If one power bank has independent probability p of failure, then n of them have probability 1 - (1 - p)^n of failure. What you're saying is equivalent to claim

If one power bank has independent probability p of failure, then n of them have probability 1 - (1 - p)^n of failure. What you're saying is equivalent to claiming that this quantity is somehow independent of n.