I believe quantifying and creating rules is a good idea, however, I would still leave room for discretion. To me, its like chess, the best players these days are the computers, but if you combine a computer with the best human player, that's actually a more powerful combination because the human knows where the computer is weak at (long-term strategic moves) and the computer fixes the human issues (short-term tactical mistakes). If I was doing macro full time I would try to move towards something like that, combining great mechanical rules with my discretion, there are a number of papers on macro 'warning signs' for economic disruptions/FX moves, etc, maybe that's a good place to start
I would agree with this. I believe risk has to considered individually. For a well paid doctor, it makes sense to run more risk in a portfolio because his income is so stable. For a singer, that is one tweet away from ruinning his career, its a different story. Also, one of my ideas in personal finance is that the first property that someone buys is virtually risk-free even though it might be almost 100% of that person's worth being invested in a risk asset. Its risk-free because people need a roof to live on and it generates 'income' in the form of rent saving (and is tax free). Effectively its an REIT with 100% occupancy. So if that person's portfolio consists in 100% of real estate but that real estate is their first property, I wouldn't think that's a mistake but rather a great idea.
Its like we as humans have permission to do risky things in certain situations because of our circumstances. There are many situations like that, probably many that I havent even thought off but its fun to them because it enables us to earn premia while taking little to no risk