Second day of trading Forex - I love stop losses!

[cynic]

Trading Station has Market Range and Range Entry functions that limit your negative slippage (but not your positive slippage) on Market Orders and Entry (Stop and Limit) Orders respectively.

FXCM's MT4 platform has a Maximum Deviation function that works similarly to Market Range to limit your negative slippage on Market Orders. However, MT4 has no function like Range Entry to limit negative slippage on Entry Orders.

The stats for orders executed through FXCM over a 12-month period from September 2013 through August 2014 show our clients received positive slippage as frequently as negative slippage overall: http://bit.ly/1qLierb

• 76.2% of all orders had NO SLIPPAGE.
• 13.5% of all orders received positive slippage.
• 10.2% of all orders received negative slippage.
• Over 58% of all limit and limit entry orders received positive slippage.
• 52% of all stop and stop entry orders received negative slippage.

That's great for those people trading two to three years ago. It happens to be May 2016 at present. Is there any chance of something more recent than a report covering a period ending August 2014?

 

[NickF]

Jason,
I thought I will ask you another question, as you may have access to the statistics data from FXCM.
I understand that the accounts at FXCM were profitable in 2010 as follows:

Equity Range % Profitable
$0 – $999 27.89%
$1,000 – $4,999 40.52%
$5,000 – $9,999 42.36%
$10,000+ 47.74%

Did this profitability change in the last years? As the machines get smarter, maybe people get dumber?

Is it possible to provide the average annual profit of the top 10% most profitable accounts in each category? I am curious if the profits of the most profitable traders are like 2-3%, or they are like 20-30% per annum

Nick

 

[Jason Rogers]

About half an hour later, I received an email from DailyFX Trading Signal Alert, about sell trade opportunity for EURJPY, USDJPY, AUSJPY.
Unfortunately I can't see from DailyFX's email what were the prices where they sell - would have been interesting to see if they picked up better entry points than my system. It would also be nice that when a trade is closed (at a profit or at a loss) to see the profit/loss made, in pips.

Hi Nick,

You can see the entry and exit price for DailyFX Trading Signals on Mirror Trader. For example, below is the Mirror Trader strategy card for the Range1 signal from DailyFX PLUS.

​

You can see the strategy opened a short position in AUD/JPY at 78.712 on May 23rd.

Jason, you were right about possibility of S&P500 to target a new high considering the SSI was strongly expecting a drop.

SSI isn't always right, but I find it can be a great filter to help me decide when to act on my trade ideas. The fact that so many retail traders have already shorted the US stock indices in anticipation of a drop told me the market might not be ready to oblige.

​

I want there to be fundamental, technical and sentiment-based (SSI) reasons for entering a trade before I pull the trigger. It's not necessary for me that retail traders flip from being net short to being net long the S&P 500, but I would like to see some reduction in the current extreme level of shorts.

 

[Jason Rogers]

That's great for those people trading two to three years ago. It happens to be May 2016 at present. Is there any chance of something more recent than a report covering a period ending August 2014?

Hi Cynic,

Thank you for your interest. We are in the process of compiling the latest stats for an updated execution study. I can let you know when the data are released.

In the meantime, it's worth noting FXCM is one of the only forex brokers to provide our clients with tools like Market Range and Range Entry to limit their negative slippage, and to publish execution stats proving our clients receive positive slippage as often as negative slippage.

 

[Jason Rogers]

Jason,
I thought I will ask you another question, as you may have access to the statistics data from FXCM.
I understand that the accounts at FXCM were profitable in 2010 as follows:

Equity Range % Profitable
$0 – $999 27.89%
$1,000 – $4,999 40.52%
$5,000 – $9,999 42.36%
$10,000+ 47.74%

Hi Nick,

Below are the data from a DailyFX report in 2011.

Did this profitability change in the last years?

Below are the data from a similar DailyFX report in 2015.



