We only have musk's word that tag was used to measure visibility. Wouldn't a random sample with users of different sizes and types make more sense for that? Ev

We only have musk's word that tag was used to measure visibility. Wouldn't a random sample with users of different sizes and types make more sense for that? Even if it was, that could have the same effect. If they test every change and only keep the ones that don't harm Musk's visibility that is in effect making the algorithm consider Musk specially.