<<Does your model give fewer scalp (long) signals during bearish phases as well?>>
My model tends to give more long scalp signals during the bear markets, mainly because the markets are in a perpetual state of oversold and do not need to travel as far to get to a buy signal.
And you are correct, I use a 10-day moving average of my OS indicator during bear markets vs only 5 days for bulls. Furthermore, the level of OS required to enact a scalp is much deeper during Bears.
During Bull markets, I set the OS indicator really shallow. As such I was able to go long several times during the summer and score nice gains by scalping these little 1-2 day selloffs when other people were still waiting.
Unfortunately, my signals are still in "Bull" mode even though it seems the market is rolling over here. It is a set up for me to get creamed, but I've been saying the same thing for what seems like a month or two, and I've always managed to escape the scalp with a profit. So far.
FWIW, one thing I can point out about this selling as of late is the anemic volume. I track daily a ratio of Nasdaq volume wrt its 40DMA. I use this ratio in my model. Rarely in the past month has this ratio been over 1.
Market Data 
Markets 
AI
