The analysis focuses primarily on the decentralization of the governance of EOS and the presumed tendencies towards centralization that could result in the loss of the benefits of a truly decentralized blockchain, for instance, resistance to collusion.
In particular, the report focuses on the high market capitalization of EOS and the fact that the largest EOS owners are further consolidating their positions and risk ending up holding all the power.
EOS is a delegated consensus network based on Proof of Stake, which means that those owning large amounts of tokens may affect governance.
Binance Research’s hypothesis suggests that the decentralization of the governance of EOS can be assessed by observing some specific parameters, such as the performance of the blockchain in terms of collusion resistance, error tolerance and resistance to attacks.
With regard to resistance to collusion, the report states that the governance of EOS actually reinforces consolidation, leading even to the promotion of vote-trading and selfish acts, since it lacks mechanisms to avoid or structure the trading of votes. In addition, individual parties, such as proxies or Block.one, appear to have sufficient power to influence votes.
As far as error tolerance is concerned, it was observed that two-thirds of the exchanges operating as block producers (BPs) had the absolute worst performance among all 21 BPs, in terms of reliability and ability to react, while two groups appear to have formed among the BPs in terms of resistance to attacks, which can be seen from the correlations between voting patterns and regional distribution.
In total, two incidents where errors occurred were detected.
The conclusion reached by Binance Research analysts is that the problems of EOS seem to be caused and aggravated by a number of issues such as low turnout, low resistance to Sybil attacks, lack of consistency, the 1 token and 30 vote system and modified block rewards.