Journal Article4 February 2026 Robert Scott, Mikhail Sher, Jonathan Daigle
Crash games are a new type of casino game offered on some online gambling sites.
The first crash game was created by the online cryptocurrency casino Bustabit.
Other cryptocurrency casinos started offering their own versions of crash games.
Now mainstream online casinos have developed their own crash games—notably
Rocket by DraftKings. Crash games are easy to learn and play. They offer the
opportunity to increase your bet by many hundreds of multiples—though, as we
show when deriving expected values, these outcomes are rare. In this paper, we
study the history of crash games and calculate theoretical values of crash game
bets. We then use data from two popular online crash games to study the
behaviors of players and return to player (RTP) percentages.Journal Article12 February 2026 Michael Alexeev, Peter Chi
In casino games using a standard deck of cards, a wash shuffle is sometimes
performed prior to the rest of the card shuffling procedure. Unlike other
methods of shuffling, the wash shuffle has not yet been well studied. To this
end, we first develop a mechanistic model of the wash shuffle based on our
observation of how cards tend to move when a wash shuffle is being performed.
Then, we use this model to simulate the card shuffling procedure used in casino
poker rooms, and explore the resulting impact on where the cards are dealt in
the context of Texas Hold’em poker. We find that while the addition of the wash
shuffle does produce probabilities in line with what would be expected from a
perfect shuffle in terms of where individual cards will end after the shuffling
is performed, it still shows deficiencies in terms of breaking up clumps of
cards that have started next to each other, which has significant implications
in games like Texas Hold’em in which several cards are dealt face up.Journal Article23 March 2026 Anthony Lucas, A.K. Singh
Modified pay tables from a licensed reel slot machine were altered to examine
whether two games with equal variance and par could produce meaningful
differences in the duration of play, at the level of the individual gambler. To
the best of our knowledge, this was the first study to examine this possibility.
Industry executives consider time on device (TOD) to be a critical measure of
gaming value, especially for losing players. Additionally, TOD experiences are
theorized to produce knock-on effects related to loyalty behaviors. The results
of 18 simulations featuring six different player engagement protocols supported
the ability of differences in the allocation of pay table variance to
significantly impact the spins per losing player. These outcomes indicated that
games with the same par and total variance should not be assumed to produce the
same expected TOD. For operators, the findings suggest that more
meaningful/precise pay table descriptors could help them populate slot floors
with games that are more likely to produce their desired TOD outcomes. For game
makers, the results present opportunities to supply games that improve the
outcome targeting of individual player experiences. For decades, par has been
the primary focus of pay table metrics and gaming value. But par is a long-term
measure of central tendency, with multiple studies indicating that differences
in pars do not generate significantly different session-level results. This
study, along with several others, indicates the considerable ability of variance
to impact the player’s visit-level outcome.