I simply believe that if some skill is used every few seconds in game, random observer can easily come to conclusion that *this* specific skill is liked and concidered fun, as it is ... and therefore you can afford to give it lesser priority, than other things. If that same skill is never used in few thousand playthrough ... the same observer can as easily come to conclusion, that there might be something wrong *this* specific skill ... and therefore they should investigate it futher.
Simply believing something doesn't make it true, and in this case your interpretation of how telemetry works is incorrect.
Let's use a more egregious example. I (and many other players) long rest frequently because otherwise we'll miss out on cutscenes. Thus, telemetry shows that players long rest frequently. Using your arguments, this is grounds for concluding that players like being at full resources all the time and would encourage Larian to automatically restore all HP, spells slots, etc after each battle. However, this is NOT the reason why many of us long rest so often. A lot of us would much prefer having cutscenes not tied to long resting, and thus being able to long rest less frequently without missing content.
Also, your proof threshold for finding something is wrong with a mechanic is impossibly high. Some players will inevitably use shove, but that doesn't mean it's widely liked or that an alternative option wouldn't be better. Especially for a game in EA where we're explicitly meant to test out all mechanics/dialogue/parts of the game, so simple use of a mechanic doesn't mean its good.
