It's interesting to compare these two games using the NO table. DDB
has a variance more than twice as large (42.2 versus 18.8), but the
higher EV of 10/6DDB swamps the variance difference. The NO table
shows it takes 1,312,000 games for 10/6DDB (versus 101,253,000 million
games for 9/6JB) to have an 84% chance of breaking even, at a CB
of .5% ... a fairly high CB. Below .5% 9/6JB is a losing proposition.
At .5% 10/6DDB is 100 times better than 9/6JB! It takes a CB of 1.5%
before the two games equalize, in terms of NO (that is, 172,000 games
for 10/6DDB versus 180,000 games for 9/6JB). How often do we find a CB
of 1.5%?
Of course 10/6DDB is not as widely available as 9/6JB, and 10/6DDB may
be unavailable at denominations above $1. But for $1 or lower
players, it sure seems to be the superior choice in LV. I was
surprised at how big a difference there is.