Trend Trading

Greetings and salutations to all fellow Forex traders. This thread is called Trend Trading for a simple reason. It is thus, either trends exist in Forex or they don't. If trends don't exist, then currency markets are simply a random walk, with no direction anymore likely than another. If currency moves are completely random, then it would be no more possible to make a consistent profit off trading Forex than making a consistent profit off the toss of a pair of dice or the spin of a roulette wheel. It's ok to believe that, but if you do, you'd may as well go to the casino. At least you get free drinks with your bets there. If however you believe that trends do exist in forex, then this thread will discuss how to find and profit from them. This argument has been going back and forth between finance and math guys for years, and we aren't going to solve it here. Here, we will just assume trends exist and try to reveal how to make a profit from them.

Just a few rules. Anyone of any level of experience is welcome to post here. I only ask that everyone remain respectful and try to keep a positive attitude. I find that the more positive my attitude, the more positive my trading results. I'm hoping experienced traders can discuss items of mutual interest here and along the way they can mentor newbies, not crush their dreams. Just one other thing, we are here to discuss trading. Discussions of other topics should best be left to other threads. That said, trolls will be immediately banned. Otherwise, any questions or comments regarding trading are welcome from anyone. Let's begin.

I'll begin by throwing out a few things for discussion and readers can chime in as they wish. We can begin by looking closely at the definition of a trend. What is a trend really? I'd say, a trend is when price has more moves in one direction than the other over a specific period of time. So, if price ticks up, down, up, down... over a period of time, there really is no trend. So we definitely can't trade it. But if price goes up, up up, down, up, up, up, down..., is that a trend? Can we trade it for a profit? Maybe, maybe not. I like to believe there are two types of trend. An apparent trend is like flipping a coin heads three times in a row. That tells you nothing about whether the next flip is more likely to be heads or tails. A real (or tradable trend) is driven by some underlying fundamental force, like hawkish comments from the RBA driving up the AUD. I made a couple thousand pips last week using a multi-lot strategy trading the Aussie without having a clue what was driving it. I really didn't care until I got curious later and went back to see. So, must we understand the fundamental reason driving a tradable trend to profit from it? No, not really, but knowledge is power, and we like power, right? More power! But let's not go power mad. As often as not the market will go against fundamentals, just to frustrate our attempt to trade fundamentals. Fundamentals can be used to confirm and understand trends, but we wont be using them to predict trends. Any thoughts on that from the readers?

On technical charts, trends are usually marked by a succession of higher or lower trading ranges. An uptrend is understood as a market that makes a series of higher highs and higher lows, and a downtrend is understood as a market that makes a series of lower highs and lower lows. Traders will find it helpful to pay attention to real-world factors that could be driving long-term trends for certain currencies and assets.

Continuing to look at trends. From my experience, about half the popular pairs appear to be trending at any one time. But about half of those are really just apparent trends. They are simply the random motion of price caused by business currency transactions and speculators all over the world that have no single underlying fundamental basis, or at least none that we can discern. Perhaps Caterpillar needs to repatriate a few billion Yen to the US for construction equipment sold to Japan, perhaps Apple needs to trade some Euros for Yuan to pay some Chinese children's parents for some iPhone fabrication. We see several ticks up as the transactions clear, but can we trade it? Not really, when the transactions clear, the next tick is just as likely to be down as scalpers take a few pips profit as up when momentum traders try to jump into a rising market. It's an apparent trend and it's whipsaws have driven more than one aspiring trader to other markets, or even worse, a day job. So, how can we tell the difference between an apparent trend and a real trend driven by underlying fundamental factors, one that is likely to persist long enough for us to extract profit from it? Any ideas? Comments?

Hello mate, thank you for starting and giving time to this thread.
As an accounting and finance guy, I really don't know if finance guys believe on trend existence or not in market but I do. Second as you said "No, not really, but knowledge is power, and we like power, right? I was watching The Mummy (2017) where Ahmanet says about power, Power is not given, it's taken. But she went mad for power. Although irrelevant but more interested to learn from you and would keep sharing my insights and learning here. Thank you once again.

Aaz

But there are trends visible in random coin toss generated charts :O

In my opinion to identify the trend correctly, we must simply understand which side of the scale is biased.
For example in the following 3rd chart (more in the libk below) you see bullish trends, created simply by the the probability of an up tick being 55% while the probability of a down tick was 45%.
http://www.forexfactory.com/showthre...83#post7985283

Very interesting thread. I would actually be more keen to discuss on the technical side of trend following; topics such as win ratio, money management, diversification and drawdown tolerance.

I would also be interested to discuss the the effectiveness of the very mechanical nature of trend following strategies. Would it be better for discretion in trend following as opposed to strict signallng system?

