Newsgroups: comp.ai.games
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!hookup!swrinde!pipex!warwick!bsmail!irix!mc4391
From: mc4391@irix.bris.ac.uk (Martin Craig)
Subject: Re: Is Connect4 a theoretical draw?
Message-ID: <D5J4yK.AMD@info.bris.ac.uk>
Sender: usenet@info.bris.ac.uk (Usenet news owner)
Nntp-Posting-Host: irix.bris.ac.uk
Organization: University of Bristol, England
X-Newsreader: TIN [version 1.2 PL1]
References: <3k8v98$itp@rockall.cc.strath.ac.uk>
Date: Thu, 16 Mar 1995 10:54:20 GMT
Lines: 19

I.H.Cornwell (cnas029@ccsun.strath.ac.uk) wrote:
: I just wrote a "connect-4" program, a fairly naive minimax implementation.
: It plays quite well despite looking only 5 half-moves deep. 

: I suspect connect-4 is a theoretical draw - can anybody confirm this?

: Ian Cornwell, cnas029@ccsun.strath.ac.uk

	Nope - person going first has a theoretical win.
	I read about this in a newspaper a while back - somebody
programmed a computer to play every possible game and worked out a perfect
winning strategy if it went first. If it went second, the opponant had to
play perfectly to beat it.
	Unfortunately the first time they tried it playing against itself,
the second mover won! However I think this was some programming error and
apparently they fixed it.

	Martin.

