I think the easiest and simplest solution is simply to reduce each system’s allocation on a pro rata basis. However, this might under reduce systems that have been around the longest and over reduce systems that are relatively new, because the former will have built up profits and the latter won’t to the same extent . Perhaps one solution to this is to pro rate only the starting allocation and not the accumulated profits?
I don’t think making some subjective decision on best (worst) performing systems is the way to go. If a system is meeting back test expectations, then it should be continued? Perhaps it’s different if so much of the capital has to be removed that it isn’t feasible to keep trading all of the systems currently being used? Then it might make sense to close down the ones that are the worst performed or were the least attractive in back testing.
I think the idea of rebalancing has a lot of merit, but there is the risk of not compounding the systems that have performed the best…
I can’t think what else could be done that isn’t some variant of these options.
Best wishes,
Terry