Lastly, I think it will be beneficial to all trend followers to discuss the various trend following signals. Cross-sectional momentum, absoulate returns and indicator based signals just to name a few. What are the good and the bad of these signals?

Superb

Thanks for the feedback and comments. Yes, I did sort of start in the middle of the story, assuming we had already identified a trend. There are many methods to spot trends. A series of higher highs, or lower lows, trend lines, moving averages, indicators, etc. All probably have some value and I promise to discuss spotting trends more as we progress. At the moment, I'm assuming we have spotted what we think is a trend, using our favorite method, or a combination of methods. I pointed out that the trend may be the result of a real fundamental force driving price up or down, like a central bank rate change, or it could be the result of random price action with a cause unknown to us. The question is then, can we tell tradable real trends from what appears to be random price action. Even more to my point, do we need to know the difference?
There are two answers to this question. First, yes, we can examine all the available news, and quite often the fundamental reason behind major moves will be splashed all over the headlines. For instance right now the GPB is trending down and the AUD is trending up. Obviously one of my best trades at the moment is short GA and it's easy to see the reason. Just type into google something like "news why is the AUD going up", or "news, why is the GPB going down" and we'll quickly see inflation fear in the GPB and a hawkish central bank in the AUD. I understand these are after the fact explanations for the move though. Still, it gives me confidence that what I'm seeing on the technical side is real and not just the result of random price action. I like being confident, it makes me happy. But I don't need to know the reason to successfully trade a trend. I just need a method to tell real fundamental driven trends that are likely to persist from random price action where the trend is likely to fail.
The second answer can be seen in the charts that Max posted above. A real fundamentally driven trend will persist over time due to the bias imparted on the price action by the fundamental driver. An apparent trend driven only by random price action will fail quickly as price reverts to a mean value. Essentially, price action tells us if the trend is real or apparent. If price continues it's trend up say, we can say that so far, it appears to be a real trend and profit can be extracted if it continues long enough. If price breaks trend and reverts to mean, we can assume that the price movement is the result of random price action and the trend isn't tradable for profit.
That leads us to a startling conclusion. Following my assumptions, 1/2 of pairs don't appear to be trending, so we ignore those. Of the 1/2 that appear to be trending, 1/2 of those are driven by fundamental forces, and most of those will allow us to take profit from them (that's 3/4ths of all pairs, for those that are keeping track). Of course we wont be 100% effective in extracting profit from every good trend, but we don't have to be, and with tons of practice, we can get better and better at it. Now, what's left is the 1/4 of pairs that appear to be trending, but are really the result of some random unpredictable forces. Generally, if those pairs are really random, about 1/2 will continue to tick in the direction of the apparent trend for some time (that's 1/8th of all pairs, if you are still keeping track), and about 1/2 will soon revert back against trend to near the mean (the final 1/8th).
So, our conclusion is 1/2 of pairs don't appear to be trending, and we ignore those. About 1/4 of pairs are really trending and we will take those trades and get rich off them, 1/8th appear to be trending and continue in the direction of the trend, we ride those for as long as we can and often profit off those when we can. About 1/8th appear to be trending but turn against us soon after we enter the trade and revert back to mean. We quickly cut those losses short and laugh in their betraying little faces. We have profited off the 1/4 that were really trending and many of the 1/8th that we were lucky enough to select even though it seems they were driven by random price action, that's 3/8ths of our trades profitable and only 1/8th unprofitable. If we cut our losses very short and let our winners run a reasonable distance (don't be too greedy, leave a little for the next guy), we can do very well with that 3 to 1 edge. Thoughts? Comments?

Excellent suggestions! Let's make sure to get to all of these. It's a big subject. But we have all the time in the world.

I think the key is to let the profit run, but you won't be profitable with random entries and a rr of 1:2 - 1:3... lol

Yes, It's important to cut losers fast. I usually cut them off very fast, like if they go against me 20 to 50 pips, very fast. Most, but not all, of those trades fall in the 1/8th of random motion apparent trends that prove themselves to be losers. Of course, a few might have been cut early, but that's ok. They can be easily re-entered. The important part is to cut losers early before they get a chance to grow out of control. It's like weeding the garden. I usually take profits off a portfolio of trades at the same time, but the average profit is much more than my average loss. To be clear, we are letting price action determine which trades to let run and which to cut off quickly. Questions, comments?

seems perfect to me.. I can't deal with losing streaks... so I do the opposite (cut the profits and let the losers run). believe it or not it still provides a small edge.. better than nothing

Hi Graviton
I followed you on your multi time frame thread.
That was fun.
Looking forward to participating in your new tread.
Thank you for all your help to so many out there.

example of trend following with bad rr... as you see those curves don't look good but diversification gives hope..





tested with 20 points of spread.

