It’s a common sight on casino floors. Customers run from slot machine to slot machine and end up jumping into the next big-money game. But can players, even regular users of certain properties, really tell the difference between the edge of any game’s home and the edge of another game’s home?
No, at least not according to a series of recent studies led by Anthony Lucas, a professor at UNLV Hospitality College and former game industry operations analyst.
Over the past few years, Lucas and his colleague Catherine Spield of San Diego State University have gone to casino floors of several properties in the United States, Australia and Mexico to investigate. Their results contradict the casino operators’ long-held belief in the player’s ability to detect differences in the amount and frequency slot machines pay.
“I think some operators are naturally and understandably cautious about new information that challenges traditional industrial practices,” Lucas said. “But we have to consider how we know what we know. Our work here has a Moneyball-like aspect, questioning the wisdom of widely accepted belief when the data shows that new ways of thinking can be better.”
In a recent study, a team led by UNLV compared two pairs of reel slot games at a “local” casino in the suburbs of Sydney, Australia, where all bets come from electronic gaming devices. 바카라
Their process is relatively simple. Choose two identical slot machines located in a similar location on the casino floor, but they change the par, which is the percentage of the total coins the machine maintains over time. If the game’s par is set at 10%, for example, the machine is expected to retain $10 of its $100 bet on average over the long term. However, this rarely happens in the short term, adding to the difficulty of par detection.
For this study, the researchers compared the daily pairing performance of “Tokyo Rose” and “Dragon’s Fortune X” games for nine months. The wave number within each pairing varied from 7.98 percent on the lower side to 14.93 percent on the higher side.
The researchers measured the coin input for each machine every day, as well as T-win, a formula that multiplied the coin input by the wave to calculate the machine’s expected value, or T-win, a theoretical victory. If ordinary players could detect the difference in the wave over the course of the nine-month test, the comparison would reveal whether (and how much) the players moved from higher to lower wave games.
As predicted by Lucas, the difference between high and low-par games remained stable during the study period, meaning there were no statistically significant signs of play movement.
And while the lower wave machine held more coin-ins during the study period, the T-win was larger on average for the upper wave machine.