You don't need "perfect randomness". The claim was about "perfect randomness". You just need a non-zero probability for the outcomes in question. And you nee

You don't need "perfect randomness". The claim was about "perfect randomness". You just need a non-zero probability for the outcomes in question. And you need independence between subsequent random events.