Understood, and apologies for passing on advice without first-hand experience. In my defence, I did actually read through what I believed to be the definitive thread on the topic (along with several others) before posting...
http://www.troublefreepool.com/threads/8265-A-discussion-of-CYA-turbidity
... although in fairness on the first read-through I missed the point that in your second simulation (post #17 test two) there was no difference noted between diluting with tap water and diluting with murky pond water. Excluding that missed data point, my take-away from the thread was that :
(a) filtering *did* make a difference for most people, albeit a small difference and from a very small number of people testing
(b) that said, if there was a difference it was sufficiently small as to not really affect the course of action
EDIT -- another point you made (which I didn't consider sufficiently) is that the difference observed from cloudy water is within the normal accuracy limits of the test, and so is even less relevant
(c) there was an open question from chem geek about whether algae and dirt would have the same effect on the test, and general conclusion that testing with actual algae would be a good idea (but since the people most likely to test are also least likely to have algae in their pools I guess this may never happen...)
In this case filtering the pool water seemed like a good way to address the OP's concerns about impact on the test, but in hindsight I agree that I should have at least researched this a bit more carefully before saying anything. I thought I had done so, but your second test in post #17 above definitely argues against the validity of filtering...
So... apologies, and will be more careful next time.
EDIT -- I'm trying to resist the urge to drop my FC levels, let algae start growing, and redo your tests using real algae and real coffee filters