home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: rec.games.go
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!sgiblab!zaphod.mps.ohio-state.edu!caen!umeecs!umn.edu!lynx!carina.unm.edu!blowfish
- From: blowfish@carina.unm.edu (rON.)
- Subject: Re: winning ratio
- Message-ID: <cllq5fh@lynx.unm.edu>
- Date: Wed, 18 Nov 92 16:46:25 GMT
- Organization: University of New Mexico, Albuquerque
- References: <1992Nov17.023826.2810@bhprtc.scpd.oz.au> <1992Nov18.021006.14737@uxmail.ust.hk> <1992Nov18.060917.27210@u.washington.edu>
- Lines: 57
-
- I found the refered to article:
-
- In article <1992Nov18.060917.27210@u.washington.edu> adrian@stein.u.washington.edu (Adrian Mariano) writes:
- >The formula was obtained from study of over 2000 games played in
- >tournaments. The study was done by Jos Vermaseren. This doesn't
- >conform to the (omitted) guidlines you demand, but it is the best
- >emperical basis you can reasonably expect. It shows (to some degree)
- >what happens when real people with real ranks (however they were
- >determined) play games.
-
- The article that Joe Vermaseren wrote on this:
-
- From: t68@nikhefh.nikhef.nl (Jos Vermaseren)
- Newsgroups: rec.games.go
- Subject: loosing at 9 stones and probabilities
- Message-ID: <1484@nikhefh.nikhef.nl>
- Date: 26 Jan 92 14:34:11 GMT
-
- One of the problems about the outcome of a game of go is, that it
- depends very strongly on the circumstances.
- I remember a series of games with a (very strong) 2 kyu (I was 3 dan
- at the time). We would change the handicap if someone would win three
- consecutive games. I got him to 9 stones. Then we played with 5 stones and I
- could not win anymore.
-
- Lightning go is also famous for handicap adventures. One evening it can
- be possible that one player is `moved' to six stones, while the next
- evening the opposite happens. So even though I am 4 dan I am sure that
- a top professional can get me occasionally at 9 stones (when the beer flows
- that is probably to his/her advantage).
-
- This introduces probability, and another discussion brought on transitivity.
- Suppose that the chance that a player at strength x wins from a player
- at strength y is defined as P(x,y).
- It is rather amaizing to me that actually P is measured to be only a
- function of x-y. I have done this measurement with a database of about
- 2000 tournament games a few years ago. Actually if x weaker than y, the
- function P(x-y) = ((1/3)^(y-x))/2 with x and y in units of half handicap
- stones and beginning gives only half a stone more on the board.
- When this function was used for the probability in a giant maximum likelyhood
- fit for all these results the graph of measured strength versus generally
- accepted strength was linear (with some scatter of course) from 5 dan to
- 15 kyu! I had not expected this in advance and it is something I still
- do not understand.
- Soem people claim that a normal exponential cannot be the right function,
- but it does have some very nice properties. And when a 6 dan looses against
- a 6 kyu (without handicap) something funny must have happened anyway.
- The single power in the exponent makes that the results are interpreted
- as `lost against a weak one' or `won against a strong one', but the exact
- difference is not relevant in the limit of a big difference. (In a
- maximum likelyhood fit at least).
-
- Jos Vermaseren
-
- ---end of included text---
-
- r.
-