I can't read the chart properly. Over which time period did you run your backtest(s)? Instead of waiting an arbitrary amount of seconds to reduce the chance of partial fills messing up your calculations you could also decide to cancel the order after a fixed amount of time, followed by your calculations. As you have then created a stable situation by cancelling a working order you also know your portfolio size with certainty.
I thought you were not supposed to post video links? You must have gotten triggered very hard on this one.
Ah, a backtest dominated by a strong uptrend. For fun and giggles you could re-run it during the 2007/2008 crisis and see what happens to your "risk of ruin".