While the percentages changed a bit over the years, the main takeaway point remains the same: It's always a good idea to use leverage judiciously, and stats show 10:1 leverage or less tends to work better.

As the machines get smarter, maybe people get dumber?

You're assuming these stats only include non-automated trading which is not the case. Automated trading strategies are becoming more sophisticated, but since they are trading against each other in the market, this would not result in an increase in average profitability percentages. Remember that 50% is the theoretical upper limit on the profitability of the average trader, since for every winning trader, there is a losing trader on the other side. The theoretical limit is actually slightly below 50% when you take into account transaction costs.

Is it possible to provide the average annual profit of the top 10% most profitable accounts in each category? I am curious if the profits of the most profitable traders are like 2-3%, or they are like 20-30% per annum

Nick

While I don't have access to such stats over the course of a full year, you can see the performance of the top 10 traders each month in FXCM's $10,000 Monthly Challenge.

 

[Jason Rogers]

You can see the strategy opened a short position in AUD/JPY at 78.712 on May 23rd.

CORRECTION: It was a long position as shown in the image.

 

[NickF]

Hi Nick,
You're assuming these stats only include non-automated trading which is not the case. Automated trading strategies are becoming more sophisticated, but since they are trading against each other in the market, this would not result in an increase in average profitability percentages. Remember that 50% is the theoretical upper limit on the profitability of the average trader, since for every winning trader, there is a losing trader on the other side. The theoretical limit is actually slightly below 50% when you take into account transaction costs.

I was joking about that. Of course the stats include the automated trading. I said what I said based on the common knowledge that generally rich people get richer. So, whoever is already successful in Forex (especially in automated trading), they would put more effort to improve their already good performance (and they may often succeed), at the expense of the others.

The risk is that although nowadays is very easy for most people to open a trading account, the poor performance discourages people to continue trading. I understand a lot of Forex brokers lost customers in the last years. This is a bit of a conflict considering that more and more people now have access to internet and could open Forex accounts.

While I don't have access to such stats over the course of a full year, you can see the performance of the top 10 traders each month in FXCM's $10,000 Monthly Challenge.

Unfortunately that performance doesn't impress me. It may be good to attract naive people to Forex It is to me much like a monthly record of top ten Lotto winners all over the world. If such a list was made, it would show that number 10 on the list earned perhaps a few million dollars dollars. Would I start playing lotto? No, I still prefer to lose my money in Forex

The major problem with the results of the contest is that the account size can be as low as $50. A lot of people would happily create a few accounts of $50 and gamble like crazy with the hope they will win the big prize. As with Lotto, very few will succeed.

It would be more interesting to see a top ranking list for the people with accounts over $10000. While perhaps some of them may show monthly profits of 100%, I don't think there would be too many people with large accounts targeting a 100% monthly profit.

Nick

 

[Jason Rogers]

It would be more interesting to see a top ranking list for the people with accounts over $10000. While perhaps some of them may show monthly profits of 100%, I don't think there would be too many people with large accounts targeting a 100% monthly profit.

Nick

You make good points. While I don't have access to such stats at this time, perhaps they can be included in a future study by the DailyFX research team as a follow up to their series on the Traits of Successful Traders.

 

[NickF]

You make good points. While I don't have access to such stats at this time, perhaps they can be included in a future study by the DailyFX research team as a follow up to their series on the Traits of Successful Traders.

Thanks Jason for all links provided. There are many interesting and useful articles written by DailyFX research team. Such as the best hours for trading for retail traders (we should avoid high volatility), the best currency pairs, etc. I will try to include some of these in my expert adviser programs, but later.

So far I am running optimizations and out of sample testing and for some of the past years (such as 2010), I get hard to believe good results. Such as, optimizing EURUSD for three months (2010.01.02 - 2010.03.01), then testing on the next one to three months. Out of the thousands of optimized parameter sets, I selected 5 points and run out of sample testing. The worst result I got for the first month was +18% and the worst result for 3 months was +24%. I am playing with $10000 and 1 lot, which is a risk too great, so in real life I would scale down, perhaps to 0.01 lots or even 0.1 if I get confident in the code.

