The JTS flips to unhealthy mode, which stops out half its position and changes the internal dynamics of the model, making it easier to generate short signals and harder to generate longs. Meanwhile, the CCI coincidentally flips to "short-biased" mode, which unlike the JTS, means very little. However, one built in feature of the CCI system is that as a precaution, it will sell any long position on the transition. This means the JT Comp will move to 50% long on the close. (JT Comp ignores the JTS half-stop).
I am cutting back to 25% long XIV/75% cash. There is a good chance we will continue to sell off early next week. (If we don't, it will be a good shorting opportunity.) We have been healthy for ~90 consecutive trading days and "long-biased" according to the CCI for about 80 days in a row. After such a run, it's highly unlikely the market will just take a straight dump south from here.
I've gone through past data to see how the market reacts after flipping to these negative modes after being positive such a long time. I will try to put together a weekend post detailing each case, but here is a quick study for your consideration.
Using only the past 2 years for SPY (I can go back farther this weekend), when CCI flips to "short-biased", the average number of days until a short-term bottom is achieved is about 3 days. (Shortest time was 0, longest was 4 days until bottom). Buy at the bottom and hold for 5 days - Avg SPY return is 5.59%. Best return 8.77%, worst return 2.24%.
So the only tricky part here is figuring out if the bottom is today or someday early to mid-next week... The best way to play this is an average down strategy. Here is my plan. I'm going to hold a small position over the weekend and if we get a big drop early next week, I will get longer. If we continue dropping into Weds-Thursday, I will go hardcore 100% long. I will exit the position the following Wednesday Oct 24th. Using leveraged and/or volatility ETF's, a bottom to top trade should be able to net at least 10-20%. If we rally hard next week, I will scrap the plan. In that event, we will surely get short signals from both models that should carry very good odds of being winners. Obviously, I'm favoring mean reversion right now. Good weekend!