"the Thunder chose depth and flexibility over a core of four stars that might have left them capped out in perpetuity. Harden needed the ball, but so did Durant and Russell Westbrook; there are real diminishing returns in pairing three ball-dominant players, and it was unclear if any of those home-grown stars had the shooting touch and deferential nature to evolve into Oklahoma City's Chris Bosh. The Warriors soared toward history the moment they gave Stephen Curry the ball on every possession. They wouldn't be able to do that with Durant on board. Paying four players close to the maximum salary leaves almost no money for nine or 10 leftover roster spots. Top-heavy teams scrounge every summer for one-year minimum bargains, undrafted rookies and aging ring-chasers. [..] filling eight roster spots on the cheap is an uphill battle. [..] Maintaining wiggle room below tax hell also allowed the Thunder to chase third-tier vets like Anthony Morrow and Kyle Singler with deals in the $3 million to $4 million range. That was the idea: The extra breathing room would expand Oklahoma City's long-term core from four guys to six or seven, and leave more money for everyone else. It also left only three stars, and when any of them got hurt at the wrong time, the Thunder were done. And that, in the end, may be the most persuasive argument for Golden State going after Durant: Adding a fourth star is a hedge against an injury to one of the other three. If any of Curry, Klay Thompson, Draymond Green and Durant get hurt in May or June, the Warriors could still have a fighting chance. In the postseason hothouse of shorter rotations, that extra dose of star power might be a better cushion against bad luck than a deep bench. [..] Another concern: The second key building block for teams over the cap -- the midlevel exception -- has been borderline useless amid the cap boom. Overflowing cap space soaks up almost every viable rotation player now."
