This is completely fabricated. Jan 24 was the line switch. Final 40 games of season/playoffs, Marner played on a 94 point pace.
Well congrats of being one of the few to get the correct date. I notice you had nothing to say about Marner having essentially the same P/60 with Kadri as without. The reality is that for the last 33 games of the season Marner's P/60 on the PP were double what it was for the first 49. The PP P/60 of the first 49 games, which is very good, is what he is likely to average going forward.
In the playoffs Marner was lucky. It happens. Matthews scored 4 goals during his first game. Great player. Great goal scorer, but hasn't come close to 4 in a game since, despite having games with far more high danger shots. Sometimes you get lucky. Most really good shots don't go in. Some really poor shots do go in.
In the playoffs most of the goals Marner was on the ice for were from medium danger chances, and most of the goals against that Matthews was on the ice were from medium danger chances. Normally those chances will find the back of net about 3.5% of the time when someone like Marner or Matthews is on the ice. For Marner they were going in 5 times that often in the playoffs. At 5v5 he was on the ice for 6 goals for and 3 against. If he maintained his on-ice shooting percentage from the regular season for high, medium and low dangerous attempts (both for and against) the expected result would have been 3 goals for and 4 against. For the PP it would have been only 1 goal for (again - lucky). For Matthews, it was the opposite. He was on the ice at 5v5 for 4 goals for and 8 goals against. If he maintained his on-ice shooting percentage for high, medium and low danger attempts (both for and against) the expected result would have been 8 goals for 3 against, and another goal for on the PP.
Players generally regress to the mean. For next year's regular season Marner will likely see his PP numbers regress to the mean, and his 5v5 numbers stay about the same. Maybe he will get a small bump at 5v5 from playing with Tavares (although his history with Hyman is very bad). Matthews will likely see his 5v5 numbers regress and that will likely be balanced by an improvement from unsustainably bad PP numbers.