1. I still don't completely understand using only weighted event counts. E.g., with a SAP count of 6 10 20 8 the conventional wisdom is to bet 2s will go to 3. But there have been an equal number of 2s and 3s, 5 each. So why is it more likely 2s will go to 3 in THIS PARTICULAR SHOE, in which they're running equal, and which shoe doesn't know or care what the normal occurrences are in the long run? 2. When counting 1s, 2s 3s and 4s does anybody have an opinion as to whether it makes a difference whether they are Ps or Bs. E.g., if only, or mostly, Ps have been 3s, would you be less likely to bet on B going to 3?