If the system suspects a cheater, maybe it can just match them up against a GM (or beyond GM) bot to confirm?
Of course the system could go ahead anyway, displaying a fake rating - within the search criteria - for this GM / bot, but given that there are always false positives (and it would have to be done for merely suspected accounts, as you said yourself), it would essentially mean deliberately pairing some innocent players against, well, effectively cheaters. Which kind of defeats the core purpose.