Should One Buy An Oversold Market?

By Lawrence G. McMillan

This article was originally published in The Option Strategist Newsletter Volume 23, No. 15 on August 8, 2014. 

We have often used the phrase, “oversold does not mean buy.” It is probably one of the most useful phrases a trader can employ. Many a would-be bear missed almost the entire bear market of 2008 because it got immediately oversold in September 2008 and stayed that way all through one of the worst bear markets ever, that unfolded over the next couple of months. So, from an intermediate-term viewpoint, one must wait for confirmed buy signals before stepping in front of a bear market roaring downhill.

Having said that, there are some very short-term oversold conditions that can be traded during market declines. It has been a long time since we’ve truly had any oversold conditions, but some arose last week. Over the years, we have published many articles detailing various trading systems. Some of these have to do with extreme oversold conditions. Since there haven’t really been many extreme oversold conditions in a long time, it’s possible that you may have forgotten about these indicators. A few of them came to the forefront this past week, issuing successful short-term buy signals.

CBOE Equity-Only Put-Call Ratio

This is one of the truly rare oversold signals. This put-call ratio considers only equity options (i.e., options on stocks) that trade on the CBOE. No futures or index options are included, nor are any options that trade on any of the other seven (or is it eight?) exchanges that trade listed options.

This buy signal is triggered when more puts than calls trade on a given day; that is, when the CBOE Equity- only put-call ratio is greater than 1.00. When that happens, a short-term buy signal is generated. This system was thoroughly described and analyzed in Volume 20, No. 12 (June 24, 2011).

Since that article was written, there have only been three days on which the CBOE Equity-only put-call ratio was above 1.00: two in August of 2011 (a very bearish time for stocks) and one last Friday (Aug 1, 2014). You can thus see how rare these signals are.

When that article was originally written, we observed that sometimes a single day stands alone, where the ratio was greater than 1.00. But often, there is a series of above-1.00 days. The article showed that buying the first occurrence when the ratio was above 1.00 was indeed profitable, but it was actually more efficient to wait and buy the second such day, if it occurred within five days of the first day. By “more efficient,” we mean that the average profit is greater, and thus it’s a better use of one’s capital.

The problem with waiting for the second day, is that there have only been 10 such sequences since 2001, while there have been 55 times where one could have bought the first day – not knowing if it was going to be part of an eventual sequence of days or not.

Last Friday, the CBOE equity-only ratio was 1.036 – setting up a buy signal for Monday. As you may recall, $SPX was up strongly on Monday, as it rose nearly 14 points.

Tables 1 and 2 show the results of these CBOE Equity-only buy signals, over holding periods of 1, 2, 3, 5, 10, 20, and 30 days. These tables are complete, with the exception of the August 1st, 2014 signal; it is shown only in Table 1, Columns 1 & 2 (i.e., the 1-day and 2-day results). Where rows refer to “points,” that is the number of $SPX points made or lost during the tenure of the signal.

The “Average Points” row is highlighted in yellow in both tables. You can see that both methods made a profit, on average, over all holding periods. In addition the “Percent Wins” (4th row in each table), was above 60% for all holding periods in Table 1 (except the 2-day holding) period, and was above 70% in all cases in Table 2. Any system that has 60% winners or more and makes a profit, on average, is a very strong system. You could pay big dollars for systems that do far worse.

We traded this signal as a one-day trade only, on Monday August 4th, in Daily Volume Alerts. Since our other indicators are heavily negative, I was – and am – reluctant to trade it for the potential longer-term time periods as shown in Tables 1 and 2.

You can monitor this data yourself, on the CBOE website. For closing put-call data, visit:
 http://www.cboe.com/data/mktstat.aspx

The equity only data is the fourth set of data on the page.

Furthermore, the CBOE publishes intraday option volume statistics, currently at:
http://www.cboe.com/data/IntraDayVol.aspx

The volume data is posted there every half hour. Note that the time of day on the CBOE web site is Central time. So the final posting shows a time of 3:00 pm. SPY options trade until 3:15 Central time, so you should be able to see the final posting and still have enough time to take a position in SPY or $SPX options at the end of a particular trading day, or overnight in the Globex markets, without having to wait for the next day’s $SPX opening.

The Total Put-Call Ratio

This is one of our favorite ratios, and its signals are highly reliable. However, they also do not occur very often. We use the Total put-call ratio in a couple of ways. Both will be discussed here, but only one of them has recently given a short-term buy signal.

The Total put-call ratio is comprised of all listed stock and index options that trade on a given day – on all of the exchanges. It does not include futures options. This data used to be available in the major financial publications, but I don’t believe it is any longer. Rather, we compile it from the raw closing price data that we purchase from www.tbsp.com.

