With all due respect, I think your analysis is wrong. He points out how much luck plays into everything and so even if someone younger with 95 year experience came out I think he'd point out that in the end there's an element of luck to everything. The world isn't deterministic.
This is why I've been so curious as to exactly what Reddit is really trying to do with this api change. If the point of Reddit is to consume content then the content itself (and the amount of it consumed) is really their business model so all actions should support that vs. getting people to arbitrarily be on their site. If your business is content, it shouldn't matter if it's a third-party app or if it's on your main site. I just cannot figure out the long-term logic behind this move (ofc short-term it's about $$ and their IPO but this'll hurt their model in the long-run)
I wrote a Go library for postgres & redis that handles all caching automatically.
where all you do is input the queries your application needs and then it will automatically determine which columns are affected & do all caching / cache invalidation for you.
I have it to the point where due to Redis being able to push onto a list that only writes hit the database & it even will update the cache keys automatically so that after an insert or update the cache is still hot.