Position Management - Pyramiding TradesBy Simon Coulter
Pyramid trading is a method whereby traders increase their position size to gain leverage when the market moves favourably after the initial trade entry. Pyramid trading requires a sound understanding of trend risk, scaling into a position aggressively when the market moves in your direction.
Constant Position Pyramiding
Initially we will look at the concept of constant position pyramiding. Looking at the graph below we can see the resistance zone on the NZD/USD between .8260 and .8320, with support between .8055 and .8105. The first short trade is indicated by the arrow at .8055, a logical trade as support is broken.
Following the support break we see a sharp impulsive move down to .7905 before minor support forms. The two blue candlesticks that follow are a corrective move against the new down trend. Another trade of the same size is entered on a break of .7905 after which a similar pattern emerges whereby an impulsive trend following move is followed by a small corrective one, this time off the .7810 area. The third trade is placed upon a break of this support at .7810. We repeat the process until 5 trades of equal size are placed, all after the break or either major support (1st trade) or minor support (last 4 trades).
The average entry on the overall position is ~ .7782. An aggressive exit on the position would see a cut on the break of the down trend line at around .7585, which would have yielded ~ 2.5% profit (unleveraged) or ~ 197 pips. Naturally leverage would have vastly increased the percentage return on your capital.
Leaving the trade open to additional downside would have most likely resulted in us finally throwing in the towel and squaring up upon the break of resistance around .7685 yielding an unleveraged profit of around half the amount or ~ 1.25%.
Pyramiding With Trend Risk
Pyramiding with trend risk is where additional positions are placed dependant upon the strength of the trend. Trends cause traders the most problems at their beginning and end as this is where they are at their weakest and most unstable points, both when an uptrend turns into a down trend and downtrend an uptrend. During the middle parts the trend is more robust and we can be confident that the risk of trend collapse is low. As a trend matures towards its end the risk of a reversal will heighten, meaning the addition of risk becomes less beneficial.
Using the charts below lets look at how we could have used the concept to enhance our returns during the trade period discussed earlier. Our first short trade at .8183 is placed after resistance successfully holds and the CCI departs the overbought zone generating a sell signal. It is advisable to wait until the indicator leaves the extreme overbought zone before trading as traders who sell when the indicator first moves into the zone are exposed to considerable risk when the trend continues and the indicator remains overbought.
A sell signal is generated around .8105 by the simultaneous MACD/Signal line crossing, and the CCI moving into the oversold territory (Lambert’s original thesis for CCI sell signals). As these signals are generated right at the top of the support zone we choose to ignore them for the time being. An additional sell signal was given some 36 hours earlier when the –DI directional movement indicator crossed above the +DI directional movement indicator, action on this was also put to the side given the nearby support zone (see second chart below)
At .8055 upon the break of support we use the earlier weight of evidence referred to in the paragraph above combined with the break of the lower Bollinger Band and weak reading relative to the EMA to place two further sell trades. At this point as the young down trend turns into a robust downtrend we have gained the confidence required to pyramid our position as we now have faith that the risk of the trend collapsing is low.
The fourth short trade at .7905 is supported by the rising ADX reading (see chart below) and a minor support break, this equates to same place as the second trade in the constant position pyramiding discussed earlier. The fifth and final trade at .7810 is based on a minor support break; it is at the same level as the third trade in the constant position pyramiding discussion. From this point no further trades are placed as the risk of a trend reversal increases, a CCI reading well of its lows and extreme oversold RSI reading of ~ 14 helps us avoid placing a sixth trade at .7620 (trade 4 constant position pyramiding). By .7520 (trade 5 constant position pyramiding) all the indicators bar arguably the directional movement system (-DI/+DI/ADX) have warned us that the risk of a trend collapse is extreme.
Exiting the position at the same levels as we did in the constant position pyramiding trade (.7585 and .7685) yields unleveraged profits of ~ 5.21% and ~ 3.96% respectively, or over double the yield for the first cut and in excess of triple the yield for the second cut. Had the first trade signal (CCI sell) been ignored and executed with trades 2 & 3 when major support broke at .8055 our average entry would have reduced to .7976, reducing the profits on the pyramiding to ~ 4.9% (first cut) and ~ 3.65% (second cut).
There are many ways in which to execute pyramiding with trend risk all of which are based around the risk of trend collapse. The key as we have seen is to add multiple positions early in the trend as the new young trend moves into its most robust phase. As the trend moves towards maturity traders should add less positions or alternatively they may wish to reduce the size of the incremental trades. Positions should not be added to until a trader is in profit. Stop losses should be moved to your breakeven (or nearby) as the position size increases after having given consideration to any key technical support and resistance points near the stop loss level. Traders should take this into account when pyramiding so that the total position stop loss order is at a sensible point, whilst also remembering to have ensured that their position is within maximum acceptable limits.