Yeah, in this case only data can suggest which would be the best UX route. Some A/B testing around percentage of visitors who come to the home page and login wi

Yeah, in this case only data can suggest which would be the best UX route. Some A/B testing around percentage of visitors who come to the home page and login with either layout (some permutations of Remember Me checked/unchecked, Remember Me above/below login) would be interesting. Perhaps instead focusing on login pages and how they layouts perform there (vs. the home page) might be more precise because I wonder about all of the other noise that may come into play on a home page and if that noise could interfere with the precision of the testing.