just a thought,
I dont think effects like death blow can go over 100%
hence its percentage would work like probability, though there is still a dimishing return rate that I have to factor in somewhere.
I think the formula would be closer to :
50% chance 1 item
2nd item of 50% chance....
50% chance of 50% that 1st occurance did not happen hence 25%
3rd item of 50% chance..
50% chance of 25% chance that first 2 occurances did not happen..hence 12.75%
though there does seem to be an added % diminishing as the data ihad for the 3 items added together ended up with a different x factor, hence there is something else affecting deathblow perhaps an attribute?
either way my method gives a much closer value as the factor hovers around 1.1-1.6
may conduct own experiment later when I have death blow items .