It takes me a long time to run even three to twelve months of optimization with 4 parameters. Usually over 24 hours. I am running on three computers. I am somewhat messy in recording the results (but I'm improving). For EURUSD, optimizing on 2015.01.02 - 2016.01.01 and testing on 2016.01.02 - 2016.05.28, I get a 52% profit, with a max drawdown of $2887. I don't suspect any fishy mistakes in my code, such as including future data (which should not be available when placing a trade or closing the trade) in my code. Anyway, I will run the expert advisers on my demo accounts and see if I get positive results - so far they look too good to be true.

Warren Buffet, here I come!

 

[Jason Rogers]

Anyway, I will run the expert advisers on my demo accounts and see if I get positive results - so far they look too good to be true.

I'm a big fan of automated strategies myself. As you know, past performance is not necessarily indicative of future results. That said, do you know if the results you have seen are for the account balance or the account equity? I try to watch out for automated strategies that only close the profitable trades to boost the account balance, all the while keeping open large losing trades that deplete the equity.

Equity and equity draw downs are what concern me more than the balance.

Also, keep in mind that 3 to 12 months is not a lot of time. I know it's more time consuming but try to get data going back as many years as possible. Market conditions change and the most robust strategies have proven themselves over the long run.

Another factor to consider is the rationale a particular strategy uses. Does it make sense to you when the trader/strategy creator explains it? Do you have reason to believe past success was based on more than luck? Do you believe the strategy remains relevant in current market conditions with a good chance to succeed going forward?

 

[NickF]

I'm a big fan of automated strategies myself. As you know, past performance is not necessarily indicative of future results. That said, do you know if the results you have seen are for the account balance or the account equity? I try to watch out for automated strategies that only close the profitable trades to boost the account balance, all the while keeping open large losing trades that deplete the equity.

Equity and equity draw downs are what concern me more than the balance.

Also, keep in mind that 3 to 12 months is not a lot of time. I know it's more time consuming but try to get data going back as many years as possible. Market conditions change and the most robust strategies have proven themselves over the long run.

Another factor to consider is the rationale a particular strategy uses. Does it make sense to you when the trader/strategy creator explains it? Do you have reason to believe past success was based on more than luck? Do you believe the strategy remains relevant in current market conditions with a good chance to succeed going forward?

Hi Jason,

At the moment I make and test my own strategies. The holding period is short, usually under one day. The account balance and account equity are close, because I only allow one transaction on one currency pair at a time.

The implementation at the moment is very rudimentary, I have a long way in front of me to bring them to a level that I consider acceptable. Your concerns about strategies keeping the losers in a hope they will eventually recover are valid, the end result could be very unpredictable.

At the moment I set Stop Loss and Take Profit, although I consider replacing them with a Trailing Stop Loss. I'm already running a strategy for a few days on a few currency pairs and I seem to get positive results - the problem is that I am doing the development of other strategies on the same computer and the balance is tampered by my testing. But I already moved the strategy on a separate computer and I'll see how it performs.

You are right that 3 to 12 months is not a lot of time. But I have this theory that a strategy may not work for too long without optimization. For this reason I will change the parameters every few weeks (and they are a bit different for different currency pairs), and I expect a profit at the end of every month. That's wishful thinking

Once I complete the first strategy, I will use it on a real account, trading in lots of 0.01 and hopefully the loss will be less than what I am able to accomplish manually. The month of May was not my best, I ended up on -5.5% - I am still holding two positions in SP500). I was able to hold them during the best week for SP500 (which was the worst week for me). I am

Your questions are to be considered and time will say if I will be able to end up with some profitable strategy for the long term.

