Then it's even less likely that a 14 vs 16 will matter. Oh, that and many of my combats may last longer ... or maybe I just don't pay attention.
In any case, I think a 5% difference is so small it will rarely matter or be noticed. Other people disagree. Without a complex combat simulator that doesn't oversimplify things like spreadsheets tend to we'll never have a concrete answer, if that's even possible. Analysis is frequently way off base. Not that it's proof of much, I've been playing Solasta (D&D video game) that tells you statistics when you save a PC at the end of a module. According to a guide's spreadsheets, the paladin in awesome, the fighter is crap. My games? The two weapon champion fighter is second only to the wizard in overall damage and because they can second wind have just as much staying power as the paladin while doing 40% more damage.
In other words, I don't see a lot of value continuing the "I'm right you're wrong" when I don't buy the simplified analysis most people apply. Play a few hundred games with a variety of groups and track detailed statistics and we can discuss again. Until then? Maybe I'm wrong, maybe I'm not. Either way I have a preference for balanced PCs most of the time.