The Tao of Gaming

	AJ642 AJ9 T753 7
Q985 53 AKJ JT64		KT83 KQT752 Q94 --
	-- 64 862 AKQ98532

Both vulnerable. Dealer was west, I made a 3rd seat opener of 1H. South jumped to 3C, passed back to me and I doubled back in, figuring I could stand any suit. Of course, I can't really stand 3 clubs doubled, which floated and made. -670 was a zero. (The top was +790, four hearts doubled making). Bidding 3H would have worked out, but it felt wrong to give up on spades or even diamonds. Perhaps partner should bid 3S ... looking only at my hand (East) I'm still not sure.

In the next hand, I opened 1 diamond with a flat 3-3-4-3 13 count, and got to play it.

	AKQ2 T84 Q96 T63
JT976 765 J82 95		543 J32 AK43 KQ4
	8 AKQ9 T75 AJ872

Three tables played in 1 diamond, off three. Getting to game isn't easy at all. I must admit, I'd be very tempted as South to overcall 1 Heart (for the lead direction value), or for north to balance 1 spade. Solid four card suits are biddable. North could also double, which is more flexible...

One poor pair had west make a weak jump shift into spades, and now N-S suddenly know what to do — take the money. Anyway, even among the N/S that bid, most missed their game, although one pair wound up in 4 Hearts (which made five...west must have led a spade). A touch over 4.5 match points.

The last hand was a relatively flat 3N.

Update: Fixed first hand ...

"My insights into Fairy Tale strategy are markedly better than random play." OK, it's not that comforting.

Based on feedback, I'm skipping the Neural Network and just going with a correlation matrix (which will eventually evolved). I've hardcoded my default correlation matrix (80 x 40) and initial weighting of cards (1 x 40). [Player #1 is the non-random one].

 INFO  9859 - GameState - Player #1 won 82 games.
 INFO  9859 - GameState - Player #2 won 9 games.
 INFO  9859 - GameState - Player #3 won 5 games.
 INFO  9859 - GameState - Player #4 won 8 games.
 INFO  9859 - GameState - Player #5 won 3 games.

Right now I'm just using a "Play the card you like best each round", but perhaps I should look at the breakpoints ... if I know I won't be playing two of the cards at all, perhaps theirs a way to let it evolve ways of ordering the rest of the card play.

Now I just have to program a mutator, a serializor and a reaper. Then it's off to the evolutionary races! (Eventually I'll need to write a GUI so I can play against it, assuming it gets good enough ...)

I'll write some more bridge in a few days. And some actual game stuff later...

Or just help it out. If you've done Neural Nets in Java, can you lend a hand?

While I figure out what to do, I'll start writing some test cases for obscure scoring issues for the game I implemented. Or not. (Don't expect a release anytime soon, this is mainly a NN testbed right now. Doesn't even have a GUI yet).