And a high sublevel of IS plus RB could push that 27.26 above 28. I don't know if that is the case here, but it's especially unfortunate if he's entirely capped since he'll never be able to train enough passing to have his first career assist.
Our DB system is correct on calculating the CAP, but if it's like you say is really incredible...
I am sure it's as accurate as possible, but with cross training involved it's no longer possible to be highly certain of sublevels of skills other than perhaps immediately after they pop. If the tool is depending on estimates of how much shot blocking has changed in the current formula, that data may not be as accurate as the coefficients originally relayed by (I think) JoeyKa, though I admit I am not sure who to properly credit with that.