An approach to the fairy tale card game: a rotating sets competitive knapsack problem with strongly stochastic rewards and item availability
Candice Schumann, Timothy Highley, Howard Stickley · Journal of computing sciences in colleges · 2015
We have analyzed a four-player card game called Fairy Tale. We created an implementation of the Fairy Tale game and several algorithms for playing the game. Comparisons among our algorithms have yielded interesting insights. For example, in some cases a player's particular algorithm choice is less important than simply having an algorithm that is different from the algorithms the other players are using. In the game, drafting a card means selecting a card so that no one else can use it. Players receive sets of cards from the deck and draft a card from each set. Each card's value is determined by a function of which other cards have been selected. Although one may expect that some cards are inherently more valuable than other cards, particularly at the beginning of the game, we have demonstrated that any algorithm for the Fairy Tale game that bases its first draft decision on a static ranking of which cards to draft will not be optimal. The Fairy Tale game that we have analyzed can be characterized as a competitive knapsack problem with 100% turnover in item availability between selections and reward functions where the value of each item is based on the set of items in the knapsack.