What Time Frame Should I Be Trading?

Anyone who has ever opened up a Forex price chart has probably noticed that you can view that action in a number of different times frames, usually from 1 minute to 1 month. All this means is; which time interval is used in the X-axis of the chart. For example, in a 1-minute chart, each candlestick or price bar shows what happened over 1 minute of time.

S which time frame should you be using to trade? This is a question whose importance tends to be underrated. The best way to answer it is to ask yourself a few questions, and be very honest with your answers.

Firstly, how much time do I have available? If you only have a few minutes to spare each day, then you really have no business using any time frame lower than the daily. Unless you want to drill down to take a better look at the details of the day’s price action in determining how “good” the day’s candle really was.

My point is, that you will make more money by competently using a daily chart, then you will be incompetently using a chart of a lower time frame.

If you have a few minutes to spare every few hours, you could use the 4-hour chart. The more time you have to spare, the lower down you can go.

You might have guessed by now that I am of the opinion that the shorter the time frame you are able to use competently, the better you should be able to do in the market. However, the key word here is “competently”.

Shorter time frames are more difficult to handle than higher time frames. It is often said that all time frames are essentially the same, but that is not true. Shorter time frames provide more opportunities to make mistakes and get confused. That is why I advise newer traders to use higher time frames and then, when that is going OK, to try to look more deeply and drill down into what they are already doing successfully.

For example, you might find that of two candles which look equally bullish on the hourly chart, the one that looks more convincing regarding the twelve 5-minute candles which compose it, will often be the more successful trade.

Of course, it is always essential to look at higher time frames. Nobody ever made consistent profits just trading a 5-minute chart unless they were also looking at higher time frames all the way up to at least a few weeks of recent price action.

The true advantage in using lower time frames, is that it can help you win the same trades, but with more precise entries and exits, which means greater overall profits.

Finally, I see the 4-hour time frame as a very good place to start. This is because, going back to what I said about not all time frames behaving equally, 4 hour candles strike a nice trade-off between being large enough to show something meaningful, but also being small enough to allow you to get into a piece of the move before it ends.

Finding Trends in a Quiet Market

It has been a quiet week in the markets so far and even as we come to the U.S. open as I write these words, the markets remain quiet. This might change later today, but as there isn’t much news right now I thought I would write about a few fruitful places traders might be able to find trending currency pairs to exploit.

Looking at the major currency pairs first – EUR/USD and USD/CHF are very flat and best ignored. GBP/USD is still down post-Brexit but there is no question that downwards trend is becoming flat and is quite probably nearly over.

The only reasonable candidate from the major currency pairs is USD/JPY. Here is the weekly chart with a standard 20-period, 2 standard deviations Bollinger Band indicator on top:


Clearly there is a downwards trend still ongoing. However, it has been many weeks since a new low was made, with lots of support at the psychologically key level of 100.00 and below. This means that shorting this pay may not be the happiest hunting ground if you are targeting a significant number of pips, and is an early signal of a possible trend change.

Next we turn to the Pacific, looking at the Australian and New Zealand Dollars. Here, the NZD/USD currency pair is certainly in an upwards trend, but this also looks a little weak:


If you look at the final eight or so weekly candles at the right of the chart, they all have long upper candles, and are at least a little suggestive of a bearish head and shoulders pattern (this would probably be more visible on a daily chart).

Looking at precious metals, Gold and Silver are still just about in the mix for bulls, but have not looked very healthy lately so there would not seem to be much opportunity there.

So, in our hunt for stronger trends – remember, the stronger the better! – we have to turn to some lesser-known currency pairs. These can have relatively large spreads / commissions, but a good trend should pay for that. You have to watch out though, because these smaller currencies can be more volatile and prone to sudden massive reversals, so stop loss orders become more important, even though they can be subject to slippage.

A good candidate is the Mexican Peso, traded against the U.S. Dollar:


This trend looks stronger, but definitely more volatile. You can see the strength by the recent new highs and the steep slope of the Bollinger Band.

The Mexican Peso has weakened enormously against the U.S. Dollar, falling by over half in value since September 2014. Much analysis attribute this to Trump’s significant prospect of victory in the upcoming Presidential election, as his policies are widely seen as detrimental to the Mexican economy. However, this trend has been running for much longer than Donald Trump.

Clinton vs. Trump – Round 1

I didn’t post yesterday as it was a very quiet Monday with not much going on in the market, and I knew anyway that after last night’s debate of the two major U.S. presidential candidates I probably would have something to say, so here it goes.

I didn’t stay up very late to watch it live so I’ve just seen the highlights. Notwithstanding the fact that the media is heavily biased in favor of Clinton out of fear of a Trump victory, I think it is true that most neutral observers would agree that she “won” the debate, so I agree with what nearly all the media is saying.

