It's tough to make good estimates without some assumptions.
Here's a Bayesian method I just used to calculate the probability of getting an FBSS.
We start with a prior probability of getting 1 FBSS every 10 hours (this is arbitrary). Since you had 109 attempts in 18 hours (or 6.1 attempts per hour), that means our expectation rate of FBSS is 1/61.
If you use Laplace's rule of succession and fit a beta distribution to that prior, we get Beta(2,61). This distribution looks like this:
[You must be logged in to view images. Log in or Register.]
Now, when we include the information that you provided (101 tries, 0 successes) we end up with a Bayesian that looks like this
[You must be logged in to view images. Log in or Register.]
In this one, the blue is the prior probability, the red is your findings, and the gold is the posterior probability. So, this tells you what the FBSS drop rate is most likely to be based on the data available.
Some statistics on that post distribution:
Mode: p -> 0.00606061
Expected value: 0.011976
From this data, I calculate the FBSS rate per bloodthirsty ghoul killed to be 1.20%.
EDIT: Here is a cumulative probability function based on a geometric distribution. The x axis is the number of frenzy/bloodthirsty you kill and the Y axis is the probability of having looted at least 1 FBSS.
[You must be logged in to view images. Log in or Register.]
EDIT: If anyone has more data, I would be happy to throw it in here.