This work addressed critical questions related to the optimization of all-important slot revenues, as they are critical to the success of most of the world’s casinos. Using data collected from a live gaming floor, the obfuscated pars of otherwise identical reel slot machines were manipulated to determine whether differences in game-level performance and/or player detection would occur. The results indicated significant and substantial increases in revenue for the high-par games, with no clear signs of play migration toward their paired low-par counterparts. These results were produced in spite of the clear disincentive to play the high-par games. This work extended the extant literature by increasing the time to discover the difference in pars to a full year, and by expanding the difference in… read more.