The bench is well below average, horrible actually. They need a front-line center with decent size that can anchor the defense and be of some value on offense, allowing Theis to play on a hopefully much improved second unit.
The bench was bellow average in his production this year, but horrible is maybe a little too much ...
https://stats.nba.com/teams/advanced/?SeasonType=Regular%20Season&StarterBench=Bench&dir=-1&sort=NET_RATING
Boston’s bench has been substantially above average, at seventh in the league.
I don't think anyone can look at our bench and say it's top 7 in the league
Well, Danny Ainge did. I don’t have a link.
I’m surprised that you are so definite about this. I think that it’s a danmed difficult question, and it seems to me that it’s on you to say what’s wrong with the official stat. I’m not attached to it being ‘right’, either; it’s a snapshot, like any stat, with a confidence interval.
I myself am not very interested in ranking teams in this way - but very interested in what lineups give you a winning formula.
You can accumulate the raw +/- numbers for each individual in every game who doesn’t start, and in some sense you’ll get a number for the bench’s net. That strips away important contexts, though - like who’s playing together, whom they’re playing against, etc.
Context matters. Take the starters and replace Hayward with Ojeleye, for example. Now Semi is a poor rebounder and super-low with turnovers, and yet a Walker/Brown/Tatum/Ojeleye/Theis lineup rebounds at 52.5% compared to the starters’ 47.5%, while turnovers get
worse, at +1% net compared to +4%.
None of the top-20 most used lineups had all subs. One of them had four, though: Wanamaker/Smart/Tatum/Ojeleye/Kanter. That lineup performed better than the starters:
Starters Ortg 1.23/Drtg 1.13
W/S/T/O/K Ortg 1.27/Drtg 1.08
- i.e., nine points better in a 100-possession game.
So should that group start? Probably not - context matters.