Not what I "said", anywhere, at all. And, irrelevant...
Simply stated, without isolating primer weight as the ONLY variable of the test, a solid conclusion can't be achieved that states "it was the primers"...
If correlation is good enough, then I'm not gonna belabor that point. Just sayin', without isolating ONE variable & establishing ALL others as 'constants', one cannot rule them out as factors that skew your data...
And that's not to discount your data, altogether. Surely, you guys are onto something. So, why not take it a step further and see about 'cleaning up' a test to further substantiate what you've already got to ponder?
What's the harm in having more, relevant data, than you already have?