The CR is just derived from the backtest. It is not 100% "safe" - there is no guarantee that you won't need more capital in live trading. The displayed "risk" is just the distance to the stop and has nothing to do with the statistical risk of the system. The system risk is normally higher, but can also be lower than the stop distance in some cases. For instance when the system is unlikely to hit the inital stop, or when the live stop was more distant than any stop in the backtest.

I can not say it often enough: There is no real "safe" balance and no "guarantee" against a margin call. Your trading is based on statistical probabilities, not on guarantees.

Thanks jcl.

Big Picture #1: Yes, I understand your comments re. statistics & safety. I understand this is not at all "set and forget". I attempt to monitor at least daily - hopefully that will be enough so I won't get burned... Have come too close for my tastes already...

Big Picture #2: CR is from backtest using the given Margin & Risk settings. If Trade and Test function the same, then those Margin & Risk settings should have produced during Test trade(s) with the same displayed "risk". Evidently the Test never closed such a trade at the worst possible point, hence that risk was never realized, hence CR is smaller than if it had been. However, in real Trading that could happen which would in fact lead to margin call if the system is started with no more than the Tested CR. Is that the correct way to view/understand Test vs. Trade?

Thanks.

Partially. A backtest drawdown smaller than the stop distance is theoretically possible, but unlikely. More likely is a higher stop distance in live trading than in the backtest.

Thanks jcl. I think I get it now!

But it leads to another question: Does it make sense to consider storing additional results from Test - for example the maximum risk as discussed above - basically a description of the Tested "world" - and then in real Trading compare the calculated values to these stored values to flag (perhaps via a pop-up) when the strategy is leaving its known "world" and entering a "new world", possibly requiring more Capital or needing some other attention?

Thanks.

This is indeed one of the goals system developers try to achieve - stopping a system when the market behaves differently that in the test. The problem is that live trading is always a new world. You must draw a line somewhere. A higher stop distance, as caused by an unusual weekend gap or a high volatility period, is normally harmless. So this certainly makes sense, but is a difficult matter.

Thanks jcl.