Both candidates managed to mostly put their best face forward, but Clinton seemed to strike the better balance. Trump was trying not to be too aggressive and he did moderate his vocal tone, but he didn’t seem to manage to coolly deliver a few telling blows in the way that Clinton did. Clinton also looked and sounded far healthier and composed that has in most of her recent public appearances. I thought she was much more convincing in this debate format than she was at the DNC rallying before a big crowd, for example.

There is intense interest in the “outcome” of this debate because the opinion polls have been narrowing for some time. Just prior to the debate, Nate Silver was showing if the election were held tomorrow, it could be swung just by the electoral votes of one small state (Colorado was the balance). A few national polls were starting to show Trump taking a small lead in the popular vote. With the odds seemingly narrowing, a clear “win” by one candidate in the debate could provide a decisive push or shift in momentum that could prove crucial in setting the tone of these final weeks.

A CNN poll of viewers gave Clinton a very healthy lead in debate performance, yet this audience is heavily skewed towards the Democrats by a net of approximately 15 point. Nevertheless, it is significant, and we have to wait and see whether Clinton benefits from this by beginning to pull head in the swing states.

What effect will the election result have on the markets? Broadly, a Clinton victory would be seen as “business as usual”, as continuity. A Trump victory would probably provide more uncertainty, fear and volatility as no candidate has won the Presidency without previously holding some kind of high national office for almost 100 years. Eisenhower won in 1952 but had been the commander of the Allied invasion of continental Europe a few years earlier, although he had never been elected to any position.

Trump and Clinton

Changes in Global Forex Markets 2013 to 2016

Last month The Bank for International Settlements released their latest Triennial Central Bank Survey which essentially tries to account for the major statistics in all global Forex trading.
Here are some of the key points, highlighting how things have changed over the years as compared to their previous surveys:

  1. Spot Forex transaction volume dropped by 19.16% between 2013 and 2016. In other words, the total quantity of spot Forex trades globally dropped by just under one fifth! Forwards and swaps, in contrast, are almost completely unchanged over the same period.
  2. The most widely traded currency pair is EUR/USD which accounts for 23% of all transactions, followed by USD/JPY (17.7%), GBP/USD (9.2%), and AUD/USD (5.2%). This should give you a very good idea of where the money is!
  3. USD currency pairs comprised 87.3% of all transactions.
  4. The most popular currency cross was EUR/GBP (2.0%), followed by EUR/JPY (1.6%).
  5. Since 2013, London’s volume of OTC FX trade turnover has fallen by 10.99%, from $2.73 trillion to $2.43 trillion. However, the United States’ volume over that same period has risen very slightly.

What does all this mean for retail Forex traders? Maybe not a great deal. In recent years though, I have noticed that the time around the London open seems to have become less important, with less volatility between the London open and the New York open. In contrast, the New York open seems to have become more crucial. This may well be reflected in the shift by volume from London towards New York that the survey shows.

The other take away for me is how important the USD is a currency within the Forex market, particularly as regards the EUR, JPY and GBP. Note how the USD/CHF pair, generally regarded as one of the “big four” does not even make the top 6 by volume, accounting for only 3.5% of market volume.
If you trade only the big three currency pairs, you are covering just under half of the entire market’s volume.

September FOMC Decision & the “New Normal”

Yesterday the Federal Reserve made its monthly announcements, which included the news that its interest rate would be left untouched, as well as some economic forecasts.

These announcements were keenly anticipated, as the question of American interest rates and monetary policy has been the center of debate in the Forex market for some time now.

It was not a surprise that the rate of interest was left untouched, with most analysts seeing only a 1 in 3 or 1 in 4 chance of a hike. What is more of a surprise, at least to me, is that there was a lot of talk pushing the idea that it was going to happen and surprise everyone. Well, the Federal Reserve almost never raises interest rates just before an American Presidential election (the vote is now less than 2 months away). It has happened only once before in recent decades. Of course, the Federal Reserve is meant to remove any political considerations from its decisions, but I think they interpret that as not inserting themselves into the political environment while an election campaign is going on. Interestingly, Donald Trump has made an allegation that the Fed are keeping interest rates artificially low for political reasons, so whatever the Fed did yesterday, it couldn’t win!

The really interesting take away from the Fed yesterday was its downgrading of both its timetable for rate hikes in 2017 (but are still indicating a December hike is more likely than not) and the economic performance of the U.S. economy generally (it cut anticipated 2016 growth from 2% to 1.8%). Of course, the downgrades were not large, and were accompanied by much rhetoric that more or less says that the economy is chugging along pretty well.

