warp, I think you'll be okay. The Naz is pretty nutty, but it almost always gives us an "out" if we have the patience and timing to take it. (All too often I don't, but I think you're better at that than I am.)
Here's a question for you. I find that the only way to avoid getting spanked pretty hard during such periods -- I'm talking about back-testing here, not actual trading at the moment -- is to lengthen the moving average quite a bit, which results in far fewer scalp signals.
Does your model give fewer scalp (long) signals during bearish phases as well?
TIA,
augie
Market Data 
Markets 
AI
