Some research could be done
Some research could be done in an attempt to quantify some of the
effects I have mentioned:
- Giving Knight odds should be more serious between stronger
players than between weaker players. Test this with thousand of
games between computers of different strengths whose ELO ratings are
known; organize a small tournament of human volunteers to see if the
results are applicable to human play.
- Weaker players err more often. Acquire a large number of games
between players whose ratings are known, ranging from master level
to beginner; run them through a strong chessplaying computer and
tabulate the frequency of bad moves.
- Larger advantages produce larger winning rates. Using several
different strong chessplaying computers, and thousands of games,
find winning rates for the standard opening position (no book),
Pawn-and-move, Pawn-and-two, ..., Knight. Try again with some book
positions evaluated += and +-.
And In Closing, May I Say
Looking for a research project?
Next Section
Other Links In these Pages
This is a Mailme.