Asymmetric training is potentially lethal. Over decades in the RAF it is a statistical fact that sim or practice asy accidents outstrip real asy accidents by a huge margin.
RAF asy rules are incredibly strict for just this reason. You never bust the asy speeds and heights. For example, the Canberra, perhaps the worst asy ac in recent years, the VCH was 600 feet. And that's genuine - I have been with a trapper who demonstrated just why it is so - and believe me, it was a hair-raising demo. The term 'parting grass' springs to mind.
I am astonished that the CAA and/or manufacturers haven't introduced rigid rules to protect pilots that have no experience or concept of the dangers involved - after all, it is exactly what test pilots are supposed to do, defining the limits to be flown by the average pilot. It should not be a question of the skill level of the pilot - some of the most experienced pilots in the world make mistakes and kill themselves because they break the rules. Rather the rules must be pitched at the average pilot; by all means add additional safety margins to the less experienced, but you have to have a definitive figure from which to start your calculations.
Having experienced years of sim, practice and real asy flying on a variety of 2 and 4 engined pistons and jets, I had developed a healthy respect for the published limits. I, and most of my friends are alive today because of them.