That video is ridiculous, as far as destructive testing and tensile testing is concerned, but the end results pretty much show what epoxy is capable of. That the wood itself failed, not the epoxy.
You got it backwards. Epoxy has a tensile strength of about 8500 psi, while liquid rubber's only about 750 psi. So epoxy is at least 10x stronger than liquid rubber. If epoxy is used as a bonding agent, it will make the plywood's joints stronger than the wood itself. With epoxy as a bonding agent, you could build a plywood tank without a single screw. With epoxy as a structural element, you could build a tank without using any wood. If you use epoxy just as waterproofing agent, then just be sure that you use enough of it.
Probably the biggest mistake is using the minimum recommended thickness or minimum number of recommended coats. For some reason, some people think the minimum should be the goal. The minimum has very little margin for error or safety build in. I've seen the same % of failures with liquid rubber for that reason. Many people don't know what thickness the product should be or how many coats they need for waterproofing an aquarium. Many people don't know how to calculate how thick each coats is. Many people don't know how to calculate/measure/weight how much product they'll need for each coat. They usually guess and go at it blind. It's really an engineering question that can be fixed with some mathematical skills, a scale, and an excel spreadsheet. With my spreadsheet, I've got my mixing calculations down to 0.1 grams in accuracy.
