Financial Sense   Home  l  Market Monitor  l  Market WrapUp  l  Storm Watch  l  About Us  l  Contact Us

"Divergences: What They Can Tell Us About Abrupt Changes in the Character of the Markets"

Everyone who has been through a few major changes in market trends knows that in order for a major trend change to take place, the prevailing "character" of market must have changed prior to the change in trend. In fact, it is the change in character which causes the change in trend. We can always tell that a major change in trend has taken place after support levels that correspond to the previous trend are broken. One of the nightmares that investors/traders occasionally experience is when they take positions as the market approaches a level that represents major support/resistance of the previous trend, under the belief that the previous trend is still intact. The implication of such belief, and expectation there from, is that the market will act in the same way it has acted previously, when contact with the same support/resistance levels was made. However, if the character of the market has undergone a material change, then the level that was expected to act as support/resistance doesn't, resulting in losses. When the trend-line that had defined the previous trend is violated, and then begins to act in an opposite manner than previously (support becomes resistance, and vice-versa), then we know that the trend has changed, and thus we conclude that the character of the market has changed as well. However, the realization that such change has taken place--after the fact--doesn't bring back the losses which incurred due to the "positioning" that was based upon our false assumption regarding the prevailing trend and market character. One of the most common reasons behind investors' and traders' wrong assumption is due to the fact that in most cases, the market tends to change character in a manner that makes the actual change undetected while it is taking place in real time, thus, we find out about it after the fact through the way it manifests itself--a trend-line that had defined the previous trend is violated--and then it begins to act in an opposite manner than previously (support becomes resistance, and vice-versa).

Why the market makes it so difficult in most cases to detect that it has undergone a character change, I do not know. However, such act does conform with what we know about markets; they never make it easy. If they did, everyone and anyone would be able to use the financial markets in order to become fabulously wealthy on demand  which has never been the case. All of us who make their living in this business have spent countless hours trying to figure out ways to directly detect a character change before it manifests itself. Over the years I have seen several indicators and models designed to meet the goal, ranging in their complexity from modest to outright incomprehensible with only a few meeting the challenge on a consistent manner. Moreover, none of them are available to the public because they are the intellectual property of institutional firms and they are considered a "competitive advantage." Thus, don't expect E-Signal, or Trade-Station, etc., to feature any of them in their charting packages anytime soon!

So, is there a way individual investors/traders can directly detect a change in the character of the market --in real time--with the indicators that are currently available? Based upon my observations as a student of the markets for the past fifteen years, I have come to the conclusion that sometimes individual investors/traders can successfully detect such change, maybe not in real time, but certainly before the fact, before the character change manifests itself as a violation of the support/resistance levels that had defined the previous trend.

Over the years, I have observed the following:

a) Entry/Buy signals tend to fail in a bearish environment; the market eventually falls below the entry level and continues lower. Conversely, Exit-Sell signals tend to fail in a bullish environment; -the market eventually surpasses the entry level and continues higher.

b) Positive divergences between price and a proven indicator result in an upside resolution three out of four times, if the divergences are taking place within a bullish background. Conversely, negative divergences between price and a proven indicator result in a downside resolution three out of four times, if the divergences are taking place within a bearish background.

Consequently, how price behaves in the presence of positive/negative divergences can give us a clue with regards to whether the overall background of the market--character--remains the same, or it has changed. For example, if the market has been in a bullish mode as it approaches a level that defines trend support, if we observe several positive divergences between price and time-tested indicators, assuming that the mode of the market is still bullish, we would expect support to hold three out of four times. If support doesn't hold, in other words, if price doesn't respond favorably to a bullish set-up that presents itself in the form of positive divergences, then in all likelihood, the character/mode of the market has undergone a material change, from bullish to bearish. Divergences represent "un-confirmed signals" when price moves in the direction of the divergence, then the signal becomes "confirmed." To put it simply, if price is making lower lows, but XYZ Oscillator is making higher lows, then we have a positive divergence and a potential "entry" signal. If price does turn up, then we have a "confirmed entry-signal." For example, take a look at the McClellan Oscillator as the NYSE was making its August 2001 lows. Price was making lower lows, while the McClellan Oscillator was making higher lows, producing a "positive divergence." However, since the market was in a "bearish mode," the positive divergence didn't have any positive effect on price, and the NYSE still plunged into much lower lows in September. Contrast this with what happened in March 2003. As the NYSE was making lower lows, the McClellan Oscillator was making higher lows. Once again we had a positive divergence, if the market was still in a bearish mode, three out of four times price would not have responded favorably to the positive divergence. The fact that it did gave us a warning that perhaps the market had undergone a character change, and thus it was in a bullish mode opposed to a bearish mode. Although further proof was needed, the fact that the market had responded to a positive divergence in a manner consistent with a market in a bullish mode, should have been adequate warning that perhaps the bear market had ended, and thus instead of looking for an entry level to short the ensuing rally, one ought to look from that point on, for an entry level to go long.

