate them they don't round the point each player has at the end of the game?

Re: Ranking

Posted: Mon Oct 05, 2015 7:39 pm
by Shirluban
Table points are points earned depending on your hanchan ranking.
The difference with uma is uma is added to your score, while TP makes a separate scoring.

FYI, at MCR, 1st gets 4 TP, 2nd => 2 TP, 3rd => 1 TP, 4th => 0 TP.

Re: Ranking

Posted: Tue Oct 06, 2015 4:15 am
by Gnom
Shirluban wrote:MCR uses table points, bt it's essentially the same: you sum table points, ties are resolved by looking at points' sum.
What about the case of players having a big points gap? The table points system reduces the benefit of winning big (especially when using small/no uma or rulesets without bankruptcy). It is not guaranteed the resulting ranking will be the same under both systems, although they will be similar.

Re: Ranking

Posted: Wed Oct 07, 2015 12:36 pm
by Mauro
Gnom wrote:What about the case of players having a big points gap? The table points system reduces the benefit of winning big (especially when using small/no uma or rulesets without bankruptcy)
About this it would be interesting to know how much big wins are linked to skill, and how much to luck: some hands require luck, other are due only to luck, so maybe table points could somewhat level luck's influence?

Re: Ranking

Posted: Wed Oct 07, 2015 2:17 pm
by Gnom
It's kind of an endless debate... Yes it does reduce the gain from making a yakuman that requires luck but it also reduces the gain from careful calculations. And many players (including me) feel that luck is part of mahjong, and also that it is a skill to be able to take advantage of luck (many beginners can't make yakuman not because they lack the luck but because they fail to recognize it), so it should be rewarded to some extent... Personally I'd be very frustrated to play in a tournament using table points...

Re: Ranking

Posted: Wed Oct 07, 2015 4:44 pm
by Shirluban
The problem with table points is it makes it impossible to recover your loss.
During a small tournament, if you're second at the table, even only once, even by only one single point, then you're pretty much banned from tournament's first place.

Re: Ranking

Posted: Thu Oct 08, 2015 2:34 am
by Senechal
The inherent problem with table points is that it is ranking a tournament as 8 mini-tournaments is not what people are playing. People are playing 1 8-match tournament. Rank bonuses reward but do not abstract the objective of the game.

Take soccer. If we gave 1 point to whoever leads the first half, 1 to whoever leads after the second half, and 1 to whoever dominates both, you can still get 3pt for winning and 0pt for losing (standard soccer value) but if a 1-0 half ends later 2-4, both teams get 1pt, not the standard 0-and-3 for a clear victory. You could then try amending it with giving both sides a point for drawing a half but then can get 2-and-2 ties, or some 2-and-1 where some draws are worth more than others. The more a system is made complicated, the more people will scream for simplicity.

MCR abstracts games to a point system. WSoM (specific Zung Jung use) roots point scores. Riichi normally uses points to rank in a wide event, or it will straight up eliminate people on table ranking... something not yet common in Western circles.

Some systems work better in a given setting, others generally do not work well outside of their framework. And some people will say stuff like "you get 4/3/2/1 points plus 0.01% of your score" when there are easier ways to express it (15/45 uma rank bonus).

Re: Ranking

Posted: Thu Oct 08, 2015 8:23 am
by Mauro
15/45 uma as in +45 +15 -15 -45? I never saw it used, which rules use it?

Re: Ranking

Posted: Sat Oct 10, 2015 2:48 pm
by Senechal
Mauro wrote:15/45 uma as in +45 +15 -15 -45? I never saw it used, which rules use it?
A Polish tournament announced somewhere on the forum here in Bielewa iirc.

Re: Ranking

Posted: Thu Oct 15, 2015 6:44 pm
by Krabman
Yeah, it was our tournament. I had no idea it could've been expressed with that uma until someone (was it you, Senechal?) explained that to me haha

Re: Ranking

Posted: Thu Oct 15, 2015 7:36 pm
by Mauro
Why did you choose an uma so high?

Re: Ranking

Posted: Fri Oct 16, 2015 12:25 pm
by Krabman
I just followed the system used by the Polish Mahjong League for non-EMA tournaments. Table points: 1st - 10; 2nd - 7; 3rd - 4; 4th - 1. I wanted to emphasize places, not points. In case of tied TP, final positions were decided based on Tenbo. The last criterion was number of 1st places each player had.

If I were to organize another non-EMA tournament (which I'd gladly do as I don't like the current ruleset), I'd go for a different uma. The problem is, some players won't bother playing non-EMA events since they can't get any ranking points that way.

So, another EMA tournament in Bielawa has already been confirmed for 2016. I still don't know about a non-EMA one. I think I will try & go for it. I should be able to get that minimum of 12 people to play, I hope.