Arm has some structural advantage in the decoding stage (instructions have the same length) and a huge lead in low power systems they got by heritage, not to me

Arm has some structural advantage in the decoding stage (instructions have the same length) and a huge lead in low power systems they got by heritage, not to mention less historical baggage (x86_64 cpus still have a 16 bit mode and a 80 bit fpu) I think the lead the M1 has is bigger than what can be attributed to the different node.