Well you need to model what happens when you get an earnings miss and the stock drops 20-30%. Your back test filters that out but you will encounter it trading forward.
What would happen I think is I could wipe out the profit for the quarter, but it hasn't happened in the last 5.5 years unfortunately, and I cannot use TOS to backtest farther than that and see exactly what would happen. I need a new, more mechanical method of backtesting