It may be confirmation bias on my part, but i tend to agree with you hunch about shoot-versus-pass innate tendencies.
Alternatively, my hypothesis is that passing is conditionally modeled for big men - i.e. basket success is first evaluated on relative skills. if likely a missed shot, then with some randomness, a pass is executed instead based on passing ability. In this model, passing would allow a second attempt in the same sequence, hence adding value. However, if your player has tremendous scoring advantage over the opponent's defense, then passing is moot. This particular model may also be applicable only for certain offensive strategies, such as look-inside.
Of course, this is just a random guess, so definitely don't quote me and feel free to disregard a totally insane idea. I have absolutely no proof - it is just how I would like to model it if I were to design a game engine myself.
Last edited by rocket.science at 10/22/2012 1:49:09 PM