New Research Shows How a Few Addresses Shape Blockchain Game Economies

Table of Contents

• Abstract
• Introduction
• Background
  • The Sandbox
• The Data
  • A: Transaction Graph
  • B: Bow-tie Model
  • C: The Sandbox Support
  • D: The Sandbox Scandals
  • E: Whales
• Methodology
• Results
• Related Work
• Conclusion
  • Resource Contributions
  • Acknowledgements and References

Related Work

The bow-tie model has previously been used to analyze the Steemit blockchain [26], where they found that the SCC made up a comfortable majority of the transaction network at 56.3%, similar to our 59.25%. The OUT group was the second largest at 43.7%, which is also similar to our 40.13%. They found that the third largest group was the IN, with a significantly lower number of nodes than in the SCC and OUT groups at 0.016%, like with our 0.62% IN group, and the rest of the groups were insignificantly small. Overall, the groups of their bow-tie model were very similar to ours, so this could be a common pattern seen in more blockchains. They also analyzed the whales of the Steemit chain and found that they held a significant amount of power, which could also be a common pattern. It is clear that this analysis strategy has its merits and could be applied to several different blockchain dApps, as we have been able to apply similar analysis strategies to The Sandbox. We have also possibly found patterns that could be universal across blockchains, so more research could be done to see if this pattern holds.

![ Figure 7:Graph showing the number of transactions per day in the 30 days before and after each of the scandals explored.

](https://cdn.hackernoon.com/images/KX9Ud2ILyvWlxGDDMx5fRwBGTZJ2-by0345v.png)

Li et al. use the Steemit blockchain transaction graph to measure how decentralized it is and analyze its bot activity [27]. Their results reveal that a few powerful addresses in reality controlled the network, similar to our analysis of whales in The Sandbox. Additionally, they also found significant bot activity abusing the reward system on Steemit. While this analysis differs from our paper’s approach, it shows the significant power that a blockchain’s transaction graph holds Fig. 11. Stacked bar graph showing the number of transactions per day in the 30 days before and after May 26, 2023, when the Twitter account of the CEO of The Sandbox was hacked and used to post a crypto scam. The transactions are split into three groups depending on whether they are part of the SCC, OUT, or IN groups of the bow-tie model. for analysis.

Guidi et al. analyzes Decentraland, another popular GameFi project where users can buy land in the form of NFTs [28]. Here, they analyze how the NFT market affects the in-game world and how users interact with the game itself. They find that the profitability of the platform works against its playfulness. Once again, this paper analyzes a different aspect of a GameFi project, showing much potential for analysis in these projects.

In [29], Lai et al. analyze the transaction graph of Axie Infinity to discover player behavior patterns and measure the effect of whales on the game. They found that, while many users were actively playing the game, there was a worrying amount of hoarding of game assets by a small number of addresses, which they believed could threaten the game’s future. This is similar to our analysis of whales in The Sandbox and their significant influence, and we additionally once again see the power of the transaction graph.

![ Figure 13:Graph showing the number of daily transactions involving current Sandbox whales in the 30 days before and after each event explored.

](https://cdn.hackernoon.com/images/KX9Ud2ILyvWlxGDDMx5fRwBGTZJ2-0b734nm.png)