sjm34 wrote:So you would prefer to use the NBA defensive rtg which only shows how the team performed with Young on the court, and does nothing to show how Young himself actually performed. Sounds like a really accurate portrayal. We certainly wouldn't want to look at defensive rebounds, considering they are only the deciding factor of getting an extra offensive possession for the opponent. Of course that extra possession doesn't show up in the NBA rtg because that is strictly based on 100 possessions.
Defensive rebounding actually affects NBA team defensive rating. Possession ends always for basket, turnover or defensive rebound. So if team has worse defensive rebounding rate with certain player on the court, it increases the opponent offensive efficiency.
If you check on/off stats from basketball reference, you see that Minnesota defensive rebounding rate is little lower with Young on the court than with Bennett but still overall defensive efficiency is better with Young because opponent eFG% is lower and turnover rate is higher with Young on the court.
I agree that it makes sense to try to neutralize the effect that other players have for team defense. One possibility is to calculate for example what is team defensive rating with certain player playing with Bennett versus him playing with Young.
I did this for Dieng and Williams (since both Young and Bennett had played decent amount of minutes with both players) based on stats in 82games.com. It seems that Dieng has defensive rating 112.9 when he plays with Young and 122.3 when he plays with Bennett. Similar analysis about Mo Williams show that he has defensive rating 121.6 while playing with Bennett and 111.1 while playing with Young. Ultimately this should be done for all players that have played with both Bennett and Young but already this brief analysis is telling similar story than what team defensive rating with Young and Bennett tells, Minnesota just seems to play better defense with Young than with Bennett.