Possible SOS Signal Generated at the Close

Update January 6th, 2016 250PM EST:
The hourly score has already fallen further to -65%, so, although I still think the SPX 1970s are likely, the score could now hit -70-80% before then allowing for a sizable rally. Of course, the hourly indicators can get extremely oversold at -100% for many hours and allow much lower prices, but my man point is that the easy money to the downside has likely run its course here in the SPX 1980s and anything lower is gravy for short positions. And, since all trends are down, by my rules, it is not a great time to initiate longs until we get a bullish swing signal and an hourly score <=50%. Good fortune.


Update January 6th, 2016 945AM EST:
Another swift drop. Many say this type of thing which has occurred 2 or 3 times recently and many times since summer usually occurs in bear markets. I'd say a cluster of them more than a few weeks apart probably does, but I have no proof. I remember the Fat Finger episode in February 2007 that bottomed within 1-2 weeks and then rallied into summer. When disruptive global events like the North Korea news this morning have led to swift drops in recent memory, they have been recovered pretty quickly. Of course, a potential bear market backdrop is making that less and less likely, but I still think it feeds into our expectation for SPX 1970s and then a 1-3 week rally. Yesterday ended up being a medium-sized SOS day confirming the signal I mentioned below. In EW terms, today's opening drop and small bounce underway look like waves 3 and 4 of a larger move, in which case anything below 1986 is fair game for a 5th wave bottom on the next intraday drop, but of course EW is fluid and the larger count is in question. The hourly score rose to around -20% yesterday allowing for today's retest as I suggested, and now the score has dropped back to -35% but looks like it will take 1970s or lower to get back to -70% or more where we'd be better setup for a large rally. Even then, a daily trend change confirmation is needed to get bullish short-term. Good fortune.
************************
Today is currently a small Selling On Strength (SOS) day for SPY. Although I retreated from publishing the SOS/BOW indicator that I put together a couple months ago due to questionable back-testing, I have still been tracking it to see if it has some worth in projecting tops and/or bottoms or if I can tweak it. If today remains an SOS day (I personally think the same applies if it is an SOS day that gets wiped out narrowly at end of day when price narrowly goes negative before the close), then my indicator as it stands would call for an imminent bottom.


The last 14 times this signal occurred last year were 11/16, 10/1, 9/9, 8/10, 7/10, 6/5, 5/19, 4/29, 4/2, 3/18, 3/6, 2/4, 1/21 and 1/8. You are welcome to view those dates on a daily chart. About 1/3 of those signals rallied hard for 1-2+ weeks, 1/3 traded in volatile fashion sideways ending up where price started after 5-10 trading days and 1/3 traded sharply down for a few days but then recovered within a week after that. Every downtrend that was sustained for 2+ weeks in CY2015 (there were arguably 6 in Mar, Jun/Jul, Aug, Sep, Nov and Dec) did not have one of these signals occur just before or just after it started. The August debacle started with 1 or 2 small down days after a bottom signal occurred 6-7 days prior. I don't consider that a failure since there were 5-6 sideways days. The indicator seems to be most accurate looking out 5 trading days or so. In 2/3 of the signals, you would have been break-even or profitable after ~5 days while in the other 1/3 you would have been  around break-even within 1-2 weeks after suffering a multi-day 2-3% drop. So, there were no real losing long trades if one had the patience to wait through occasional 1-3% multi-day dips. Of course, 14 is a small sample set and one would want a stop at some point below such as 2-4% and one would probably want to only use this to augment another tool.


Having said that, applying this 5-10 day bottom signal to today's situation, it indicates SPX has great odds to trade sideways-to-up for 5-10 days and even if it dips first the remaining odds are that it will not drop far and it will recover within 1-2 weeks. That suggests my other evidence for a drop to SPX 1970s followed by a rally that lasts 1-3 weeks with SPX 2040-2050 around OPEX Jan 15 is almost a worst case scenario for the next couple weeks and that a more shallow dip or perhaps no significant dip at all could be seen for 1-2 weeks. Personally, I feel like the signal strengthens the case for SPX 1970s followed by a strong rally, but I'll change my mind if the trend or hourly score tells me to. Good fortune.