The weasel word beneficial. "Non-beneficial" iterations are also saved. I stated I did not accept the "velcro" concept.(due to serendipity). I clearly stated that. That is not disingenuous and the calculation nevertheless describe the universe being discussed. And that is just a 52 item sequence(of course there are 52 different cards and only 21 AA).
Okay let us calculate a particular protein sequence of 52 AA. That calculation involves a universe of 2152. That is 52*log(21) expressed in Log10 which is 68.7554033. That is compared to the shuffled deck which is log(8.06581752) + 67 = 67.9066484. Now tell me how many proteins are useful, realizing that the tertiary structure is what makes the proteins useful.
You know that beneficial iterations are saved. You then immediately return to citing statistical methods where beneficial iterations are not saved. Do you even realize what you are doing?