How is it possible for chesss.com to stop low level players cheating with sub optimal moves

If a low level player used some kind of an engine which shows several options to them in a parallel game, and they select sup optimal moves and use them sparingly, how is it possible for chess.com to detect their cheating?   And would not this be of massively disproportionate benefit to a cheater at lower levels compared to higher levels because sub optimal moves are often much better than standard moves at say 700 - 900.

For example, instead of playing the first or second recommendation, they use the third frequently, and use this somewhat sparingly for when they get into trouble. How would chess.com reliably be able to detect this cheating?

I am not a good player, but I find on chess.com compared to Lichess (1) my rating is 500 points lower in 10 minute rapid, and (2) My opponents respond in the same 2 or 3 ways in almost all of my games on chess.com for the first 10 moves or so even when I play fairly uncommon openings, and on Lichess there is way more variability in how my opposition responds.  

The rating part doesn't really matter to me, but I find the lack of variability really boring, and it has become substantially less fun playing on this website, so I am wondering if there is actually a way of them stopping this kind of thing happening at lower levels. 

You can't discuss cheating in the regular forum.

Join the Cheating Forum club if you want to discuss.