I wonder how much the multiple-runs eliciting more yield has to do with more agitation, and how much of it has to do with the fact that there are more trichomes being allowed to free-fall to the bottom of the bags on each run. Basically what if the first run of agitation breaks all the trichomes off, but a high percentage of them never actually reach the screens because they don't free-fall through the ice/water/plant mix. Then ever time it's remixed, it's basically just churning the trichomes that were stuck in that ice/water/plant mixture allowing them to free-fall.
OK this is, possibly, the most important post I've ever seen on this subject. I've been deep-diving the methods and techniques of hash making for months, and will soon start growing and making my own.
In support of your theory, if the heads indeed, settle to the bottom, you would only need to drain a few seconds of water from the bottom, then wash again, right? But that's NEVER the case... everyone drains the entire container!
So far the Frenchy Cannoli method is at the top of my list, and the main takeaway I get is that he does a gentle and short first run, so the "ripest" heads fall first, then he increases subsequent runs, so 1 minute, then 5, then 10, 15, etc until a certain level of green contaminants show up in the bags. He uses 2 sieve bags, a 190 and 45 and keeps only the 45. So the separation in quality is the length of wash, NOT the micron size!
So if I take YOUR theory into consideration, there can either be a continuous drain, or, a waiting period between washes, and draining only part of the water.
Have you experimented with this since you made your post in Feb 2019?