It might be that the Fed is achieving what it truly wants: to give the impression that the economy is doing well, that interest rates will be raised and were close to being raised yesterday, but not quite yet. As it happens, this is the best of all possible worlds: it promotes confidence without any of the real consequences that would probably arise from the implementation of a rate hike. At the time of writing this, the U.S. Dollar is down, but stocks are up and risk assets are up and sentiment is quite sunny.

Nevertheless, things are just not that great. From time to time we get economic data releases that look bad, like the confidence measure we saw a couple of weeks ago that came in well under par. If you are old enough to remember the 1980s, you know what a real strong economic and market boom looks like. What we have now, is nothing like one. In Forex though it is all relative as currencies are always traded against other currencies, and the U.S. economy, in spite of its problems, is doing better than more or less anywhere else.

The Chair of the Board of Governors of the Federal Reserve System.

The Chair of the Board of Governors of the Federal Reserve System.

FOMC Day Arrives; Japan Confusion

The big day is finally here: within a few hours, we will all know whether the Federal Reserve is raising interest rates this month of September. It has been the major topic of fundamental debate in the Forex market for quite some time now.

The market’s consensus is that there will be no rate hike today. Although there was an interesting story in Bloomberg saying that some of the Fed’s major dealers appeared to be positioning themselves for a surprise rate hike, with just a few hours to go opinion seems to be shifting even more in favour of no rate hike, with Bloomberg giving odds of about 2/1 for such an eventuality.

The major reason it is such an important question for the market is that America is seen as the strongest major economy in the world right now, and as such, traders are prepared to buy into the U.S. Dollar. For longer than one year the U.S. Dollar has been choppy against most currencies, so traders are waiting for a trend to emerge. The endless will-they-won’t-they debate over the rate hike and mixed economic data releases which vary from month to month keeps the USD choppy.

One thing that might help your trading if you like to trade based on economic releases, is to consider “which side is the bigger surprise”. The bigger surprise here would be a rate hike. This suggests that if there is a hike, the USD will rise by proportionately more than it will fall if there is no hike. Another thing you can consider is how new data releases such as this one tend to boost prevailing trends when they support the trend. That doesn’t really apply here as the USD is not in much of a trend right now – well, maybe a small bullish trend, which would also suggest that a hike would produce the bigger move.

Successful trading is more about spotting where the potentially big payoffs lie, than about being right. If you are right half the time but triple your money when you are right, well, you will be doing absolutely fine.

As for Japan, the Bank of Japan released their monthly report earlier and pulled a lot of complicated monetary manoeuvers that I won’t pretend to fully understand. The result was that the Yen weakened, and then later began to strengthen. The persistent strength of the Japanese Yen doesn’t make a lot of sense vis-a-vis the Japanese economy. The Bank of Japan is trying hard to cause inflation and seems prepared to keep injecting new money into the system endlessly, yet the Yen gets stronger.

When technical analysis and fundamental analysis are completely out of sync like this, I say go with what the charts are saying. If people are persistently buying Yen, it doesn’t necessarily matter why, unless there is some factor that can flip very quickly and cause a huge reversal in the blink of an eye.

Now it is over to you Ms. Yellen….

Time Series Momentum

Over the past few days I have been writing about strategies that buy the currencies that are going up the most strongly and sell the currencies that are going down the most strongly.

I showed how some simple strategies based on this principle would have performed over approximately the last 21 years.

For example, the 3-month version of this strategy yielded 94.45% over a 21.75-year period (meaning each trade made a profit of 0.36% on average). The maximum peak to trough draw-down was approximately 27%.

Here is another variation: instead of picking the strongest and weakest currencies to trade, how about trading all of the currencies and just buying each currency when it is going up and selling each currency when it is going down?

This kind of strategy is known as “time series momentum”, as opposed to “best of momentum”. It just means that the price has to be higher or lower than it was, not outperforming or under performing any other currency.

Applying a 3 month look back period here over the same overall time period, the strategy returned 357.24% (meaning each trade made 0.20%). The maximum peak to trough draw-down was approximately 131%. The equity curve is shown below:

The maximum peak to trough drawdown was approximately 131%.

The maximum peak to trough draw-down was approximately 131%.

As you can see, the time-series strategy had a larger proportionate draw-down (36%) than the best-of strategy (27%). It also made a higher average return per trade (0.36% compared to 0.20%). So the best-of momentum strategy is better, right? No, not necessarily.

The reason for this is that best-of strategies built from time periods other than 3 months, do not perform as well as time-series strategies built from other time periods. This suggests that time-series strategies are more robust i.e. more likely to keep working. This is probably because with best-of strategies you are putting all your eggs in one basket, whereas time-series strategies will keep you more diversified.

To give an example, the 6 month look back applied to a time series strategy over the same time period gave a very similar return to a 3 month look back. However, a 6 month look back to a “best of” strategy actually produced a loss overall of -56.02%.

