Even if you don't taste the sweetness difference in these beers, you note the body difference. It may not be pronounced in a 2 oz tasting you do while drunk and eating nachos in a loud pub, but then the same could be true for plenty of moderate differences between beers, which is exactly why I consider XBMT data spurious.
It's dangerous (to the extent that bad brewing advice can be dangerous) misinformation because the conclusions are presented in a way that indicates they can be relied upon and the ubiquity of this mostly bad advice-giving blog causes tons of not-quite-novice brewers to brew worse beer. Read any of the Reddit comments on these suspicious XBMTs and you will see plenty of brewers announcing, "wow, I'm done building starters" or "now I know I shouldn't bother controlling fermentation temperature." Even if these XBMTs are meant to be taken with a grain of salt, they are treated like gospel because they present bad data as though it was bona fide conclusive scientific evidence.