I’d say this. An AF ratio gauge should be used as a reference tool to tune. It’s not the be all and end all and in the end, it’s only as accurate as it’s tolerances.
And they are only accurate as long as the flame is burning as expected/predicted.
Too bad that MSV post by shrinker is gone. But this was one of the things he showed. If the burn, or even part of the burn is cold for example, then the products of combustion result in left over O2 that is different than a good complete burn.