cwazy
cwazy
  • Threads: 14
  • Posts: 178
Joined: Mar 18, 2016
July 16th, 2016 at 5:17:40 PM permalink
I spend a fair amount of time counting single deck games in Northern Nevada. In these casinos, they don't wash the cards when they change them, and shuffle machines are the the exception, not the norm. The simply shuffle once or twice extra with a new deck. This invariably leads to card clumping. From a counting perspective, the counts tend to get very extreme - either very negative or very positive, for the first 2-3 shuffles of new, unwashed, manually shuffled cards.

Since the normal shuffle procedure (at least in the games I play) clearly does not come close to fully randomizing the cards from their starting point through the shuffle, is there a way to classify the distribution in a given deck and then make a judgment as to whether or not it would be smart to play after the next shuffle? For example, if one could estimate that after the shuffle the current deck will have larger than normal clumps of small cards, that may not be the best table to play at for the next shuffle (or few shuffles) because the dealer will tend to bust less often.

I'm aware of shuffle tracking, but that's not really what I'm talking about. More like....if someone has run the numbers to positively say "if the count on the previous deck ever got to plus or minus 5 on a hand-shuffled single deck game, you shouldn't play the next deck because the clumps will tend to stay together at least through the next shuffle". Or perhaps the opposite might be true. I know there is some correlation from one deck to the next, because they simply don't shuffle enough to randomize the cards - and there has to be a way to take advantage of that.
DRich
DRich
  • Threads: 86
  • Posts: 11756
Joined: Jul 6, 2012
July 16th, 2016 at 7:58:29 PM permalink
I can't answer your question but I will agree it does have an effect. The real question is does it have enough effect to overcome the house edge.
At my age, a "Life In Prison" sentence is not much of a deterrent.
  • Jump to: