Well you heard it here. Stats are in no way indicative of current or future performance. They’re also made up and inaccurate.
I don’t know how one would make up the stats I quoted. It’s just being on the ice when a goal is scored divided by the ice time. Not complicated. Whatever. You’re free to disagree, but the numbers are what they are.
That isn't what I said. At all.
I said you are using that data as a faulty tool. It doesn't say anything at all, its just a collection of numbers that you are choosing to interpret in a manner that is in no way realistic.
Stats are interesting. That's it. They are not a predictive tool, they don't understand context, they don't offer an opinion. Its just a collection of data. The problem is their constant, consistent misrepresentation. Its become a lazy way of trying to justify something, proof of an opinion.
By no means am I singling you out here. A couple of weeks ago another poster tried to use data to suggest that Matt Roy wasn't really a second pairing defender despite the actual fact that he has been deployed as one for three years now. He used "quality of opposition" to show that he faced a lower rated group of players to justify that opinion. Well, no shit. A 2nd pairing defender is going to play fewer minutes against top opposition, especially when the top pair defender on your team leads the universe in time on ice for over a decade. That stat meant nothing, but it was used to back up an opinion.
Dustin Brown is the 4th best PP scoring forward according to the info you posted. Cool. But does it mean that nobody else would do better given the same opportunity? No, not at all. Its just collected data interpreted poorly when the reality is that his spacing, positioning, passing, shooting and puck recovery skills have only resulted in 2 points in 64 games - all of which he played on the PP.
The game is played differently depending on the circumstances. Home, road, leading, trailing, different linemates, different times, back to backs, long layoffs, all kinds of variable conditions in which players do different things under similar circumstances. You will need oddly specific data to quantify those situations, all independent of each other, then filtered by what is still a very human personal opinion to reach a conclusion that is almost always just a digit or two after the decimal to separate the wheat from the chaff.
I find the numbers interesting, but gee whiz, they are far from accurate tools to decipher what's going to happen next.