Hard for me to see the trends there, but I do believe in diversification. So, for instance, I start out monitoring 28 pairs. About half show no clear trend, so I ignore them. I'll take a position on the remaining 14. Of those 3 or 4 will quickly (within hours) go against me. I'll quickly bail out of those as they get to -20 to -40 pips. I rarely let them go to -50. So that will leave me with about 10 trades. I know a few of those are just random motion pretending to be trends, so I wait them out. The good ones rise to the top and I cut out the bad ones. In the end I'll wind up with 4 to 8 good trades and maybe a few just slightly negative, but well ahead overall. My best trade is now GCad short, at +112 pips. My worst is AU long at -13 pips. I'm currently trading 13 different pairs and I'm about +290 pips ahead for the day. I suspect I'll carry all these trades until Friday unless they go bad, in which case I'll bail out quick. I often bail out on Friday as I don't like the opening gap risk from the weekend, but often my best trades are when I risk the opening gap and stay in the trade over the weekend. Still, it's an unbounded risk that I'd prefer to not take.
My point is, my perception of something looking like a trend puts me in the trade. My understanding of the fundamentals gives me some confidence to stay in, but it's ultimately price action that determines which trades I stay with and which I cut short. I play no favorites here. If it wins I stay in it and for the best of the best even add to it. If it loses I cut it off at the root and say good riddance.

Great to see you back G and with a great new *always enlightening* thread.

Tough question......however let's assume just for an example that of all extended directional moves that get classified into our definition of trend, 70% of them are attributed to random price movements (fake trends) with the balance being attributed to fundamental drivers and deliberate rational behaviour of investors (real trends). If we trade all of them using *non-predictive* trend following rules irrespective of their fundamental basis....then we would profit from all of them......however one way to distinguish between *fake trends* and *real trends* is through their duration and extent of uni-directional momentum. This is quite hard to do visually (probably best left to statistical methods) unless you use methods (such as your multi-timeframe approach) that can evaluate their momentum and directional extent (refer to the chart below that compares a random price series (orange) against the same random price series with a minimal drift applied of 2% attributed to underlying factors (blue)).

Fake Trend versus Real Trend


Trading trends of longer duration and momentum extend your profitability given their higher Reward:Risk ratio. While it is difficult to distinguish between *fake trends* and *real trends*, it is the real trends that offer durability and this can only be determined as a hindsight measure. A way around this is to avoid any form of prediction and simply trade what you observe.

IMO that's why following price action is essential for trend following. As a trend follower, it is self-defeating to attempt to predict future price action. Rather you simply follow price action with standardised repeatable rules applied. This therefore only exposes you to times when uni-directional moves are not sufficient to profit from them (eg. sideways ranging with mean reversion). By diversifying you can then further restrict your trend following rules to only trade uni-directional moves of a more *significant* character that reduce the *random trend* to say 40% of the mix of trend trades with *real trends* contributing to 60% of the mix. As a result you progressively improve your performance metrics *generally* as you diversify.

What is your the percentage of your winners? As I personally view 20-50pips stop is VERY tight. Daily market volatility can very well hit these numbers.

Also, say for example we use a MA crossover strategy (It can be any signallng system but the concept applies). It signals us to go long. We do.
However price reverses intraday and hits our stop before bouncing back to close higher at the end of the day.
Should we enter long again or do we wait until our MA gives us another signal?

I ask because this is one of the common problems trend followers face. Personally, I will wait for another signal. At the risk of missing out a big one.

Good questions. I'm currently winning about 42% of my trades, so I'm taking lots of small losses. I currently have an emergency stop loss at entry of 150 pips. My goal is to never hit it. I try to manually exit all my trades, well before it hits the stop, usually when they are in a 20 to 50 pip loss range. As you said, a tight stop can get taken out with a quick spike. I do slowly move my stops up in the direction of the trade as time goes on. I hit one stop for a loss of 71 pips out of my last 50 trades. I had tightened up the stop over a period of 5 days and a spike took me out. That was all my fault, but that's ok. A good edge covers up many sins.

Yes, I will re-enter after I have manually exited or stopped out, if the trend resumes. That happens a lot, but I don't worry over it. I just trade what I see. Usually if I manually exit, or stop out, price has broken out of the trend I'm following, and it usually takes a day or several days for my trend criteria to be satisfied again. When it is all lined up again, I re-enter. I might lose 20 pips 2 or 3 times, but then I'll make it all back and some extra on a single 120 pip win.

Oh, FYI, in my trading world, there is no cost to missed opportunities. For instance, there are a million good trades a day somewhere in some market that I miss. Am I supposed to assign a cost to all those missed opportunities too? So, I don't care at all if I exit a trade for a 20 pip loss and then it turns around for a 200 pip win. The only point is that I exited when my system says I should and if the trade resets the entry criteria, then I re-enter it. Thanks for the good questions.