With two decks (one for the Player and one for the community cards) five of a kind ranks highest.

See http://www.pagat.com/vying/pokerprob.html

per Yoda.  But subject to that change, the concept is sound.

I had made the decision to put five of a kind below straight flush based on a different website: http://www.homepoker.com/more_poker02.asp?ID=21

But, that one was based on one deck with wilds, not two decks.  I’d be happy to change this proposal if everyone prefers Yoda’s suggestion.

But this isn’t a pure two deck hand. Your hand is one deck and its hand is the other. Now I want to write a program that simulates a bunch of hands and sees which ones have the best chance…

this makes as much sense as anything, unless clucky finds otherwise.

It’s interesting.  Normal poker logic won’t quite apply due to our screwy deck rules.

Another solution if others prefer it is I could code into the GNDT a single shared deck of cards, if someone would tell me what language it’s written in.  That would solve the problems of cheating, multiple instances of the same cards, and save the dealer the trouble of dealing and checking for cheaters.  Just a thought, but I’d be happy to do that.

Yes, simply draw cards, if it can tell you publicly a code, then privately the cards, and then The Dealer can check the codes against deck, or something like that.

Might not need to be in the GNDT, if there was another way.

Where there’s a will…

(...there’s a deed to the farm?)

Assuming I wrote my script right, in the hole cards from one deck, community cards from another deck game, straight flushes are quite similar in commoness to five of a kind.

Five of a kind appears to be a little more rare.

The interesting thing is that the order of rarity of straights, flushes and full house appears to reverse, with a full house actually being the more common of the two.

Its pretty interesting. Someone else should write a sperate program to make sure I was doing it right =P

It works.