When Should Selfish Miners Double-Spend? (2501.03227v1)

Abstract: Although, both double-spending and selfish-mining attacks have been extensively studied since the Bitcoin'' whitepaper of Nakamoto and themajority is not enough'' paper of Eyal and Sirer, there has been no rigorous stochastic analysis of an attack that combines the two, except for the complicated MDP models. In this paper, we first combine stubborn and selfish mining attacks, i.e., construct a strategy where the attacker acts stubborn until its private branch reaches a certain length and then switches to act selfish. We provide the optimal stubbornness for each parameter regime. Next, we provide the maximum stubbornness that is still more profitable than honest mining and argue a connection between the level of stubbornness and the $k$-confirmation rule. We show that, at each attack cycle, if the level of stubbornness is higher than $k$, there is a risk of double-spending which comes at no-cost to the adversary. The result can be seen as a guide for picking $k$ in the $k$-confirmation rule in a blockchain design. At each cycle, for a given stubbornness level, we rigorously formulate how great the risk of double-spending is. We provide the minimum double-spend value needed for an attack to be profitable in the regimes where the scheme is less profitable than honest mining. We further modify the attack in the stubborn regime in order to conceal the attack and increase the double-spending probability. Finally, we evaluate the results and provide the optimal and the maximum stubbornness levels for each parameter regime as well as the revenue. As a case study, with Bitcoin's $k=6$ block confirmation rule, we evaluate the revenue and double-spending risk of the attacks for each pool parameter.