Cool initial look at the problem space. I think a Nash HearthStone deckbuilding solution is very possible, but it's the definition of building to one client giv

Cool initial look at the problem space. I think a Nash HearthStone deckbuilding solution is very possible, but it's the definition of building to one client given the aggressive legal stance Blizzard tends to take. The best commercial chance would be to try to sell it as a balancing tool so devs could make sure all classes stay viable for new releases. Given board + life + hand information is clear, alpha beta for the gameplay strategy combined with a simple genetic algorithm for deck peturbation, initialized with the best decks from current players, could probably be superhuman competitive.