A rare two-block Bitcoin chain reorganization has brought fresh attention to one of the network’s most widely cited principles: the six-confirmation rule. On March 23, at block height 941,880, Foundry mined six consecutive blocks while AntPool and ViaBTC briefly extended a competing branch. In the end, Bitcoin behaved exactly as designed, with the chain backed by the greater hashpower winning out.
Still, the event exposed an important weakness in how the market talks about transaction finality. The six-confirmation standard has long been treated as a rule of thumb for safety, but its logic originally depended on assumptions from Nakamoto’s model, including a scenario where an attacker controls only around 10% of total network hashpower. Under those conditions, reversal risk after six confirmations remains very low.
That picture changes when mining power becomes concentrated in a small number of pools. According to the figures cited in the report, with Foundry holding a 32.2% share, the modeled reversal risk after six confirmations rises to roughly 18.9%. At the same time, the top three mining pools account for around 60% of total block production, making finality assumptions far more sensitive for large-value transfers.
Mining economics are also adding pressure. On March 21, Bitcoin difficulty dropped 7.76%, February hashprice averaged $32.31 per PH/day, and transaction fees accounted for just 0.57% of block rewards over the last 24 hours referenced in the article. When margins get tighter, smaller miners have stronger incentives to join large pools, which can further increase concentration.
This does not mean Bitcoin has become broken or that mining pools are inherently malicious. What it does mean is that the six-confirmation rule should no longer be viewed as a universal guarantee. For ordinary retail payments, it may still be sufficient. But for exchanges, OTC desks, and institutions handling high-value transactions, confirmation requirements may increasingly need to reflect both transaction size and the real-time distribution of network hashpower.
