This is a fair take

Dismantling other bench units to the degree Quickley has done is noteworthy, and usually only done by elite players. It should be noted that when Quickley played WITH Brunson, the Knicks were even more dominant (+13 net rating in the sample this year. of course this is a small sample but we are speaking the history of what happened). You could filter the data however you want for the number of starters on the floor. The on/off signal is so strong it does not matter. Also, rotations across the league are more staggered than ever, so I don’t think the quality of his opposing lineups really differ that much from a starter. You’re pretty much always playing against a few starters at meaningful points of the game. He also played extremely well in his 21 games as a starter himself. He played well as the on-ball PG, the off-ball SG, against the bench, against starters. He performed poorly on offense in the playoffs vs the Cavs, but it should be noted Cavs historically poor offensive series mostly occurred when Quickley was on the court.

You can still have good advanced stats if you are on a really bad team. You know sometimes I actually hear the opposite argument – he only looks good in on/off or RAPM data because the rest of his team sucks so he sticks out. If the team is improving on offense and/or defense when you are in the game, with a big enough sample it will pick it up. For instance, Robert Covington was on the process Sixers. Surrounded by godawful talent. RAPM recognized him as a top 50 player when he was surrounded by G leaguers. Why? Because he’s super impactful defensively. Forced a bunch of steals and deflections. Opponents just scored less when he was out there. When he left, the Sixers went back to getting destroyed.

Now how the coach does rotations can definitely impact how well a player’s signal gets picked up in on/off data. If you’re doing full 5 man unit subs at a time consistently for years, i think on/off data starts to lose a bit of value. The more the lineups mix the more reliable on/off data is.

You may think they are better players, but do know every single advanced stat model disagrees with you. Should be noted there are many differences in methodologies across those models and they all have the same conclusion.

You may think those players are more likely to improve from here, but Quickley has been better than either of those guys for the last 3 years. Almost all of the Knicks success on the court has occurred with Quickley on the court. They’ve been mediocre without him.

Also, his position is extremely important. He has the ball in his hands a much higher percentage of the time than either of those guys, and while he isn’t the POA defender those guys are, he is an extremely good team defender and communicator. Think Quickley has far more offensive upside than either of them. Not buying Jaden’s O.