« The Four-headed Monster, Called 'Technical Analysis' (Part 2 of 4) »

[This is Part 2 of a 4-part series. Follow the links for | Part 1 | Part 2 | Part 3 | Part 4 |]

Trend Analysis

One of the primary pillars of technical analysis is ‘trend analysis.’

Without going into the details of how various durations of trend are deciphered, suffice it to say that the ‘intermediate trend’ was neutral, during this phase, and the ‘minor trend’ was bearish. The contention of the proponents of the head and shoulders top was that the intermediate trend was set to change from neutral to bearish and the minor trend was going to experience a ‘bearish continuation.’

(For reasons that will be elucidated later, skilled technical analysts were preparing for the minor trend to actually reverse to bullish and for the intermediate trend to remain neutral.)

Now, aside from duration of trend, there are several other considerations in the area of trend analysis. We’re going to focus, for a moment, on one aspect from within this area: Support.

‘Support’ is a level on the chart under which buying pressure is intense enough to overcome selling pressure. A support level typically shows up as a lateral level on a chart, at which a security that is falling tends to reverse course and rally. In other words, it is a level at which several “reaction lows,” as they are called, have formed and, as such, might be seen as a level at which traders would want to consider closing bearish positions and/or initiating bullish positions.

Illustration 2: S&P-500 Index (SPX), as of Aug 31, 2010, highlighting intermediate support level.

Now, needless to say, support levels have a shelf life. Sooner or later, the bears muster enough strength to push prices through the level and render it defunct. However, one of the tenets of technical analysis is that traders should not try and anticipate a breaking of support; rather wait for price action to confirm a bearish breakout (for prices to fall through support), before actually initiating a short position - or a bearish options position - on the security.

As such, analysts who were taking a bearish stance prior to the actual breaking of support in the case of the purported head & shoulders pattern on the S&P-500, were misusing the respective tools of technical analysis and, thereby, doomed to end up paying the price.

As illustration 2 shows, the S&P-500 found support at around the 1045-mark in Feb and then again in late-May and early-June. On each occasion, a rally of over 50 points took place on the index and not once was the level breached, over that period. Clearly, the 1045-level was acting as a source of ‘support’ to the index.

As such, when prices tested the support level in late-June, no one would have faulted a trader for looking to buy stocks, in the hope of a rebound. As it turned out, support was breached on this occasion and unless he had set a wide stop, it is quite possible that the trader’s stop loss would have triggered and he’d have taken a small loss on the trade. That’s okay; losses are a part of trading and, as long as one is entering and exiting positions based on properly laid-out rules, the law of averages should end up working out in one’s favor, over time.

But look at what happened within a week of the breach of support (late-June)… The index ended up finding support near the 1010-level and quickly bounced back above the old support level at 1045, on its way to a peak at 1125 (which has since shown up as a significant “resistance” level).

Short-lived breakouts such as the one that took place in late-June are known as “false breakouts” or “fake-outs.”  Fake-outs are relatively common on the charts of stocks, stock indices and other securities. For this reason, the seasoned technical analyst not only waits for a breakout to actually take place before taking on a position (rather than pre-empting a breakout, just because a potential pattern has developed) but also utilizes tools from other areas of technical analysis to either confirm the breakout or raise the possibility that it may be a false move.

Coming back to illustration 2, we see that after rallying to 1128 and finding resistance at that level, the index started to fall back towards the lows, once again. Once the index started to approach 1045 in the latter part of August, several market watchers started to raise the alarm that a potential head and shoulders pattern was showing up on the charts.

As we’d mentioned earlier, the form of the pattern that had showed up between Nov ’09 and Aug ’10, was close enough to the textbook form of the H&S top, to leave room for the possibility that the pattern was legitimate. However, an essential element was still missing…

Now, the cardinal rule of chart patterns is that a breakout must take place, in order for the pattern to be considered complete and for a position to be taken, thereafter. In the case of a head and shoulders top, there has to be a breaking of the neckline, in order for the pattern to be considered complete and for a bearish position to be taken.

In other words, a breaking of the support level that is identified as the neckline is required. Now, if you were to glance at illustration 2 in search of a neckline for the potential H&S top, you would arrive at two potential candidates – 1045 and 1010. The big question would be: Which level should be chosen?

It’s simple, really. The 1045-level had held up to scrutiny a good handful of times and, on nearly every occasion, it provided a decent bounce to the index. On the single occasion that the index fell through 1045 (when it fell to 1010 in early-July), the situation reversed itself relatively quickly and the move proved to have been a false break. As such, the support level at 1045 should have been deemed to be the neckline of the pattern.

Coming full circle, it would have taken a break of support at 1045 on the S&P-500, in order for the head and shoulders pattern to be seen as complete and, as a result, for the ‘intermediate trend’ to have been seen as having changed from neutral to bearish.

As of the end of August, support had in fact not been broken, despite the fact that the index had been sitting promptly at or just above 1045, for a week or so, and, as such, the analysts who were taking on a strongly bearish stance, at that point, were being premature.

Momentum Oscillators

So far, we’ve illustrated how a study of the S&P-500, as of the end of August, may have painted a picture of ambiguity for a trader who only used ‘trend analysis’ and ‘chart patterns,’ as a basis for his/her investment decisions. Now let’s add another layer to the mix and take a look at what a couple of momentum indicators were saying about the state of the markets, at that point in time.

Illustration 3: S&P-500 Index (SPX), as of Aug 31, 2010, highlighting momentum indicators.

'Momentum oscillators’ are a group of technical indicators that are used to measure the momentum of a trend. They are extremely useful in non-trending market phases, when prices are trading within a narrow range.

As it turned out, the S&P-500 was trading in a range (mostly between 1045 and 1125) for just over three months, leading into the end of August. As such, the momentum indicators were generally well suited for use, during this period.

Illustration 3 provides a close look at price action and the corresponding developments on two leading momentum indicators – ‘Relative Strength Index (RSI)’ and ‘Moving Average Convergence-Divergence (MACD)’ – in the several months preceding August.

There are a good number of interpretations that can be derived from momentum indicators. For example, you can see a ‘positive divergence’ on MACD (third pane from the top) in early June, which gave risen to a rally of 90 points from low to high over the ensuing two weeks.

Additionally, you’ll notice that each time RSI (second pane from top) reached ‘oversold’ territory between May and July, a tradable bounce took place on the index. In fact, those developments provided the alert technical analyst with further evidence that perhaps the index was not likely to fall much further and instead might even rally, when RSI got very close to oversold levels in late-August (SPX was testing the 1045-level, at this point, as well)…

Furthermore, MACD provided the well-rounded technical analyst with another potential signal that perhaps support was going to be found: Now, while there are other more “sexy,” if you will, interpretations of RSI and MACD, one of the basic interpretations is the relevance of trendlines, drawn on the indicator itself…

Notice that if you drew a line joining the three lows that formed on MACD in late-May, early-June and early-July, respectively, and extended that line to the right, you’d find that MACD ended up “testing” the trendline in late-August (once again, at the time SPX itself was testing support at 1045).

The technical analyst would have noticed the possibility that support might not only be found on the index, but on the indicator itself, and that was another sign that confirmed the potential for an imminent bullish reversal! 

[This was Part 2 of a 4-part series. Follow the links for | Part 1 | Part 2 | Part 3 | Part 4 |]

Sign up today to gain immediate access to our well thought out Stock Picks, Options Picks, Model Portfolio and Market Commentary.