Lessons from the Pros

Questions on Correlation and Fibonacci

We have some great questions this week relating to Fibonacci and correlation. Read on!

Q) Hello Ed, I do have a question regarding correlated currency pair trades. You have stated in the past that highly correlated pair trades can be dangerous. My question is this; what if the trades are not so correlated such as: LONG GBP/JPY, SHORT USD/JPY or LONG USD/CHF, SHORT USD/JPY? Would such trades be acceptable even though both have the same base or counter currencies? Thank you for your time.

Ed Ponsi) Thank you for your email; I do have strong rules when it comes to correlation and currency pairs. My main goal is to avoid entering trades that are too similar to each other; the problem occurs when we place multiple trades that are on the same side of the market. For example, the Japanese Yen has been very strong for the past few months, and there have been numerous opportunities to go long the Yen vs. the Euro, the Great Britain Pound, the Australian Dollar, and other currencies. But if a trader goes short EUR/JPY, GBP/JPY, and AUD/JPY at the same time, meaning that the trader is long the JPY on all three trades, does that really constitute three separate trades? In my opinion it does not; instead it is one big basket trade featuring a long JPY position. If a trader takes all three of these positions, and the JPY just happens to have a bad day that day, guess what? All three trades become losers. That is exactly what I’m trying to avoid.

Here is the rule we must remember – in every currency trade, we are long one currency and short another. If we are long the currency pair, it means that we are long the first member of the pair (the base currency) and short the second member of the pair (the counter currency). Let’s take a look at the examples you gave, and see if they constitute correlated positions.

Trade 1 = Long GBP/JPY (this means that we are long GBP and short JPY – see figure 1)
Trade 2 = Short USD/JPY (this means that we are short USD and long JPY – see figure 2)

Figure 1: GBP/JPY has been in a relentless downtrend. Source: Saxo Bank

Figure 2: USD/JPY has been trending lower on the daily chart. Source: Saxo Bank

In this example, we are short JPY in the first trade and long JPY in the second trade. This is perfectly acceptable; not only are the positions not correlated, they are hedged to a degree. In other words, since you are long JPY in one trade and short JPY in the other, the overall account is somewhat neutral regarding JPY as the positions may cancel each other out. Let’s take a look at the second example:

Trade 1 = Long USD/CHF (this means that we are long USD and short CHF)
Trade 2 = Short USD/JPY (this means that we are short USD and long JPY)

Again, this is perfectly all right because the trader is long USD in one trade and short USD in another, meaning that this trader is hedged to a degree regarding the USD. Please keep in mind, there is nothing intrinsically wrong with placing basket trades – as long as the trader is aware of it and has taken the necessary precautions. This could mean taking on all three JPY trades as mentioned above, but with the added precaution of entering trades that are one-third of their normal size. Good luck!

Q) Ed, since you feel like the technical aspects of Fibonacci have not been mainstream until about 15 years ago, it might be worthwhile to explore data from 20 to 30 years ago to see if Fibonacci levels worked. If they are obviously present, then you could conclude that their existence is not a self-fulfilling prophesy, but truly the result of these ratios being an integral part of the way God made us and hence inevitably showing up on the charts.

I tend to favor them being real and not self-fulfilling because I have experienced enough relationships that exist when retracements are drawn NOT from a peak, but rather from an Elliott Wave failed 5th. These "not so obvious" fib retracement levels were achieved by the "unconscious" minds of traders and not from lines on their charts. Hope you enjoyed my commentary about your well-written article.

Ed Ponsi) Thank you for your email. I believe that Fibonacci has worked for much longer than the past 15 years in the Forex market, but I also stated that the software to draw all kinds of exotic variations on Fibs was not available to the general public until the past 15 years – about the time that the general public went online en masse. The reason why it has been successful for many years is because institutional traders – not the general public – have used it for many years. These big traders have the power to move markets due to the size of their orders, including causing the price to bounce off of a Fib support level. On the other hand, your order or mine is not sufficient to cause a change in price direction. So, I would disagree that Fibonacci should only have worked over the past 15 years, because the market doesn’t care about individual traders like you and me, or what type of fancy new software we own. It is driven by the big players, and they have been using Fibonacci for many decades.

Here’s another point; even though Fibonacci is found in all aspects of nature, from architecture to music to natural science, does that really mean that a currency pair has to bounce off of a Fib level? For all of the times I’ve seen Fibonacci work effectively, I’ve seen plenty of instances where the price completely ignored Fib levels. I guess the Great Fibonacci Debate will continue for a long time to come!

Clarification

I received a few emails from viewers about my CNBC Squawk Box appearance the other day, so let me be clear – I never said that Citibank was going to zero. Citi had closed the previous trading day at $3.77 per share. When asked by the host where support on the stock was located – meaning an area under $3.77 where the price would likely stop falling – I replied, "I hear there is strong support at zero". This tongue in cheek response got a big laugh from the host and many of the CNBC cast members, but it was not meant as a prediction. What was a prediction – and a pretty good one at that – was when I told viewers to short Citibank at $22.50 on the same show back in April. Booyah!

Have a question about Forex trading? Send an email to eponsi@tradingacademy.com and we may use your question in an upcoming newsletter. Until next time, best of luck to you in trading.

Ed Ponsi