Some people draw conclusions based on analytics alone, and I think that's what the backlash might be against.
A good MD would use both subjective and objective findings to come to a diagnosis. Same goes for someone assessing a hockey player's performance, IMHO
My main issue with analytics in hockey is that I don't really like the ones that use a humans interpretations of an event or what impact it has. That's not scientific.
The industry I'm in uses data analytics heavily, but we use metrics that are scientifically quantified.
Even in my industry, analytics are used selectively to paint whatever picture we want. To our clients, we obviously want to put our best foot forward, but internally, we use them more critically to analyze the real story.
So, I do understand the difference between interpretation of data vs. quality of data.
To be fair to garret, I'm not intimately aware of all of the various datasets used in these charts and graphs - some are likely much better than others as they likely depend less on someone interpreting an event using thier inherent bias and understanding of what they're measuring.
There are other datasets that are missing a lot of context in them as well. A shot from the slot does not = a shot from the slot. There are far too many variables in hockey for that.
I guess what I'm saying is I am very wary of hockey's advanced stats because I feel like they are still in their infancy, and a lot of work still needs to be done. That's not to say they're all useless, but they are hardly at a place where they would be considered scientifically sound.