What we see is that in situations where there is not a bust, most of the time players will either stand pat or draw one card. It turns out that we can treat draw two or more cards the same and the gain is barely affected. The difference is so slight that I did not bother to include a table.

Further complication

In the situation where the player draws exactly one card and does not bust, we could further break down the data by the card drawn. In the following table, we see the dealer upcard in rows and the drawn card in columns. Cells indicate the average count of the player hidden cards. It turns out that the extra gain from this information is somewhat under 5%. Not worth it in my opinion, unless you have a computer making all these calculations.

Conclusion

To gain these results, you must be using a very large number of indexes. Also, the gain will not be as high in other combinations of players and rounds. I would not use this technique. But as the rules become more difficult, who knows what we may be forced to do.

 © 2009 Norman Wattenberger