Anyone who has traded forex knows the highs and lows that this market can generate in a short period of time. I was first exposed to the forex markets in the mid-1980's, before the famed Plaza Accord in September 1985. Back in those days, all of the cash trading was done through banks, so many individuals traded futures. I, however, favored the cash markets because of cross rates like the GBP/JPY, which were tough to trade in the futures. This cross between the British pound and the Japanese yen is one of the more volatile cross rates as a 1% move in 20-30 minutes is not unusual. For those of you who have traded the GBP/JPY, I am sure you know both the elation of seeing quick 1%-2% gains as well as the pain of having those profits evaporate just as quickly. Therefore, trying to refine both entry and exit points using Fibonacci analysis has always been appealing, because when I have ignored these numbers, I have often regretted it! 

In part two of this series (click here for part one), I would like to further demonstrate how Fibonacci analysis can help traders improve their entries and exits in this cross rate as well as in others. First of all, many of you may be familiar with some of the basic Fibonacci relationships, but not some of the more esoteric relationships, so we'll begin with a brief refresher.

Most of you know that the Fibonacci numbers start with 1, and then adding 1 to 1 gives us 2, then you add the prior number to the current number (1 + 2) to get 3,  (2+ 3) then to get 5, and so on. In the financial markets, the golden mean is especially important, and it is obtained by dividing one Fibonacci number by the previous number, say, 55 divided by 24 = 1.618. Just as important is the reciprocal of 1.618, which is .618. There are some other relationships that many have found to be important in the financial markets and will be used in our discussion as well. They are:

  .382 = .618 squared
  .500 = 1 ? 2, the second and third numbers in the series
  .786 = square root of .618
1.000 = 1.618 x .618
1.272 = square root of 1.618
2.618 = 1.618 squared


Figure 1 - Click to Enlarge

This hourly chart covers the period from early January through early February 2009. The cross had hit a low of 13000 on December 30th and reached a high of 14140 on January 7  (point a). The cross then reversed from these highs and dropped to the 13700 area, where it consolidated for two days before accelerating to the downside. By January 13, the cross bottomed at 12880 as the late-2008 lows were broken. After just two days of basing action, the cross started to rebound. In light of the prior decline, this was a rally that appeared to be a selling opportunity. The 50% retracement of the decline from point a to point b came in at 13520, with the 61.8% resistance at 13670. The rebound was very sharp as the cross quickly reached 13570 (point c) and then, after a sharp setback, made marginal new highs at 13586 (point d), which was 84 ticks below the 61.8% resistance.

The violation of the low between point c and point d at 132 confirmed that the decline had indeed resumed. Using the decline from point a to point b, you can project down from the high at point d (these targets are in blue) to find the 61.8% target at 12795 and the 100% at 12390. Alternative downside targets can be determined using the distance from point b to point d and then calculating the 127.2% and 161.8% retracement levels, giving you 12660 and 12420, respectively. When combining the two methods, you get downside targets ranging from 12790 to 12390. Personally, if I am fortunate enough to be in such a trade, I take one-third to one-half off at the 127.2% retracement level and try to stay with the remainder. Looking at the hourly chart, the lower highs are obvious, but if you were watching a five-minute chart, the wide swings would likely have made you close out your short position well above the eventual low at 11933 (point e).

NEXT: Analysis Continues Using 15-Minute GBP/JPY Chart |pagebreak|


Figure 2 - Click to Enlarge

This chart covers an even shorter period of time, from June 16 to June 22, 2009, and the 15-minute chart further illustrates the volatility of the GBP/JPY cross rate. The daily chart of the cross shows a doji formation at the June 12 highs as the cross rate hit a high of 16256 before reversing sharply over the next two days to reach a low of 15640. The cross rallied sharply for the next few hours, briefly moving above the 61.8% resistance level after the May housing data release on June 16 (point a), before reversing. Before the opening in Japan, the GBP/JPY was trading close to 157 (point b), but just eight hours later, the cross had popped above the 61.8% level (point c) before reversing to the downside. Using the decline from point a to point b and projecting downward (in red), you get the 61.8% target (15720) and the 100% target (15595). Then you also can use the rally from point b to point c to determine (in blue) the 127.2% (15634) and the 161.8% retracement levels (15550).

Let's take a look at how this worked. The cross bounced 90 ticks from the 127.2% level before resuming its decline, reaching both the 100% and the 161.8% targets (point d) before turning higher. The following rebound lasted the rest of the day and carried over to early European trading (point e) as the 50% retracement resistance of the decline from point a to point d was reached. Just over 24 hours later, the cross was 300 pips lower as it tested the 127.2% retracement of the point d to point e rally (point f).


Figure 3 - Click to Enlarge

Of course, this methodology can work on any time frame, and the addition of a momentum indicator can give traders the additional confidence needed to execute. This five-minute chart of the GBP/JPY covers less than one full day (all times are EDT). The cross peaked around 15870 (point a) and pushed quickly down to the 15750 level before retesting the lows at point b. The RSI3 formed second positive divergence at the lows (line 1), and so a rebound was likely. The roughly 70-tick bounce stalled at the 50%-61.8% retracement resistance of the decline from point a to point b, and the second higher peak was accompanied by a short-term negative divergence (point 3). Further indication that the rebound was over was the breaking of the RSI3's uptrend (line 1). As discussed in earlier articles, often times the RSI3 will rally back to the 55-70 level. This was the case here, as the rebound to the 55 area created a good selling opportunity. Two hours later, the RSI3 once again bounced to the 55 area (see circle) as the cross retested resistance at 158 before turning lower. By 10 am EDT, the cross was testing the 161.8% retracement target at 15680.

MORE: Lesson Concludes Using Hourly AUS/USD Chart |pagebreak|


Figure 4 - Click to Enlarge

The examples of the GBP/JPY focused on corrections within a downtrend, so in this example, using the Australian dollar, I will look at corrections in an uptrend. This chart covers from April 28 through May 22, 2009. From the lows at 6987 (point a) the Aussie had a dramatic rally to a high of 7381 (point b) before a brief and shallow setback that held above the 38.2% support level (point c). Using the length of point a to point b and projecting up from point c, the 61.8% target comes in at 7487, just above the high at point d, which was 7478. The ensuing correction from this high was a bit sharper as it retraced 61.8% (point e) of the rally from point c to point d. The uptrend quickly resumed, and using the distance from point c to point d, the 100% target was at 7570, while the high at point f was at 7587.

After a 38.2% correction, the Aussie exceeded the 100% target from point a to point b, but just touched the 161.8% target (point g)using the length from point c to point d at 7714. The correction from the highs at point g was more complex, as a very nice flag formed. The correction ended on May 17 at 7467, which was just above the 38.2% support from point a to point g and the 161.8% projection from point g to point h (measuring down from point i), which were at 7437.

I hope these articles have encouraged you to consider adding this type of analysis to your trading. I will likely continue the discussion next time and hope to put some long-term examples together that we can track the rest of the year.