So, the terms “U.S. dollar”, “fixed rate mortgage”, “401k”, “capital gains”, etc. don’t really have any inherent meaning beyond the set of rules that govern their interactions. This rule system itself can change over time depending on politics and human emotions, so you have to try to model the dynamics of this meta-system as well, even though that’s pretty difficult.
Once you’ve gathered all the different rules into some programmatic representation, then you decide what it is you want out of life and at what points in your life, then you start optimizing. For example, would you rather have a 10% chance of having $1,000 on some future date, or a 1% chance of having $10,000? The expected return is the same in both cases, so your optimization should really consist of targeting a specific shape for the entire distribution of outcomes rather than a single number (or specific assets).
Not surprisingly, the average person (including me) does not have the right combination of 1) detailed knowledge of the global financial system and its rules, 2) how all these rules are likely to change, or 3) the technical background to implement the advanced mathematics required to optimize all of this in an accurate way that actually makes valid predictions.
So most people instead just guess and base their financial plans on their hunches about how “money” works, and large organizations that have the time to do all of this modeling in a more rigorous way come out much further ahead.
I still think simplifying the “rules” of finance would be one of the best things we could do to reduce wealth inequality, as inequality seems to largely be hidden in the highly volatile landscape of the cost function for the optimization problem I described.