Understanding discrepancies in Dragon (7) tables.

Maybe the answer to this is already somewhere; if it is, then I cannot find it.

When evaluating expectations from counting the Dragon (7) side bet in EZ Baccarat, I've noticed minor discrepancies in the table offered by Mr. Jacobson on this page and the table offered by Mr. Shackleford when comparing the Panda 8 expectations to the Dragon (7) here.

Specifically, I'm attempting to find the expectations resulting from a cut card depth (penetration) of one-and-a-half decks (78 cards).
One shows a profit expectation of .176 units per shoe, while the other table shows a profit expectation of .21 units per shoe.
Assuming the units are $100, the difference in expected profit per shoe between the two tables is $3.40.

With a penetration of 78 cards, the two tables mentioned above have the same bet frequency (5.37%), and very similar advantage per bet made (4.79% versus 4.82%).
The very minor difference in advantage per bet made is not enough to explain the difference in expected profit per shoe. The only thing I can think of is that the two tables are using a different amount of average hands per shoe for the calculations.

Anyone have any input on this?

Thanks.

Post edited 12-24-2013 so as to clarify and eliminate confusion.

Quote: rob45

When evaluating expectations from counting the Dragon (7) side bet in EZ Baccarat, I've noticed minor discrepancies in the table offered by Dr. Jacobson on this page and the table offered by Mr. Shackleford when comparing the Panda 8 expectations to the Dragon (7) here ... Specifically, I'm attempting to find the expectations resulting from a cut card depth (penetration) of one-and-a-half decks (78 cards). One shows a profit expectation of .176 units per shoe, while the other table shows a profit expectation of .21 units per shoe.

Anyone have any input on this?

Note that at 78 cards, we have exactly the same bet frequency (5.37%) and our average edges are within epsilon (4.79% vs. 4.82%). Mike gets a win-rate of 0.259 units per 100 hands, I get a win-rate of 0.257 units per 100 hands at 78 cards. The win-per-hour (60 hands) results are 0.16 for Mike and 0.154 for me. I don't see a meaningful difference in those results.

It seems like the answer has to be in the # of hands per shoe. I get 68.32 hands per shoe. I am not sure what Mike's value is.