Problem is, most people who want to "start a startup" start from a baseline scenario that is already very risky. Especially the standard "let's start a B2C comp

Problem is, most people who want to "start a startup" start from a baseline scenario that is already very risky. Especially the standard "let's start a B2C company in a field I know nothing about, because I can write code". You're completely right that most first-time entrepreneurs should be mitigating risk [1], but the easiest ways are usually not part of their thought process. Things like "let's start a consultancy", or "let's build a B2B company in a business that I validate in advance by getting people to actually pay for my product's development", or even "let's take a working business model and import it to a new geographic location". [1]: I assume here that most first-time entrepreneurs care more about the expected value of their business, rather than other things like "changing the world" or "running a 100-person business". This isn't entirely accurate, but is very close for most non-rich people, and actually a lot of the ways to mitigate risk can also increase the chance of these things as well.