Eoin Treacy's view David has long described reversion to a mean as "the closest thing we get to a natural law of physics in the social science of investor sentiment". At Fullermoney we have a strong preference for using the 200-day moving average to represent the trend mean. There are a number of reasons for this, not the least of which is its popularity among investors and traders as well as the fact that it reflects the average of one year's trading.

As a trend smoothing device, moving averages are at their most useful in consistently trending markets. This is why we tend to employ them when talking about some of Fullermoney's long-term investment themes which have been trending steadily higher for more than two years; some for considerably longer. Moving averages are at their least useful when prices have been rangebound for a prolonged period or are forming a base. This is because such price action tends to be less consistent. Therefore, moving averages are most suitable once a consistent trend has been identified.

We regard a significant divergence from the mean as a warning that the risk of a reversion is rising. When one sees a powerfully trending market, the temptation is to extrapolate the price action and expect even higher levels. However, if one is prepared to draw such a conclusion it needs to be on the understanding that the risk of a reversion is rising. Sound money control tactics and trailing stops are usually appropriate at such junctures.

In a consistent uptrend, the farther prices rise away from the trend mean, the greater the likelihood that the current spell of demand dominance is being expended. The higher prices rise, the nearer they get to the next reversion. This is as applicable to the downside as the upside, provided the trend is reasonably consistent.

At Fullermoney, we generally view a bullish divergence from the mean as an opportunity to take profits in investment positions or to open hedge shorts once short-term trends lose momentum. Alternatively, we tend to view reversions back towards the mean as favourable entry points, once support has been found.

A mean reversion can take any of three shapes. Prices can pull back sharply to the mean and find support. Prices can enter a period of consolidation and allow the mean to catch up. Alternatively, it can be some mix of the two.

There is a considerable difference between a reversion to the mean following an over extension within a consistently trending market and an overextension that occurs following a breakout from a lengthy base. In the latter case, prices break out from a prolonged period of ranging because they have hit a vacuum of supply above the market which sends prices higher in search of an offer. This in turn prompts those with shorts to cover and prices can move even higher. In this case, a new trend may just be starting so it would be incorrect to expect an imminent pullback [pullback] to the mean, not least because powerful breakouts often rally faster and farther than many anticipate and because the mean does not reflect the changed perception of market participants.

Considering the heightened anxiety evident across asset classes and the choppy action this has given rise to, I thought it might be instructive to examine a number of markets with a view to their prospects for mean reversion. :

The German DAX Index has been trending steadily higher since March 2009 and has found support in the region of its 200-day MA on a number of occasions, which is to be expected in such a consistent move. It last concluded a reversion to the mean in September, following a relatively lengthy consolidation within the medium-term uptrend. It rallied persistently for 5 months and has completed a reversion to the mean in less than four weeks which is an example of a swift pullback. It is currently testing the 200-day MA and will need to find support in this region quickly if the medium-term uptrend is to remain consistent.

The UK's FTSE-100 has had a less consistent medium-term uptrend and spent four months ranging below the MA from last May. It was less overextended when it began to pull back four weeks ago and has now also completed a reversion to the mean. This was a less consistent chart pattern to begin with and the MA has not offered as meaningful an area of support as the DAX above. This would suggest that the FTSE-100 may need additional room for manoeuvre in the context of the overall medium-term uptrend. A clear upward dynamic sustained from more than a week or two will be required to begin to indicate that demand is returning to medium-term dominance.

The S&P500 ranged in a similar pattern to the FTSE-100 until September but its subsequent rally has been more like that of the DAX. It has also pulled back over the last four weeks but not nearly to the same extent as the above indices. The MA is rising quickly to meet the price and is currently above 1200. Provided the Index finds support above or in the region of 1200 the medium-term uptrend can continue to be given the benefit of the doubt.

