I don't have a problem with the newer analytics themselves (admittedly I'm still sorting out what all the stats measure). The problem I have is how routinely they're misused and misinterpreted, including this article.
They call Sheahan Detroit's best shot suppressor, then point out he starts more than 62% of his shifts in the offensive zone. So in one sentence they've drawn a faulty conclusion from the statistics and even included the reason why they're wrong.
Very likely Sheahan isn't Detroit's best shot suppressor. When you start the vast majority of your shifts in the offensive zone, that means your opponent has to go a long way to register a shot attempt. Glendening on the other hand, starts in the offensive zone less than 33% of the time. Two-thirds of his shifts, he's starting in front of his own net, so there's going to be a lot more shot attempts from his opponent.
Then in subtler ways, they often use language that draws conclusions not supported by the stats. It's not that Sheahan himself only "allows" 37.7 shot attempts per 60 minutes of play. It means that when Sheahan is on the ice, the other team only gets 37.7 shot attempts per 60 minutes of play. That's it. But people constantly make the logical leap that the statistic says something specifically that a player is doing on the ice. That's a HUGE difference.