In the first snapback rally at the beginning of August, the S&P 500 and Nasdaq 100 gained more than 25% in a period of just three weeks. That led me to indicate were now in an overbought environment, at least short-term (1 week to 1 month), as the McClellan Oscillator had reached its highest, most overbought reading since October 1998. When the Oscillator reaches extreme readings, it can reflect an overbought (extreme positive reading) or oversold (extreme negative reading) condition, the latter generally in the area of -100 and below. Sure enough we peaked in late August and pulled back. High oscillator readings again in late November preceded the recent sell-off.

The extremeness of that October-November rally was illustrated by the fact that the Nasdaq 100 advanced from under 800 in the beginning of October to the 1155 area at the end of November, a 360-point gain in less than two months, or about 40%. If you draw lines across the lows starting in mid-October and then across the same tops you get a near parallel channel. Channel breaks can also help predict market tops and bottoms. After moving 360 points, the Nasdaq 100 gapped up to the top of that channel and then the next hour came down sharply. In other words, it reached resistance at the top of the channel by gapping up into it, or what we call an exhausting gap, and then the next hour came down sharply, which triggered a sell-off that was followed by 10 days of selling as of the close on Friday, December 13.

Not surprisingly, we had a snapback rally that saw the S&P 500 and Nasdaq 100 gain more than 25% in a period of just three weeks. That led me to indicate were now in an overbought environment, at least short-term (1 week to 1 month), as the McClellan Oscillator reached its highest, most overbought reading since October 1998. When the Oscillator reaches extreme readings, it can reflect an overbought (extreme positive reading) or oversold (extreme negative reading) condition, the latter generally in the area of -100 and below. (View table of McClellan Oscillator data.) Sure enough we peaked in late August and pulled back.

On the rally after the pullback I was looking for the market to at most double-top on the S&P 500 and on the Nasdaq 100, and perhaps not even do that. One of the patterns I noticed in early September was a head-and- shoulder top forming on these indices’ hourly charts. This meant we could top out at the shoulder and not double-top at the head, which is exactly what we did.

The Nasdaq 100 broke the bottom of that channel and then on three different occasions over the next two days got right up to that line. It also got up to the declining 40-day moving average and to the mid-channel resistance, all three basically converging at one point, and then failed on Thursday December 12. On Friday it gapped down and followed through to the downside to make a new low for the move, and also broke short-term support in the 1020-25. That led me to believe we might go down to test the double-bottom we had at the end of October and mid-November at around 970-75 in the ensuing few days.