Tearing Your Hair Out? Part 3

I thought I would continue from yesterday and try to conclude what I was writing about: an academic study testing “best of” momentum, where we look at a whole bunch of currency pairs and buy the one that has gone up the most over X months and sell the one that has gone down the most over X months. Hold for 1 month.

I made my own (more limited) test and showed how using a look back period of 3 months with the 7 major currencies against the USD produced positive results over the period from 1993 to 2010 and also from 2010 to 2014.

I tweaked the test to see what the results were for a look back period of 6 months. The authors of the study claimed this was profitable, but less so than 3 months.

My results here were poor. The strategy produced a total return over a 21-year period of -56.02%. That represents a compounded average annual return of -3.84%. However, while the period from 1993 to the start of 2010 produced a compounded average annual return of -5.36%, the “out of sample” period from 2010 to 2014 actually produced a (barely) positive result of 0.83%.

I’ll emphasize here that we are using only 7 currency pairs as against the more than 40 that the authors of the study used, and also that they found a 6 month period was poorer than the results for a 3 month period.

This leads me onto another subject: “time series” momentum – meaning not picking historical best trend performers but just following all trends – typically gives better results. We’ll explore that later.

Tearing Your Hair Out? Part 2

Yesterday I wrote about a basic strategy that has been proven to “work”: “best of” momentum, where we look at a whole bunch of currency pairs and buy the one that has gone up the most over X months and sell the one that has gone down the most over X months. Hold for 1 month.

I mentioned yesterday a back test done by some academics using more than 40 USD currency pairs and testing them with this method between 1976 to 2010.

I am not quite able to match this easily, but I can test the same strategy using the 7 major USD Forex currency pairs, from 1993 until September 2014. This means we also get to see how the strategy would have performed “out of sample”, i.e. after the period for which the test was conducted.

I chose to test firstly the 3 month look back period, as the study claimed this period produced the best average results.

So what were the results? Well, excluding any trading costs, the strategy produced a total return over a 21-year period of 94.45%. That represents a compounded average annual return of 3.22%. However, the period from 1993 to the start of 2010 produced a compounded average annual return of 3.63%, while the “out of sample” period produced a smaller equivalent of only 2.44%.

The evidence we have in front of us suggest it is a robust strategy. I am sure if I included more exotic currencies against the USD, the results would be better as their price movements tend to be larger. In my study here, it is just the 7 major currencies against the USD.

The equity curve showing the entire period from 1993 to 2014 within the back test is shown below:

The equity curve showing the entire period from 1993 to 2014

The equity curve showing the entire period from 1993 to 2014

Tearing Your Hair Out?

These days I am fortunate enough to struggle with the issue of whether I make as much profit as I “should”. When I was starting out as a trader, and this might be familiar to some of you, I struggled with trying to break even, let alone actually pocketing any profit. I tried a whole lot of stuff and would arrive at a “tearing my hair out” moment where I wanted to yell “Can somebody just tell me what the hell works???”

If this describes how you are feeling right now about your trading, then I can tell you something that does work, statistically speaking – following trends. That might sound dull or boring, but maybe the problem is really that everyone has a different definition of what a “trend” is. It can actually be extremely simple.

What really influenced me to go in this direction was finding some academic research papers which seemed to prove that just buying currencies that had been going up and selling currencies going down produced pretty good returns over a recent multi-year period. One of the best studies performed the following experiment:

1.    Each month, they looked at the exchange rates of more than 40 currencies against the U.S. Dollar. Historical data from 1976 to 2010 was examined.
2.    The currency that had risen the most against the USD over defined look-back periods (1 month, 3 months and 6 months) was bought and held for one month.
3.    The currency that had fallen the most against the USD over defined look-back periods (1 month, 3 months and 6 months) was sold and held for one month.
4.    The average return (not accounting for trading costs) was about 7.50% per year. This may not sound like a great deal, the issue here is really whether there is a mathematical strategy that has a positive expectancy over a large data sample, and the study seemed to show this is indeed the case.

This is actually a “best of” momentum strategy, where assets with the most positive or negative extreme performances are selected and followed in that same direction of performance.

I am not convinced that this strategy has performed as well since 2010 as is claimed for the 24-year period leading up to it, as the major trends of the past few years have been driven by fluctuations in the U.S. Dollar rather than other currencies, and the strategy outlined above is U.S. Dollar neutral. However, I will check this against historical data since 2010 and see how the strategy would have performed, and report detailed hypothetical results at a later date.

A final word – “time series” momentum strategies generally work even better than “best of” momentum strategies. In “time series”, you just buy X asset if it is up or sell it if it is down without reference to any other asset. More about this another time.