They should (in advance) post the exact algorithm by which they choose the pseudorandom results, and (on lottery day) post a truly-random seed selected by a pro

They should (in advance) post the exact algorithm by which they choose the pseudorandom results, and (on lottery day) post a truly-random seed selected by a process performed in public or verified by independent observers. Then anyone would be able to verify the lottery results after the fact.