More on Gap Bands

A few nights ago I looked at upside gaps outside of Bollinger Bands for SPY. Tonight I will show them along with downside gaps below Bollinger Bands. I am using the standard 20ma and 2 standard deviations bands for the study. As was pointed out, the band levels are fixed until the close, so the gap criteria is simply a gap beyond yesterday’s closing band. I tested using SPY going back to 7/1/98. I’ll let the table speak for itself tonight.

Equity Put/Call Ratio Spikes Down

Dr. Brett Steenbarger over at Traderfeed has done some excellent work with put/call ratios over the last few years. One way he measures it is by comparing a short-term moving average of put/call volume to a long-term put/call moving average. He has demonstrated numerous times how spikes in the short-term averages over the long-term averages tend occur near market bottoms.

The options market is now experiencing a spike down, rather than a spike up, in the put/call ratio. An example of this could be seen by considering the 5 and 65 day moving averages of the CBOE Equity put/call ratio (two averages that Dr. Steenbarger sometimes uses). As of tonight’s close the 5-day stood at 0.62 and the 65-day at 0.80. This is over a 20% gap from the 65 to the 5.

If spikes up in the put/call can predict a bottom, could a spike down predict a top?

Not based on the limited data available. The equity put/call data has only been tracked back to 2003. Below are the results:



The dates for the 4 instances were 11/10/04, 12/16/04, 5/24/05, and 7/14/05. Basically it happened at times where the market rallied strongly following a pullback or correction. I suspect the sharp drop in the average put/call ratio represents a shift in attitude of market participants. Coming out of a corrective period it appears this buying enthusiasm is a good thing and not representative of complacency.

I also ran the same test using the Total Volume Put/Call Ratio back to 1996. Results there leaned to the bullish side as well with between 57% and 67% of cases showing continued gains over the next 2-3 weeks.

The low put/call ratios may become an issue if they remain low for an extended period. I’d be careful of trying to call a top based on any spikes lower in the ratio, though. Spikes may represent enthusiasm rather than complacency. Now if we could just get some volume we could really make a case for enthusiasm.

Quantifiable Edges Subscriber Letter April Results

I’ve presented below the summary results of the Quantifiable Edges Subscriber Letter trade ideas for April and since inception. April was a decent month. Although there were fewer trade ideas and rewards were a bit smaller than March, it was still strongly positive.



In addition to the trade ideas, the Subscriber Letter provides additional research beyond the blog and shows CBI analysis down to the sector level. More features will be announced soon. Should anyone wish to receive a free three-day trial to the Quantifiable Edges Subscriber Letter, just send an email to QuantEdges@HannaCapital.com and include your name and email address.

A Look At Light Volume

The market pulled back today on light volume. Often heralded as a good thing, readers of this blog understand over the short-term it isn’t.

So I spent some time looking at volume tonight. I’ve been noticing how it has just fallen off a cliff since Easter. I decided to compare the volume the last 30 days to the volume of the previous 50. The last 30 days have posted average daily volume less than 75% of the average of the previous 50. I thought this unusual and did a study to find other times this may have occurred.

Looking back to 1960 I found 10 other instances. Over the next 2-6 months prices rose 7 of ten times. There was also a general decrease in historical volatility 8 of the 10 times. Gains were generally in line with random and 7 for 10 isn’t mathematically significant, so I’m not sure there’s a lot to be learned here. I did find it interesting that all prior instances happened between 1960 and 1980. Dates are listed below:

6/26/61, 8/10/62, 7/26/63, 6/29/66, 2/25/74, 8/25/75, 4/19/76, 7/27/78, 12/13/78, 4/21/80.

If someone sees something I don’t I’d love to hear about it. I did find it interesting that many of the occurrences happened in a generally difficult period for the market but returns were pretty good. For instance, the 70’s had 5 instances – 4 of which led to higher prices over the next two months.

