The lead into graphic in The Athletic article says that those metrics are over the course of the last 3 seasons where and only include defensemen with greater than 1500 total minutes. Although I think your point still stands, I was surprised to see this.
This is a good point. It makes me wonder how Merrill was used the two season prior to this last one. Was he getting a good amount of minutes as a #5-6, or did he have one really big season in there somewhere? 1500 minutes over 3 seasons basically just means "full time NHL defenseman". For reference, Pateryn picked up 1370 of those 1500 minutes in 18-19, when he was getting 17 minutes per game on our third pairing.
Ok, I decided to just look rather than wonder:
Looking at Merrill, it appears he did have a really nice season for Vegas that first year (of the last 3), but it apparently also saw his least, and easiest, usage per game. Also worth noting, he saw the largest majority of his playtime in that one good season, as it was his last "full" season. I put quotes around that, because he only played 57 games, but it wasn't COVID shortened. As his play-time per game has gone up, and his usage has gotten more difficult, he appears to have played significantly worse by both traditional stats and advanced analytics. I think both of these guys, Pateryn and Merrill, are being buoyed by a single good season.
I also came to the conclusion that a half-time 3rd pairing defender could make it to 1500 minutes over three healthy seasons, but it'd be close.