Absolutely correct - a two day sample has no statistical significance whatsoever. All it can show at this stage is that some folks are running Chrome - wee that is hardly a surprise given the level of interest from software development professionals.