The Gap Band

I received an interesting note from Dr. Steven F. over the weekend who observed the SPY gap open put it above it’s upper Bollinger Band on the gap. For an index to gap open outside of its Bollinger Band, price will almost always need to be somewhat stretched already. It would seem logical that a gap up into an already short-term overbought condition would be a likely candidate for a reversal. I decided to test.

The criteria was simple. A gap up over the upper Bollinger Band would signal a short entry. The trade would be exited near the close of the day. Looking back to 1998 in the SPY I was able to identify 79 such instances. There were 43 (54%) winners and 36 losers. The average winner made 0.56% and the average loser lost 0.50%. The profit factor was a modest 1.34.

I then broke it down by instances above and below the 200-day moving average. Above the 200-day moving average there have been 62 instances of a gap up above the Bollinger Band. Shorting these and covering on the close resulted in 53% winners. Winners outsized losers by 0.57% to 0.38%. The profit factor was a decent 1.7.

The trouble occurred with this less than “Outstanding” Gap Band strategy when it was attempted below the 200-day moving average. There were 17 instances. Ten winners, but the average loss was 1.0% vs. an average gain of 0.5%. Overall a losing strategy below the 200ma.

As with previous gap studies, it appears gaps up in long-term downtrends are dangerous to try and short. While it would've worked out on Friday, you always need to be wary of a short-covering rally or trend day up.

Initial results of buying a gap down below the lower Bollinger Band appear better. I will look at them in more detail at a later time.

The 200ma Cross

The Nasdaq 100 led the way today as it rose over 3% and closed decisively above its 200-day moving average. It is the 1st major index to retake the 200-day (the S&P mid-cap 400 also did it today if you consider that one major). Some people believe the 200-day moving average is an important technical measure. Some suggest it’s a somewhat meaningful psychological level. Others see little value in looking at it. Rather than discuss and postulate on the merits of an indicator, I prefer to test it. Let’s see what a cross of the 200-day moving average in the NDX has led to in the past. (Test period is mid 1986 - present.)

Over the period tested the average gain per day in the Nasdaq 100 was just under 0.09% ($90). As you can see, the market seemed to gain steam after crossing the 200-day moving average and strongly outperformed the typical period.

What if we add a filter to eliminate those times where it barely peeked across? I filtered to only look at crosses that finished at least 0.5% above the 200ma. Results below:


These results are even stronger.

How about if we also require that the NDX makes a good-sized move on the day of the cross? I used a 2% gain to test this. More results:


Whatever the reason, a move through the 200-day moving average has provided the NDX some extra fuel in the past. When the move was strong and decisive like Thursday, that made for even better results. A cross above the 200-day moving average isn’t a magic buy signal – but there are worse ones.

Image courtesy of DialBforBlog.

Fed Reaction Not As Ugly As It May Seem

Last night I showed that when the Fed disappoints and the market drops by 1% or more in reaction to it, then the market has generally recovered and worked its way higher over the next couple of weeks.

The Fed disappointed today, although not to the degree we looked at last night. The S&P finished down a relatively mild 0.4%. While the end result wasn’t that poor, the fact that it was up 1% shortly after the announcement and then faded late left many traders with a bad taste. Measuring where in its range the market closed the day is one way to measure the mood near the end of the day. Today the market closed very close to its lows. I ran a test to see how the market performed after closing near its lows on a Fed day. Results below:


Looks fairly positive. Since I’d already established moves of 1% down or more could be a positive, I decided to exclude those instances from the results and re-run it:


Not quite as strong, but still a slight upside edge seems apparent.

Also notable about today was the fact that the market made a recent high. I ran a test to see the following 1) Fed day 2) Made 20-day high. 2) Closed in bottom 10% of range. I was only able to find 3 instances. Therefore I removed the Fed day requirement. Below are those results:


The first few days were choppy, but certainly not bearish. After that the market generally rose.

Some technicians may suggest today was an ugly late-day reversal. I’m having trouble finding much ugly about it.