People rarely go out of their way to put in positive review (or at least neutral reviews, "works as expected"). The current cheats implemented by apps probably affect the ratio of positive reviews per user a lot, making it indicative only of how effective the cheat is, and not the quality of the app.

I wonder if you could get away with only negative reviews? Would a "negative review per total install" metric be worse than "positive reviews per review" (e.g. current ratings)?