We have a tendency on this forum to lean heavily on evaluating trades based on the time that they were made and downplay the hindsight evaluation years down the road. I think you have to evaluate them at multiple intervals. Rarely is a trade ever going to look bad when it's made (some certainly do), you can almost always see the justification for trades in the moment. But you also have to look at them a year down the road, two years, five years, ten years...
I mean, if you hired an investment firm, and they lost all of your money, would you give them a pass because the stocks looked good when they bought them?
So we can say both that at the time the Turris deal looked like the right trade, and we can also say that right now it looks god awful as Colorado has Girard and we're paying Turris $2M/season to play somewhere else.
We can look at the Jones for Johansen deal and say at the time it was a very good trade for us to make, and we can also say that the further away from the deal we get the worse it will look, unless Joey finds a way to consistently produce at his highest levels.
Giving up Subban to NJ has potential to go the other way. It looked kind of risky at first, but I think with his decline (even climbing back a little bit) and we ended up taking Afanasyev and Campbell with the 2019 2nd, both of whom could have an NHL impact, Afanasyev especially. We used the other pick to help land Kunin, so there's some waiting to see how he turns out there. Plus Davies could still end up in the NHL, though I think the hype has really died down on him. Finally, it opened up room to sign Duchene, which is a chapter that is still being written. But this trade probably ends up looking really good in the long run.
As with most things in life, there isn't a single bullet point that this topic can be narrowed down too. Not everything fits into 140 characters (or is it 280 now?). We (not just HF) need to get away from distilling everything to black and white, right and wrong, this or that, one extreme or the other. Hockey trades are complex issues that need the perspective of multiple point in time to truly evaluate.