As a coach, you need to know in advance of the game what percentage you need to convert to make it worth it. Then decide if you think your chances of converting against that team are greater. It would be more of an educated guess, but still better than even a season average, which has nothing to do with what your average would be vs that particular team.
Either way, it's a very interesting concept and most teams averages would fall above the threshold for making it worth trying. Therefore, since no coaches do it, I think it's reasonable to assume most coaches make bad, uneducated decisions at times. So we should quit saying if all coaches do it, it must be right.
I agree, you need to have an understanding beforehand so that you are prepared to make a decision should that scenario present itself. However, the game is played in a serial manner, you don't know what will happen on your CURRENT possession, let alone your NEXT possession or whether you will even receive a next possession, you only know the situation now. So the discussion involving the situation where you are 14 points behind is purely hypothetical and statistically nothing more than an exercise because you have no ability to constrain the game to achieve those necessary conditions. It's playing the coulda', woulda', shoulda', 20/20 hindsight game, which is a fool's game. Therefore, the proper play is to make the best decision based on known parameters, not to speculate on future conditions for which you have virtually no control.
The situation of being down 14 with a good likelihood that 2 possessions can be expected in the time remaining is significantly different than what KF faced against Purdue. Obviously, if you make your decision based on a dissimilar scenario, you more than likely will make the wrong decision. Of course, you might luck into it working out for you, but strategically, to depend on luck, is effectively failing to coach.
Finally, whatever your decision was, you need to be accountable for it, and if it was based on poor information or logic, you need to admit the mistake. The idea you should not be held accountable for your decision-making is a ridiculous notion, especially when you are paid top dollar to produce the best outcome possible. No other CEO of an organization could expect to be treated otherwise, why should KF be treated any differently?