Approaching a Level That Normally Holds in a Cyclical Bull Market

[asomani]

Just a heads up:

The low of day is 1212 as I write. Thus, the S&P is now approaching a level which, the vast majority of time should not be materially breached on a weekly close basis, when the S&P is in a cyclical bull market. This is the lower of the 20-month moving average and two standard deviations from the 200-day moving average...so about 1195.

Chart attached.

Personally, I'm still of the (perhaps now minority) opinion that the S&P 500 is still in a cyclical bull market - as I believe the weight of the evidence still favors this conclusion, at least on a technical / quantitative basis.

[asomani]

A few observations I would like to share, merely representing my own humble opinion:
As you may have noticed, the usual lower boundary for a cyclical bull market (that I had highlighted in the chart above, that I had sent last week) was closed below today on a daily basis (see attached). A weekly close below this lower boundary would confirm to me that, if one has not already fully prepared for a cyclical bear market, then it is prime time to start completing your preparations (not necessarily now, when there is blood running in the streets, but perhaps at the next test of your favorite moving average of 50 days or longer, just as an example).

This is the case even if you're like me and have a hard time believing that the cyclical bull market coming off the March 2009 lows ended back in early May. I feel this way because I have done some research on past cyclical bull market tops in the Dow and the S&P, and the May price peak - like the April price peak last summer - did not have any of the three breadth-related signatures that I look for to accompany a cyclical bull market top. It seems that every cyclical bull market top since at least 1929 has had at least one of the three breadth-related signatures I refer to.

Anyway, I think we can be sure that the extent of selling we have been seeing since Thursday is typically not sustainable (it usually ends in 7-8 trading days or less based on my research, and I would say today's close marks day 3) and will be followed by a large bounce-back rally (not so unlike the one that began in May last summer) - if the limited number of past occasions I can compare this selloff to are a guide. Technically speaking, the ~1080-1120 level (which we closed in today) and the ~980-1020 level in the S&P mark important support levels and may provide the launching point for the large bounce-back rally that history suggests is coming.

Today the VIX also hit 48 (and 50 is not a level you should see get broken above, at least in a material way, in a cyclical bull market). A VIX at 48 matches the high of last summer's downturn, which was hit on May 21, 2010. When this level was hit last summer, the S&P stopped dropping, backed and filled for some days and ultimately rallied strongly over the next month (about 7%). It then spent the following two weeks coming down and re-testing its May lows (breaking below them briefly and then closing back above them to make a successful re-test by early July...July 6th marked the low-point of the year).

[asomani]

Just to be clear, as a few people have emailed me and asked about this (as what I posted here was also emailed to several people): please remember that the chart showing the usual lower boundary for a cyclical bull market is a weekly close chart and not a daily close chart. I sent an email yesterday partly because the S&P had made a daily close below the usual lower boundary for a cyclical bull market, and this suggests that there is a good chance that at some point in the next few months the S&P will make a weekly close below the boundary - but, yes, until there is a weekly close below the lower boundary (and not simply a daily close below it), the lower boundary has not actually been broken.

By the way, if the S&P does make a weekly close below the lower boundary, a secondary signal one might wish to monitor as even further confirmation of the need to completely be ready for a cyclical bear market (even if one may not be starting, and I gave a few reasons for why I have a hard time believing that one is starting in my last post) is if the S&P closes below its 20-month moving average at the end of this month. The following is a very brief excerpt of a research article from MarketTells.com completed several weeks back: Historically, crosses of the 20-bar average on the monthly S&P chart (84-bar on the weekly S&P chart) have proven to be more reliable than crosses of the 200-day average. Out of 24 signals since 1950, note that only one led to a meaningful loss when the signal was closed out, and even that was less than 10%. By comparison, there were fourteen signals that generated a profit of more than 10%.

And, for those who asked for the settings on the chart, they are attached. In these settings, the 40-week moving average represents the weekly version of the 200-day moving average, and the 84-week moving average represents the weekly version of the 20-month moving average (84 weeks * 5 days per week = 21 days * 20 months) .