home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ogicse!emory!wupost!zaphod.mps.ohio-state.edu!news.acns.nwu.edu!casbah.acns.nwu.edu!jzy
- From: jzy@casbah.acns.nwu.edu (Jim Yu)
- Newsgroups: rec.games.go
- Subject: Re: winning ratio
- Message-ID: <1992Nov15.163127.26895@news.acns.nwu.edu>
- Date: 15 Nov 92 16:31:27 GMT
- Article-I.D.: news.1992Nov15.163127.26895
- References: <Bxq5pn.D89@nic.umass.edu> <1992Nov15.142906.30095@dickens.com>
- Sender: usenet@news.acns.nwu.edu (Usenet on news.acns)
- Organization: Northwestern University, Evanston Illinois.
- Lines: 49
-
- In article <1992Nov15.142906.30095@dickens.com> fred@dickens.com (Fred R Stearns) writes:
- >In article <Bxq5pn.D89@nic.umass.edu> s9874203@titan.ucc.umass.edu (s9874203) writes:
- >>What is the expected winning ratio for two players? ... ... ...
- >>Are there such statistics data? Is there a table to list these winning
- >>ratios?
- >>
- >>bn 2k
- >>
- >>
- >
- >Somewhere (I can't remember where now) I saw the following
- >formula:
- >
- >P = 1/2 * (2/3)^(2*d)
- >
- >where d is the absolute rank difference and P is the probability
- >that the weaker player will win.
- >
- >--
- >Fred R. Stearns -- fred@dickens.com
- >Everything's a kludge!
-
- Every time I see a formula, my eyes start to grow (near-sighted). This is
- a very interesting one! I can't wait to apply it ---
-
- *Assume* that I am a 2 kyu (shaking now :-)
-
- Opponent d 2d (2/3)^(2d) P My wins/100 games
-
- 2 kyu 0 0 1 0.500 50
- 1 kyu 1 2 4/9 0.222 22
- 1 dan 2 4 16/81 0.099 10
- 2 dan 3 6 64/729 0.044 4
-
- Conclusion: If a player is of your same rank, he/she wins as much as you do
- (between you two), needless to say; if he/she is one rank above you, he/she
- would beat you 4 out 5 games; two ranks, 9 out 10 games; three ranks, 19
- out of 20 games. That's of course the case of even games.
-
- (Note: Please don't apply the formula to UFO 30k; we will have no dans on
- IGS.)
-
- I know this formula is meant to be an approximation, but the results look
- good to me. Maybe this formula would make people to think again when they
- promote or demote themselves. I wouldn't hesitate to demote myself to 3k
- if bn (the solid 2k who started this interesting topic) beat me 4 out of 5
- even games. :-)
-
- jzy 2k (currently unofficial rank)
-