Life In 19x19 :: classification system for go tournaments

Author:

Elom [ Tue Jan 02, 2018 9:31 am ]

Post subject:

Re: classification system for go tournaments

Maybe goratings just might implement a weighting system for tournaments, when it is able to do so (it may require inclusion of the go4go tournament sheets?).

The classification would probably need to be rather simple, for a crass example:

Author:	event_horizon [ Fri Dec 29, 2017 2:22 am ]
Post subject:	classification system for go tournaments
I've met with some difficulties when trying to improve Ke Jie's and Park Junghwan's wikipedia pages. I think the key problem is there's no established classification system for go tournaments. @WindCaliber Can anyone offer some advice? Link to wiki talk page: https://en.wikipedia.org/wiki/Talk:Park_Junghwan Any suggestions on Ke's, Park's or related pages are welcome, e.g. better discriptions of their styles of play.

Life In 19x19 http://www.lifein19x19.com/

classification system for go tournaments http://www.lifein19x19.com/viewtopic.php?f=13&t=15342	Page 1 of 1

Author:	bugcat [ Fri Dec 29, 2017 12:06 pm ]
Post subject:	Re: classification system for go tournaments
I'll try my best to add my penny's-worth. Are you looking for a classification like in chess, with quite a lot of discrete numerical grade relating to the Elo or Elo-style rank of the participants? That can be done with GoRatings, which not to say that I suggest it. If you're looking for a broader classification, there are two variables that describe a tournament: 1. Scope. This is who is allowed to participate. Firstly geographical scope: tournaments are typically international or national (generally Japanese, Chinese, or (South) Korean); though there are exceptions such as the Okan and the Kansa Ki-in No. 1. Secondly age scope: Some tournaments are restricted by age, such as the Shinjin-O (younger players) and the Igo Masters Cup (older players). And there are invitational tournaments such as the China-Japan-Korea Super Meijin. 2. Time control. Generally, especially in the Orient, this is in main time with byo-yomi, but there are also variant time controls like in the NHK Cup. For instance, this could be crudely split into Blitz (< 20 mins main time / player), Fast (20 mins -- 1 hr main time / player), Fast-Classical (1 hr -- 2 hr main time / player) and Classical (2 hr or more main time / player). In example, one could (if one wished) describe the Honinbo title as (I think) a "Japanese classical / fast-classical pro tournament".

Author:	hyperpape [ Mon Jan 01, 2018 5:49 pm ]
Post subject:	Re: classification system for go tournaments
So far as there are established classifications that are used in CJK go media, we should follow them. Otherwise, I think this is likely to be a mess.

Page 1 of 1	All times are UTC - 8 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/

Author:	Elom [ Tue Jan 02, 2018 9:31 am ]
Post subject:	Re: classification system for go tournaments
Maybe goratings just might implement a weighting system for tournaments, when it is able to do so (it may require inclusion of the go4go tournament sheets?). The classification would probably need to be rather simple, for a crass example: (xX) represents multiplication of rating change. Standard: Open Professional International = A* (x2) A* = (x2.00) A = (x1.00) B = (x0.50) C = (x0.25) E... and so on in powers of 0.5. For every one of these restrictions, the grade of the tournament is lowered. 1-To just one particular country (e.g. Japanese tournament) or contingent (e.g. European-wide tournament) 2-To a certain age bracket or gender 3-If it uses 'fast' (define it in a way you want) time limits, with no differentiation between blitz for simplicity. For example, the NHK cup ticks the downgrades of National and Fast and suffers two downgrades to B. If a player plays in a closed system and acquires a high winning percentage against other players in the system, the rating mathematics may under estimate the difference in rating required to obtain such a high winning percentage, thereby under-rating the high performing players and over-rating the other players in the closed system. However, if the rating mathematics over estimates the required difference in rating to obtain such a high winning percentage, it over-rates the high performing player while under-rating everyone else in the closed system. Do we know which direction elo swings? Ideologically speaking, the ideal of a tournament classification system is so that dampens the effect of closed-system results by weakening the effect of a game result on the rating system the more 'closed' it is (closed vaguely describes what is meant).