Two separate things. If there is to be meaningful discussions about 'risk' with others, then we have to be on the same page as far as definitions. Controlling risk is, of course, completely different all together ... unless one wants to discuss 'controlling risk' with others.
I don't disagree with your definition of risk in post 162. Given that definition is it possible to measure your risk aside from you can lose everything?
That definition states that risk includes the possibility of losing "SOME or all" ... That, in and of itself, implies that risk, according to that definition, isn't relegated to only meaning an 'all or none' risk. I believe the problem here is that people think risk is one dimensional. There is a certain probability that you can lose all, in one trade. There is ALSO--at the same time--a certain, greater, probability that the price will jump over your stop, but you will lose less than all. There is ALSO--at the same time--a certain, even greater, probability that the price won't jump over your stop, and you'll lose even less than the previous value. It's more so a risk profile, rather than the concrete, "I'm only risking so-and-so per trade" that people are used to thinking.
You can define your risk. Just know that an outlier can occur outside of your (simple) definition. So far: 1. Risk, as originally defined, should continue to be defined that way. 2. That definition is not saying that risk only includes RISKING IT ALL. 3. Risk, as traditionally defined, doesn't speak to how a measurement is obtain. I can say, you're overweight; without speaking to how that conclusion is achieved. IOW, there are many ways risk can be measured, for better or worse. 4. You can define risk as a multidimensional vector; or you can average it to a scalar. Regardless, it can be defined. The real concern should be how well that definition works for your system.
Not using standard language just makes it harder to make comparisons across systems. Generally, folks that don’t like their performance against a benchmark like to tweak language to make their systems look better. But that’s only fine if you care more about your ego than your pnl.
If it truly works. But that's the question. Working so far, doesn't necessarily imply that "it works." For example, a trader can martingale for years. And it can work. Until it doesn't.