Chart courtesy of Carl Swenlin and decisionpoint.com

To sum it all up, if an index/stock responds to a number of divergences in a manner opposite than is usual up to that point, then there is a good chance that the index/stock has undergone a character/mode change, and thus, a change in trend should be forthcoming. Is it 100% certain that such behavior is due to a character change? NO, but in my observation it is 75% certain. Consequently, if you see an index/stock that has been in a bullish mode unable to respond favorably to positive divergences, you may not want to go short, but also, it may not be a good idea to go long either as the stock/index approaches trend support, because that support may not ultimately hold.

The reason for bringing up the subject has to do with my observations over the weekend. Based on Friday's closing prices coming into this week we had two very important developments:

1) Almost all major domestic indices were testing support at their break-out points, and the same held true for several foreign markets as well (see Germany's DAX index at the bottom)

2) There were steep positive divergences between price and the McClellan Oscillators, our own Summation Indexes, the Thrust Oscillators, the Buy/Sell Equilibrium Indexes, and the Quantifiers (see charts below)

Given that the indices are testing support while we have numerous positive divergences we had to expect a rally, even if Monday turned out to be a down day, assuming the overall environment is still bullish. As I already explained, all these positive divergences amount to unconfirmed "buy signals" (a signal from a divergence becomes confirmed when price actually follows suit). Buy signals--bottom picking--tend to fail in bear markets, just like sell signals--top picking--tend to fail in bull markets. Consequently, this week we should get some very important clues regarding whether the bull is still in charge, or the bear has quietly taken over. If the indices continue to decline and they close below support, despite all the positive divergences, then we must seriously consider the possibility that the cyclical bull is over and thus, we ought to be shorting rallies instead of buying dips.

(For an explanation of what the following indicators represent please visit: "Aegean's Market Indicators")





All charts are property of Aegean Capital Group, Inc., unless otherwise noted. All Rights Reserved, Reproduction without written permission, is strictly prohibited.

Charts courtesy of stockcharts.com

One may wonder what is the cause behind a change in the market's mode. Speaking strictly on a quantitative basis, the cause is always due to one reason, and one reason only: net flow of funds. Are investors putting more money into the market than they are taking out?

Take a look at the chart below. Notice, that for the first time in 12 months, NASDAQ is experiencing a negative flow of funds, more money is leaving NASDAQ issues, than money being committed to them. The last time we had a similar pattern was at point "A" in January of 2002, which turned out to mark the top for this market. Keep in mind that in order for prices to move higher, we need to have more buyers than sellers. If price can't respond favorably to the positive divergences currently in place, it will be because the net flow of funds will remain negative. If more money is leaving the market than money coming in, "positive divergences" won't make any difference.

One thing that needs to be kept in mind is that during times of negative liquidity, the markets do bounce when they get oversold, but because 

One thing that needs to be kept in mind is that during times of negative liquidity, the markets do bounce when they get oversold. But because there is not enough liquidity to sustain the bounce, it ends up being a few days affair. 

In summary, this week we should get some very important clues regarding whether the bull is still in charge or the bear has quietly taken over. If the indices continue to decline, and they close below support, despite all the positive divergences, then we must seriously consider the possibility that the cyclical bull is over and thus, we ought to be shorting rallies instead of buying dips.

SUPPORT AND RESISTANCE LEVELS TO WATCH FOR AS OF THE CLOSE 3-22-04

MARKET TIMING INDICATORS:

CURRENT MARKETVIEWS.TV GUEST CONSENSUS (3-22-04): 
BEARISH (10) BULLS (32) BEARS (14) NEUTRAL

CLICK HERE TO SEE THE CONSENSUS'S RELATION TO THE SP500 IN  CHART FORM 

Ike Iossif

All charts are property of Aegean Capital Group, Inc., unless otherwise noted. All Rights Reserved, Reproduction without written permission, is strictly prohibited.

Home  l  Broadcast  l  Market Monitor  l  Storm Watch  l  Sitemap  l  About Us  l  Contact Us