[lamford]

In online chess and backgammon, two games I also play, the strongest computer is quite a bit better than the strongest human, but the strongest "team" of all is a combination of man and machine. I had a discussion with FIDE in Belgrade a year ago about how to run an online World Chess Championship and the only method we could think of was to have, say, the last 64 played in hotels with an arbiter and spectators, so that Magnus Carlsen might be on the stage in Oslo playing Fabio Caruana in a hotel in St Louis. That way one can stop computers being consulted. A pre-game strip search, looking for communication devices, is also needed.

Collusive cheating is almost impossible to stop in bridge, unless one has a TD in each home, and the hard-line cheater will get a "message" from either his partner or a confederate that has seen the hand on Bridgebase, or played it already in a tournament or match. Online bridge is not the only form that is vulnerable to this. In the European Championship in Ostend, one team bid a slam against us off two key-cards, and the person who bid it knew they were off two key cards. Of course the missing key card was the king of trumps and the finesse was right. The TD investigated but the player just said that he thought that they were doing badly in the match and he wanted to gamble. Nothing could be done. If one compares other sports, I would expect that the cheating incidents we have "proved" so far in bridge are just the tip of the iceberg. The estimate in cycling and athletics is that 5% of participants cheat. I would be surprised if it was any different in bridge.

Entering a hand into K-R, or a better future version, or doing a SIM with Bridge Analyser can easily be set up so that it is the push of a button. When computers are strong enough, compound trump squeezes will be found instantly and the exact percentage of each line will be calculated immediately. When a psychological line is needed, the human will step in. One can plug in the information from the auction and find the percentage opening lead. FIDE's approach to chess is wrong. You can be banned if your moves have "too high a similarity to those chosen by Stockfish, the strongest commercial program". I don't think this is right at all and will lead to miscarriages of justice and some of the bridge prosecutions have been based on inadequate statistical evidence and a misunderstanding of Bayes' theorem. You don't prosecute Mrs Brown of Neasden for fraud when she wins the lottery.

"It is better that ten guilty persons escape than that one innocent suffer".- Commentaries on the Laws of England, published in the 1760s.

[hrothgar]

This topic got beaten to death on the forums about 15 years back

You can probably find some of the old threads is your google-fu is strong enough

[KingCovert]

I think it would be rather easy to write a program that allowed for descriptions of hands, based on the bidding, and then the refinement of those descriptions as cards became known in order to provide analysis on the lines of play that worked best on average. It's basically an extended double dummy solver. Definitely doable, and it would be an effective learning tool in addition to it's obvious potential usage for cheating purposes.... Sadly.

I've coded up a hand generator before that would be able to randomly generate hands based on specifications. Wouldn't have been much of a leap to take the extra step and do this I guess.