An interesting concept. If you want to succeed but cannot see the solution, spend a lot of time thinking about how you might fail, then just don't do that.
The only problem I see with this is the potential ways to fail are unbounded, maybe possibly infinite. While the ways to succeed may also be unbounded its growth metric is much smaller. That is, some infinite sets are larger than others.
A bit stronger: Think about all the ways you could fail, then make sure those things don't happen (to the extent you can).
That is, failure isn't always caused by me doing something. Sometimes it's caused by something external. My wonderful new box might not ship if we can't get the chips to make it? I should probably look at lining up a second source. (Yeah, a couple of years ago we saw that that may not work. You can't always prevent everything bad that can happen. You can prevent some of them, though, and it makes enough difference to be worth trying.)
Not only did I buy a Zune, I was an early enthusiastic customer of their streaming subscription service and thought that it would topple the iTunes sales model. We all know how well that went.
Same here. "Split brain," logical, analytical person. We aren't even a rounding error in market numbers. If I like it it is probably the superior product providing the most value...rarely ever matters.
I see the value here being: when you can’t understand the causal relationships for success, use the causal relationships for failure that you can understand and then avoid specific failure modes.
“Friend gives generally bad advice” is not that kind of a clear causal relationship.
I’ve often thought in life that I’ve learned most of what I know by seeing examples of what not to do. If for no other reason than that there are many more people running around without a clue than truly wise individuals making well-reasoned decisions. At least that’s how it’s seemed IME; I’m sure in reality people are just doing the best they could and making decisions in the moment, but they often seem to be thinking in the extremely short term (and then sometimes, still wrong even for that).
