NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/rec/games/go/2311

Internet Message Format | 1992-11-19 | 4.1 KB

Path: sparky!uunet!paladin.american.edu!news.univie.ac.at!hp4at!mcsun!news.funet.fi!hydra!klaava!olounela
From: [e-mail] (Olli Lounela)
Newsgroups: rec.games.go
Subject: Re: Change my strength?
Message-ID: <[e-mail]>
Date: 19 Nov 92 14:07:07 GMT
References: <[e-mail]> <[e-mail]>
Organization: University of Helsinki
Lines: 75

In article <[e-mail]> [e-mail] (Tero Sand) writes:
>In article <[e-mail]> [e-mail] (Korhonen Mika) writes:
>>
>>Hi!
>>
>>I wonder when I sould change my rank in IGC?
>>
>>Should I change my rank right after the game, or should I wait until I have
>>played several games? How much sould I change my rank at once?

You should wait a few games; one game is a toss of coin.  Two games can
be random.  Even three games can be badly wrong.  I would recommend
changing the rank at earliest after four straight wins, or if a
sufficient number to give some confidence.  Say, a minimum of two thirds
of the games won (perhaps 7-3) would be good enough for a(n IGS)
promotion.

Make it a rule of thumb: You should have at least four wins before even
considering the change, and even then the winning percentage should be
better than 67%.

It is, naturally, completely different if somebody else (who knows what
he does) is doing the promoting.  One might, on occasion, get a
promotion on the spot in Finland.  Example: if you, Mika (~16 kyu), beat
a dan level player, you would be in for an instant promotion or two.  If
it happened without a handicap, you would have to be re-ranked to around
4 kyu, and would have an instant (temporary) rank of at the very least,
say, 7 kyu (that's 9 promotions, folks :-).

>If your rank was estimated at a club, I'd recommend sticking with it.
>You can possibly use results you get in IGS to get your club to promote
>you, though.

Then again, even in Finland we occasionally see players who don't come
to "official" clubs too often, and in the meanwhile the skill can change
dramatically.  Not to mention those who never went to a club...

However, using IGS results to rank oneself is strictly not to be
recommended, at the _*very*_least_ not in Finland.  One should bear in
mind that there is some wild misranking going on all the time there, and
then again there is the disparity of ranks across the world.  Example:
Jan van der Steen (Dutch 3 dan) uses IGS rank of 4 dan and *still* seems
to win much more than 50% of his games -- I'm actually waiting for the
day he decides to be 5 dan (sorry Jan, I couldn't resist :-)

I cannot quite remember who said this (I believe it was Tim Casey): "If
you think you are stonger than your rank implies, you are probably
wrong." The thing this implies, is that everybody considers himself
stronger than he is (which is a fact), and that only the real results
should be used as a basis for a rating.  I could add to it: "If you have
sufficient proof, go ahead and promote."

I don't, repeat, *do* *_not_* count lots of wins in the IGS for anything
else (anything beyond lots of wins in IGS, that is :-).  Several times
I've played people who claim about 3 dan, beaten them badly, and later I
find them 4 dans.  Furthermore, I seem to remember a case of one
promoting to 5 dan.  Does this mean I can beat most 4 dans (which
actually is not the case, not in the Europe at the very least) and get
promoted to 5 dan?

No wonder there were several 8 dans and all too many 7 dans in the IGS
just a while ago.  I'm still waiting to see just how many 7 dans will
remain when (and if :-) IGS rankings get organized to some (probably
more demanding) standard -- when I last visited, there seemed to be no
8 dans.

I am, BTW, a member of the Finnish Rating Committee.  While that is not
a guarantee of anything, we've been discussing grounds for promotions
and all kinds of misrankings and (mis)ranking (mis-)systems.

--
    Olli, 3 dan

E-mail: [e-mail]           ! .sig still under construction.
Blame me only for any opinions expressed. !  Never you mind, I don't either.