In a uniform random distribution, every outcome is equally likely. For example, each card in a 52 card deck has the same chance of being selected (assumes the deck is thoroughly shuffled). Another random distribution is the bell curve. Values at the tails of the distribution are less likely to occur and those at the center are more likely. Both are random distributions. And so it is with Cashman - in a grid at reset, the distribution used more closely resembles a uniform distribution. However as the grid becomes more populated other random distributions are used where it is more likely to select a populated square than an empty square. Without the par sheet, the slot AP must guestimate these distributions. But random they are - just not uniform.