IT's usage rate just isn't compatible with Steven's offensive system as a starter.
On what do you base that assertion? How does his offensive system differ for starters?
The ball has to move and everyone needs to be shooting. One player shouldn't be dominating the ball and the shooting.
What if that one player is by far the most efficient scoring option?
If you have 4 players who score at a TS% of .450 and one player who scores with an efficiency of .570, then I don't care what system you are running. If you don't run a system that has makes far, far heavier USG of that guy than the others, then you are a terrible coach and should be fired.
I totally disagree with the idea that 'everyone needs to be shooting'. That's a recipe for a mediocre-to-awful offense. Oh, wait -- that's pretty much what we have had the last couple of seasons with the _exception_ of when Isaiah arrived and was on the floor.
Should Kendrick Perkins have been shooting equally as often as Pierce, Ray and KG?
I'm not saying that Isaiah should be taking every shot. No offense is going to succeed in the long run without more than one scoring option. Defenses will adjust and take that away, especially in the playoffs.
So whether IT starts or not, we have to have at least a couple of others emerge who are legit scoring threats. But that isn't likely to be "everyone" and it doesn't need to be. Ideally, the team needs to resolve who the 2-4 most efficient scoring weapons on the team are and make sure that those guys are getting the lions share of USG.
First off, Isaiah's usage rate doesn't fit in with Stevens system because Stevens still has a system that is very college oriented and that stresses perimeter ball movement until an inside pass for a quality shot can be found, a lane can be found for a driver or an open shot shows for whoever might find themselves open. Thomas is very much a player who offensively has to have the ball stick in his hands to make him and his scoring most effective. That is most probably better for the team as a whole if he is playing with players, that as you said, have poor TS%s and most often those players are on the second team.
This may be what you imagine Brad's system to be, but I don't know of anything in the Read & React / Motion offensive systems that says that shots should be distributed equally among all players and not weighted heavily towards your more efficient players.
What about his system makes it bad for Isaiah to be a starter but okay as a reserve?
Second, Thomas might have a TS% of 57% but he isn't playing with players with TS%s in the 45% range. Last year, AB had a TS% of 51% and would have had a higher TS% if he just stepped out and shot more three pointers instead of long 2 pointers. His FG%, 2PT%, 3PT%, and FT% are all in the same exact range as IT. Olynyk and Zeller both had +55% TS%. Turner and Sully both has TS% around 50%. Amir and Lee are also high TS% options. So IT wouldn't be starting with four guys that have TS% that were at 45%.
Third, I agree that your best scorers have to have your highest usage rates. Your argument over that is with Brad Stevens and his system and not me.
The .450 vs .570 comparison was an illustration of a point (why an equal distribution of USG makes no sense), not an assertion about what the exact TS numbers of the Celtics were.
However, the point it illustrates is still very much valid. Whenever you have a wide variance in efficiency, you should strive to weight your usage towards your most efficient players. That's not always possible (the other skills of the various players may not make it easy to set them up to make scoring attempts) and certainly the efficiency of some players may not sustain with increased USG. But nevertheless, it is a direction you need to move in or else, as I said, you will have a mediocre offense.
The TS% of the Celtics as a whole was .523, which was ranked #23 in the NBA last year. League average was .534.
The TS% and USG% of the main rotation players on the Celtics:
Player TS USG%
Zeller: .594 19.6
Thomas: .579 32.1
Olynyk: .558 21.1
Bass: .557 19.5
Jerebko: .533 17.6
Crowder: .512 17.9
Bradley: .507 21.0
Sullinger: .503 23.6
Smart: .491 15.1
Turner: .482 19.5
Looking at this, the problem wasn't that Thomas' USG was so high. The problem I see here is that guys with low efficiency numbers at the bottom of that list have similar USG numbers to guys at the top of that list.
Saying Avery's efficiency
could have been higher if only he had shot more of his shots from a stride back isn't very useful other than to hope that he will do so in the future.
It's certainly true that the Celtics did have some players who posted high TS numbers, and it is great that we are adding Lee and Amir -- they should help a LOT. There is definitely good reason to hope that this coming season the offense will function much better.