The Canadian TSX pulled away from the 200-day MA in late August and posted a remarkably consistent rally which took it back to test the region of the 2008 peak above 14,000. By last week it had become quite overextended relative to the MA and the weekly key reversal was the second trend inconsistency (The first was the overextension relative to the MA). The MA has been rising rapidly to meet the price and with the sharp price decline, the overextension has been almost completed. A clear upward dynamic held for a week or two would help to indicate that support is returning in this area.

The Brazilian Bovespa Index rallied impressively from its late 2008 lows and retested it pre-crisis peak by late 2009. It has now been ranging between 60,000 and 70,000 for more than year. This lengthy period of consolidation has seen the price dip below the MA on at least two occasions and the merit of using the MA as a trend smoothing device in such a marked loss of momentum is questionable. The Index will need to sustain a move to new highs if it is to reaffirm the medium-term uptrend. Alternatively, a sustained break below 60,000 would complete a Type-3 top as taught at The Chart Seminar.

Australia's ASX200 rallied impressively from its early 2009 low but has been ranging below 5000 for 17 months. This represents a marked loss of momentum and it is a stretch to say this market is in a medium-term uptrend. In the absence of a clearly defined trend, a 200-day MA is less useful. A sustained move above 5000 will be required to confirm a return to medium-term demand dominance.

The Shanghai A-Share Index has also experienced a marked loss of momentum since Q3 2009 and has ranged with a downward bias since. The MA is not useful in this type of environment. Action since the Chinese New Year holiday has been encouraging but the progression lower highs, currently near 3350, will need to be broken to confirm a return to medium-term demand dominance.

The Indian Sensex completed a Type-2 bottom in early 2009 and rallied to test the pre-crisis high by late 2010; finding support at the 200-day MA on a number of occasions along the way. The pullback over the last three-months has been deeper than any seen in the course of the uptrend. This might be rationalised as to be expected following a retest of an important peak but it is an inconsistency nonetheless. The Index has been ranging below the MA, in the region of 18,000 for 6 weeks. This could be a distribution below the MA, or a consolidation prior to a reassertion of the medium-term uptrend. A sustained move back above the MA, currently near 18,700 will be required to indicate demand is regaining the upper hand beyond the short term.

Indonesia has been perhaps the best performing market in Asian since late 2008. As a result, it also has one of the most consistent uptrends. The Index has found support in the region of the 200-day MA on a number of occasions over the last 30 months. The most recent test was in January and a sustained move below the MA, currently near 3300, would be required to begin to question the consistency of the medium-term uptrend.

Silver hit a peak just above $22 in early 2008, bottomed later than year and has posted a progression of higher reaction lows since. Prices have found support in the region of the MA on a number of occasions. Following a doubling since August, prices are now overextended relative to the mean by just about any measure. A reversion appears to be getting underway. The last major reaction low, currently coincides with the MA near $26.50 and silver would need to sustain a move below that level to derail medium-term upside potential.

Brent crude oil found support in the region of the 200-day MA on a number of occasions since mid 2009 as well as holding an almost unbroken progression of rising reaction lows. It successfully broke upwards to new recovery highs in December and has since accelerated away from the trend mean. A reversion appears to be getting underway but a sustained move below $100 would be required to question the consistency of the medium-term uptrend.

Copper has also found support in the region of the MA on a number of occasions since early 2009. Prices paused below $4 for a brief period in November and it subsequently accelerated through the pre-crisis high. By the time it encountered resistance near $4.50 it had become quite overextended relative to the 200-day MA and looked susceptible to a reversionary pullback. The MA currently coincides with the old high and the psychological $4. A sustained move below that area would be required to question the consistency of the medium-term uptrend.

The above examples illustrate that the 200-day MA is a useful tool but is subject to the existence of a consistent chart pattern. One might well ask how one can ascertain when a reversion to the mean is a major trend reversal as opposed to a pause within an overall uptrend. The answer lies more in the deteriorating consistency associated with every peak of any significance rather than the MA which is but a component of one's overall analysis and will lag my definition.