Introduction
In August 2017, the Bitcoin network underwent a significant hard fork, resulting in the creation of Bitcoin (BTC) and Bitcoin Cash (BCH). While both networks share the same SHA-256 proof-of-work (PoW) algorithm, they compete for the same pool of computational resources. This research employs game theory to analyze the long-term stability of such a dual-coin ecosystem, focusing on the phenomenon of "fickle mining."
The Problem: Fickle Mining and Instability
Because BTC and BCH use compatible hardware (ASICs), rational miners can switch between chains to maximize short-term profit. This behavior, termed fickle mining, is driven by the fact that mining difficulty adjustments are often more predictable than market price fluctuations. Miners frequently wait for a "difficulty drop" on the smaller chain (BCH), flood it with hash power, and then exit as soon as the difficulty adjusts upward.
This cycle creates massive instability in hash power, leading to inconsistent block times and, more critically, network centralization. When fickle miners leave, the security of the smaller chain rests entirely on a few "loyal" factions, making the network vulnerable to 51% attacks.
Methods and Empirical Validation
The researchers developed a formal game-theoretical framework to model interactions between three types of players: Fickle Miners (profit-driven switchers), Loyal Factions (entities that stick to one coin for long-term ecosystem health), and Automatic Miners (software-driven switchers reacting to real-time price and difficulty).
To validate this model, the study analyzed historical data from major mining pools like ViaBTC and BTC.com. The data clearly illustrates the inverse relationship between the hash power of the two chains as miners migrate based on profitability.
Key Insights: The Centralization Trap
The core of the research identifies different "Zones" of stability based on the ratio of fickle and loyal miners. The model demonstrates that in many scenarios, the only stable state for the smaller coin is one where only loyal factions remain. This leads to a severe centralization trap; for instance, at one point in BCH's history, just two accounts were responsible for generating 70% of all blocks.
Key findings from the model include:
- The 5% Tipping Point: If just 5% of the total ecosystem hash power engages in "automatic mining," it can be sufficient to drive away all loyal miners from the smaller chain, effectively destroying its decentralization.
- DAA as a Stabilizer: Bitcoin Cash's transition to a per-block Difficulty Adjustment Algorithm (DAA) in late 2017 was a critical move. By making difficulty adjustments more reactive, the network moved closer to a "stable coexistence" equilibrium, reducing the exploitability of difficulty drops.
- Hash War Dynamics: The model accurately described the 2018 "hash war" between Bitcoin ABC and Bitcoin SV, where massive influxes of loyal hash power from major factions temporarily forced the system into states of extreme centralization to survive.
Conclusion and Future Outlook
The threat of fickle mining is not unique to Bitcoin; it applies to any rival coins sharing a PoW algorithm, such as Ethereum vs. Ethereum Classic. The research suggests that for a smaller chain to survive alongside a dominant rival, it must either cultivate a massive loyal following or implement highly sophisticated difficulty adjustment algorithms to mitigate the impact of profit-seeking miners. Future work may explore the use of proprietary hardware or merged mining as potential countermeasures against the instability caused by fickle mining.
Summary updated