Distributed gaming apps are a new way for people to play online. Distributed gaming programs differ from traditional online casinos in that players use cryptocurrencies as items. Furthermore, instead of being stored on a central server, distributed gaming applications are stored on a cryptocurrency blockchain.

Previous literature on player behavior monitoring has examined gambler spending patterns in traditional online casinos. However, similar work has not occurred in the scattered play areas. Therefore, the profile of crypto-gamblers on distributed gaming programs is unknown. This article discusses 2,232,741 transactions from 24,234 unique addresses in these three applications that worked on Ethereum cryptocurrency online in 583 days.

We find that typical gamblers spend around $ 110 the equivalent of a median of 6 bets in one day, although highly engaged bettors spend around $ 100,000 the equivalent of a median of 644 bets in 35 days.

Our results show that the average player of a distributed gaming app spends less than other online casinos in general, but those interested in this new domain spend much more. This study also demonstrates the use of this application as a research platform, particularly for large-scale in vivo data analysis.

Hypothesis

Previous work on the analysis of the gaming business in life varies according to the type and characteristics of the group [5]. Given that neither of them focused on the use of distributed gaming programs of any kind – other than the work of Gainsbury [1], which uses advanced usage statistics – we expected to find gaming behavior consistent with casino player analyzes similar to those used by them are described in detail below.

We attach a table from one of those studies to Labrie et al. In the Appendix for a quick comparison. Secondly, we do not expect the data collected directly from the Ethereum cryptocurrency network to be useful for researching player behavior without first implementing a data cleansing method. Of particular importance is the potential presence of non-human players, known as robots, in the dataset.

Robots can be here for a variety of reasons – for example, to artificially increase the popularity of the program they trade, or to try to remove a jackpot from an app where it’s statistically worth pursuing. We cannot infer the reasons behind the existence of robots, but we can build evidence to identify their presence by estimating how different the “player” behavior of each game is and the behavior of other similar games.

Study presented

This work describes the behavior of a large group of users of a game program distributed in 583 days, from the conclusion of the smart contracts for all programs until March 9, 2020. Using cryptocurrency data collected directly from the Ethereum cryptocurrency blockchain, we can calculate behavioral metrics using individual mortgage rates versus total amounts of some kind, e.g. weekly daily.

Behaviors are described, including a typical player description (median) of each game offered through each app. We performed four separate analyzes after identifying potential non-human players:

  • Statistical comparison of player-player behavioral measurements;
  • An epidemiological description of the gaming behavior of (human) distributed gaming programs;
  • A statistical evaluation of the relationship between the actual behavioral measures of actors in this new field.