In another thread (now closed), there was reference to an article by Dellow that cited how bad the Jets 3rd line was last year in goal differential.
I had a quick look at Corsica data to investigate, and thought the results were interesting.
Here are the on-ice goal differentials (5v5) for lines centered by Jets' centres last season. Note, for Perreault and Burmi I only looked at the data for the lines on which they played with two wingers, to isolate lines on which they played C. Occasionally, it was ambiguous (e.g. Copp with Burmi), so I omitted those data from the summary.
5v5 on-ice goal differential (2016/17):
Scheifele +15
Little -1
Lowry -2
Copp -2
Burmi -5
Perreault -11
What's clear is that a lot of the poor goal differential 5v5 happened when Perreault was at C. But GD only tells you what happened, and isn't a very good way of forecasting what will happen. So I dug a bit deeper into Perreault's role as a C and looked at the total raw Corsi (5v5) for line combinations where he was C.
Perreault at C: CF 177 CA 210 CF% 45.7
Given Perreault's remarkable performance on the wing, clearly the most important message is that Perreault should not play C.
Oh, and Little needs a bounce-back season if the Jets are going to make a lot of noise.