First of all THANK YOU for the a truly amazing, well-done test, and for taking the time to document it and post the result. This is really useful.
I'm still not entirely satisfied with my MES 30 Gen 1, specifically because of the smoke quality. I'm pretty sure that many things I did when I was a newbie fifteen months ago ended up creating creosote. I sure did get some tongue-numbing nuts, and believe me, you don't want your nuts to numb your tongue.
I switched to using the
AMNPS, and that helped some, but I found that I still got bad smoke at low temperatures. Your test reveals some of what may be happening, although I always keep the vent wide open, and also added a mod in place of the chip loader which significantly increases the airflow through the smoker
I know there are threads about the importance of TBS, and other posts about improving the quality of the smoke. Some of these talk about the water pan (pretty clearly a bad idea to use in an MES); some talk about temperatures (the colder the smoking temperature, the more likely to have creosote problems); the venting (the subject of this thread); the source of the smoke; and more.
Of all of these, the mailbox mode seems like the most likely thing that would further reduce the bitterness. Do you have a mailbox mod and, if so, would you be willing to do one more test using that? I have been meaning to build some variant of the mailbox mod, in hopes that creosote would condense in the mailbox and/or the duct connecting the mailbox to the smoker, but since it requires an hour or two of work, I was looking for some additional, concrete reasons why it would be a good thing.
If you do the test, I'll pay for the whiskey.
[edit]P.S. I'm simultaneously following this thread:
Clean, Cool Smoke via Thermodynamics
That thread leads to this blog post:
Smoke Direct vs. Indirect
which begins to answer some of my questions about using an external smoke box.
These two threads together are providing all sorts of great ideas of how to get a better result.