This article doesn't really say much, does it? Averaging fuel consumption in daily driving isn't a very good testing method because of variations in conditions.

This article doesn't really say much, does it? Averaging fuel consumption in daily driving isn't a very good testing method because of variations in conditions. It's really not that difficult of a decision for me. According to the engineers that designed and built my engine, I should use fuel rated at 91 octane or better. Why should I presume that I know better than those engineers? Imagine if you, as an engineer, told a client they should not use a particular technology because it is unsuitable. Then, imagine the client flatly ignored your suggestion. That'd be pretty frustrating, right? So why do that with the advice of automobile engineers?