@Kathele et al
I appreciate the effort you have put in trying to elucidate this. However testing against a target dummy is not a viable method of testing. If you search these forums you will understand why.
But in short, you need to model this with theorycrafting because of the way stats and talents scale and interact with each other, these are not linear relationships, hence target dummies can only be used for analysing specific mechanics ie proc. rates etc.
not damage output.
In general when performing analysis from combat logs, or indeed analysis on any measured data you have to understand how to do statistical analysis. Go and read up on
Statistics - Wikipedia, the free encyclopedia.
In this case you would have to construct a null hypothesis of there not beeing a difference between the two poisons and then have a sufficiently large sample to calculate to what degree this is statistically significant. Sufficiently large is a matter of some debate but most people have sample sets obtained from several
hours of testing.
However in this case you would still have only answered the question if it was better with WP attacking a target dummy in the nude which I think holds scarce interest to most players.
I am sure I will get an infraction under the "do not reply to stupid posts" rule but maybe some would-be-posters will read this and reconsider posting their target-dummy-dummy-theories.