Just a quick error correction - around 26 mins in I mistake the amount villain needs to risk on a shove - he risks his whole stack so he would need us to fold more than what I said in the video.
You base a ton of the analysis on the assumption that his 3bet stat is accurate... and you have a 100 hand sample on what seems to be a very tight player. I would assume his 3bet stat hasn't converged and is probably above the true %.
All we can go on is the info we have - it is possible that he doesn't 3bet that much, but based on statistics I think the observed frequency is the most likely frequency. I also don't think I went overboard on the assumption - I was still fairly conservative in my defending range, so even if he is a bit tighter in his 3bet range we still should be doing ok vs it.
Fair enough, by the way I do enjoy this format a lot.. maybe try and spend less time zoning in on every possible detail in order to have more time to examine other hands though.