Pseudo-Random Number Generators in Slot Machines: Balancing Randomness and Profitability

Authors: Lola Melis, Irina Shablinsky

SUNY Campus: Purchase College

Presentation Type: Poster

Location: UU 111

Presentation #: 65

Timeslot: Session C 1:45-2:45 PM

Abstract: Slot machines, a cornerstone of the gambling industry, rely on Pseudo-Random Number Generators (PRNGs) to create the illusion of chance while ensuring profitability. This study examines the mathematical and computational foundations of PRNGs, focusing on the Linear Congruential Generator (LCG) and the Mersenne Twister (MT). While the MT demonstrates superior randomness in statistical tests, the LCG remains widely used in slot machines due to its efficiency and regulatory acceptance. Using the Dieharder test suite, this research evaluates the statistical properties of both PRNGs, revealing that while the LCG fails multiple randomness tests, it is still sufficient for slot machines since cryptographic security is not required. Instead, casinos employ PRNGs to balance unpredictability with profitability, ensuring that outcomes appear random to players while maintaining the house edge. The research explores how probability distributions and expected value calculations influence payout structures, demonstrating that even statistically weaker generators can effectively serve the industry's needs. Beyond technical analysis, this study also considers the ethical implications of randomness manipulation in gambling. By examining the intersection of mathematics, probability, and business strategy, this research provides insight into how casinos engineer chance to maximize profits while sustaining player engagement. Understanding PRNGs in slot machines offers a broader perspective on the role of randomness in commercial applications and risk management.