As a general rule of thumb, far more equity calls trade than puts on any given day, but far more index puts trade than calls. Even so, it is rare for this ratio to be above 1.00 on a daily basis. It is also rare for its 21-day moving average to be above 0.90.

Daily Signals

In Volume 21, No. 8, we detailed the single-day system for the Total put-call ratio.

It can be summarized in this manner:

A daily Total put-call ratio of 1.00 or higher is 

• a) the previous trading day’s Total put-call ratio
• or b) the previous time that the Total put-call ratio was > 1.00 was between 3 and 28 trading days ago.

Note: we do not take the trade if the previous occurrence was exactly two trading days prior.

We keep accurate statistics on this system. It has produced a gain of nearly 640 $SPX points since March of 2000. The system statistics are in Table 3. The bottom four rows are in terms of $SPX points.

Note that the system produces 58% wins, and the average trade is a small win. What is also quite noticeable is that both the average win and the average loss are quite large. Hence, this system is active during volatile markets, and thus should be traded with options rather than futures. The problem with doing that though is that if you buy calls during these times, your are most likely buying a call with a somewhat inflated implied volatility. Then, if the market rallies the next day, implied volatility (i.e., $VIX) will drop, weighing against your call purchase. The best way to combat that is to buy a slightly in-the-money weekly option, which has less responsiveness to a drop in volatility and more of a positive response to an increase in the price of SPY or $SPX. In that way, you know your risk in advance, and so those huge down days (average loss –16.4 $SPX points) won’t hurt you as much as the large profitable days will help you.

Since we published the first article, in April 2012, that defined this system, we have been keeping track of statistics on the days that were “skipped,” either because the preceding signal was 2 trading days prior or was more than 28 days prior. For the record, there have been 11 such days that we “skipped” and didn’t take the trade. There were losing trades on 6 of the 11 days, and the average trade was a profit of only 0.25 $SPX points. Hence, this reinforces skipping those trades, as they are not performing as well as the trades shown in Table 3.

Total put-call ratio 21-day moving average

The more spectacular use of the Total put-call ratio is the intermediate-term system signaled by its 21-day moving average. When a buy signal occurs here, it means something: a 100-point rally in $SPX or more.

This system is activated only when the 21-day moving average of the Total put-call ratio rises above 0.90. That ratio is not yet in range, currently standing at 0.833.

However, if this market continues to be bearish for a while, as we suspect it will, then the Total put-call ratio’s 21-day moving average buy signal should eventually come into effect.

We will detail the strategy when the time comes, but for now, here is the system:

1) Once the 21-day moving average of the total put-call ratio reaches 0.90, the system is “on alert” for a buy signal.

2) a buy signal occurs when either:
a) the peak of the 21-day moving average has lasted for 10 days, or
b) the 21-day moving average falls below 0.90

...whichever comes first.

3) stop yourself out if:
a) the 21-day moving average moves below 0.90 and then moves above 0.90 again,
or b) a new high is made in the 21-day moving average.

Note that if you are stopped out by rule 3), you will automatically be back at rule 1) and therefore will be awaiting the next buy signal.

These signals have a terrific track record, although they are not frequent, as you might have guessed. The results are shown in Table 4.

There have been 19 trades since March 2000. Six have not reached the 100-point target and have been stopped out. A couple of the ones that were stopped out actually made money, though. There were 13 successful trades for a total gain of 858 $SPX points. We would certainly relish a trade via this system in the future, but you must understand that the stock market would have to be in a bearish mode from here going forward for a while before this buy signal could set up.

$VIX Spike Peak Buy Signals

A potential buy signal is setting up from this system since the recent increase in $VIX means that it is “spiking” by our definition (a gain of at least 3.00 points in 1-, 2-, or 3- days, measured using closing $VIX prices). Once it is spiking, it is inevitable that a buy signal will occur; we just have to wait until $VIX closes more than 3.00 points below the highest intraday price it reached while it was in “spiking” mode. This system is discussed in further detail on page 12.

“modified Bollinger Band” (mBB) Buy Signal

The last mBB signal was a sell signal back in late June. It didn’t appear to be doing well for a while, but now it has blossomed. An mBB buy signal would occur if $SPX closes below its –4 standard deviation mBB. It touched that Band last Friday (August 1st) and and again today (August 7th). However, it has not closed below it.

That lower –4-sigma Band is currently at 1906.50 and dropping daily. It is dropping faster now because a) the market is dropping and b) volatility is increasing. $SPX may not be able to close below it at the rate it is currently dropping. But if $SPX does close below the –4-sigma Band, then a buy signal would be generated when it subsequently closes above the –3-sigma Band.

Summary

In summary, it is not a good idea to buy the market merely because it is oversold. You can get overrun that way. But oversold conditions can lead to buy signals, and when those occur, they are often quite profitable.

This article was originally published in The Option Strategist Newsletter Volume 23, No. 15 on August 8, 2014.