[qwerty49]

+/- per 40 minutes for conference games (missing Washington game, fromStatbroadcast)
...
This doesn't adequately take into account the quality of the opponents.
...

With respect, I think this stat is interesting but somewhat misleading because we are so deep that our reserves are generally way better than our opponents' reserves.  I think this stat tends to reflect who is doing well during garbage time, when the opponents are not too motivated.  Just my opinion but I don't think +/- per 40 min holds much water.

What would hold water?
Not shooting percentage as that is offense only.
Not blocks as that is defense only and depends on playing time (and quality of opponent).
Not points/game as that depends directly upon playing time which varies wildly from player to player and neglects defense or role on the court (play maker).
Not points/40min as that neglects defense.
Not +/- because that again depends directly upon playing time.

Please suggest what would hold water for you. 

I'd love to find a way to compare the relative value of a post, a defender, a sharpshooter, and a play maker from the same team.
(This does NOT work to compare players on different teams.)
+/- per 40 minutes is the best way I know of but maybe you've got something better.

I don't have an answer, largely for the reasons BC cited above.  This is like pushing a piece of string.  I don't believe the "magic stat" you describe exists.  I actually think +/- is the best of all evils here.  Yes, it reflects playing time, but playing time is also in itself a reflection of value, and of matchups, at least in the coaching staff's eyes.

[qwerty49]

+/- per 40 minutes for conference games (missing Washington game, fromStatbroadcast)
...
This doesn't adequately take into account the quality of the opponents.
...

With respect, I think this stat is interesting but somewhat misleading because we are so deep that our reserves are generally way better than our opponents' reserves.  I think this stat tends to reflect who is doing well during garbage time, when the opponents are not too motivated.  Just my opinion but I don't think +/- per 40 min holds much water.

What would hold water?
Not shooting percentage as that is offense only.
Not blocks as that is defense only and depends on playing time (and quality of opponent).
Not points/game as that depends directly upon playing time which varies wildly from player to player and neglects defense or role on the court (play maker).
Not points/40min as that neglects defense.
Not +/- because that again depends directly upon playing time.

Please suggest what would hold water for you. 

I'd love to find a way to compare the relative value of a post, a defender, a sharpshooter, and a play maker from the same team.
(This does NOT work to compare players on different teams.)
+/- per 40 minutes is the best way I know of but maybe you've got something better.

I don't have an answer, largely for the reasons BC cited above.  This is like pushing a piece of string.  I don't believe the "magic stat" you describe exists.  I actually think +/- is the best of all evils here.  Yes, it reflects playing time, but playing time is also in itself a reflection of value, and of matchups, at least in the coaching staff's eyes.