Jim,
The assumption of my statement is that a tournament can generate reliable statistical results about each program's strength. I think it's true to a very large extent. The problem hapens when the strengths of two programs are very close. If a program is much stronger or weaker than GnuGo, it should have no problem.
Daniel Liu
_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/