Video poker with persistent bonuses - Mathematics
Session Title
Session 1-3-D: Cards and Dice
Presentation Type
Event
Location
Caesars Palace, Las Vegas, Nevada
Start Date
28-5-2019 1:45 PM
End Date
28-5-2019 3:10 PM
Disciplines
Probability
Abstract
Action Gaming has created a number of new video poker variations that contain persistent bonuses. These are bonus features that are available through multiple rounds of play and whose value and frequency depend on decisions made by the player. A typical example of such a game is the popular Ultimate X Bonus Streak game. The optimal strategies for such games are obtained by modelling the game as a controlled Markov chain (Markov decision problem), but there are technical challenges created by the very large state-spaces in these chains. This presentation will briefly cover the mathematics of analyzing these games, but will mostly discuss how one can determine the optimal hold strategy for some very specific dealt hands, and how this strategy will change due to the current state of the persistent bonus. The presentation might also discuss a new persistent-bonus game currently under development for which the optimal holds are completely counter-intuitive to one's common sense, at least until the mathematics leads one to see hidden value in a feature of the game that is not obvious at first glance.
Keywords
Markov chain, video poker, Ultimate-X, optimal play
Funding Sources
Action Gaming, Las Vegas NV, funded the original investigations into the games discussed. Action created the game concepts and I and three colleagues computed the optimal game returns and determined the final pay-tables and multiplier schedules.
Competing Interests
None.
Video poker with persistent bonuses - Mathematics
Caesars Palace, Las Vegas, Nevada
Action Gaming has created a number of new video poker variations that contain persistent bonuses. These are bonus features that are available through multiple rounds of play and whose value and frequency depend on decisions made by the player. A typical example of such a game is the popular Ultimate X Bonus Streak game. The optimal strategies for such games are obtained by modelling the game as a controlled Markov chain (Markov decision problem), but there are technical challenges created by the very large state-spaces in these chains. This presentation will briefly cover the mathematics of analyzing these games, but will mostly discuss how one can determine the optimal hold strategy for some very specific dealt hands, and how this strategy will change due to the current state of the persistent bonus. The presentation might also discuss a new persistent-bonus game currently under development for which the optimal holds are completely counter-intuitive to one's common sense, at least until the mathematics leads one to see hidden value in a feature of the game that is not obvious at first glance.