I was the one (btanchors) who did the long chain of UOAs with the particle counts (the link was posted by steveh)
Particle counts are very tricky. In the long series of UOAs in the link posted by steveh, I was attempting to use particle counts as a way of identifying differences in filtering abilities between different oil filters.
In my case, the data was so widely scattered, I was unable to reach any real conclusion on the differences in the filters.
The only part that was repeatable in the tests was when I tested two Amsoil EaO filters. For some reason, I obtained worse results using those filters. But even then, I don't think it was enough to reach a firm conclusion (in fact I use Amsoil filters myself).
I'm not terribly surprised you obtained a different result by having the particle counts run by another lab. I suspect you would get different results if a sample was re-run by Blackstone - although it would not be wildly different.
In the end, I gave up using particle counts as I just couldn't seem to "master" them well enough to make use of the data, and it was getting expensive to have them done all the time. I was very careful in how I drew each sample, and tried to maintain as many variables as possible constant during these tests, given we're dealing with a passenger vehicle that is in normal use (in other words, not in a lab environment).
If I could come up with a new "angle" on how to use Particle Counts more effectively, I would do it - but for me, it just wasn't worth it to continue to use them.