Nick

 

[Jason Rogers]

Hi Jason,

At the moment I make and test my own strategies. The holding period is short, usually under one day. The account balance and account equity are close, because I only allow one transaction on one currency pair at a time.

The implementation at the moment is very rudimentary, I have a long way in front of me to bring them to a level that I consider acceptable. Your concerns about strategies keeping the losers in a hope they will eventually recover are valid, the end result could be very unpredictable.

At the moment I set Stop Loss and Take Profit, although I consider replacing them with a Trailing Stop Loss. I'm already running a strategy for a few days on a few currency pairs and I seem to get positive results - the problem is that I am doing the development of other strategies on the same computer and the balance is tampered by my testing. But I already moved the strategy on a separate computer and I'll see how it performs.

You are right that 3 to 12 months is not a lot of time. But I have this theory that a strategy may not work for too long without optimization. For this reason I will change the parameters every few weeks (and they are a bit different for different currency pairs), and I expect a profit at the end of every month. That's wishful thinking

Once I complete the first strategy, I will use it on a real account, trading in lots of 0.01 and hopefully the loss will be less than what I am able to accomplish manually. The month of May was not my best, I ended up on -5.5% - I am still holding two positions in SP500). I was able to hold them during the best week for SP500 (which was the worst week for me). I am

Your questions are to be considered and time will say if I will be able to end up with some profitable strategy for the long term.

Nick

It sounds like you are being very methodical with your approach which is good.

 

[NickF]

It sounds like you are being very methodical with your approach which is good.

In the last few days I left two strategies I am working on to run by themselves, on two computers.
On one computer, I had the strategy running on a few currency pairs including all major currencies. On Friday (Jun 3, 2016), when USA published the unemployment rate, the strategy started buying and selling and I was curiously watching the action. After it made a virtual profit of about $100 (with 0.1 lots), I realized I could trade in the same direction on my real account. I didn't know how far was will go, so I only traded 0.01 lots. In the end, I made about $10 or so using a trailing SL. Next day, I had on my $5000 virtual account a profit of $483 (almost 10% in one day, trading 0.1 lots). I discovered a bug in my strategy, which traded on the same bar after it closed the trade, so I am fixing the code. I left the transactions that were still open until Monday (wanted to see what the strategy can do), when the profit dropped to about $180 or so. Today there was some activity (probably the RBA deciding the rate) and now I have a few open trades. The demo account is at $5270.

My second strategy tries to catch quicker the significant moves in the market and needs many improvements, mainly a trailing stop-loss (I am struggling to write code for it). However, today it managed to bring my other demo account from $4935 to $5015 (just closed manually, because I need to continue coding).

So, both my systems performed better than my manual trading (in my real account I still hang on two positions on SPX500, which stubbornly continue to rise - from my total deposited funds of $3000, I am now down to the lowest ever of $1666). I bought the two positions on 19/05/2016 at 2027 (could not get too much worse than that!) and the other one at 2036 - I was desperate not to lose the big down-trend and I was very convinced there was only one way to go from there and that was down Well. I'll just sell them at 1950, if I can survive that long

 

[NickF]

Just an update for the people who wonder if I'm still alive. This month so far I managed to recover some of the big losses from the previous three months. I am up about 9.5% for the month (of course, who knows what will happen till the end of the month).
I'm also working intensely on my Metatrader 4 scripts. I am pretty pleased with the last results of the simulation and I fantasize about the millionaire life. Which, according to the tests results, is just around the corner! After fixing quite a few bugs in my latest script and implementing the trailing stop loss strategy, now I have to figure out what's causing these too_good_to_be_true results. No doubt, I will be a bit sad to fix a script that works so well! After all, how many people get their scripts to generate $240000 of pure profit from $10000 in a year? Or maybe I should sell the script before I fix it

Nick, trying to keep himself sane...

 

[cogs]

Hi Nickf,

Do live trades using the script match the simulation trades?

I have coded plenty of EA's that work well on simulation but only three so far match live trades placed.

Any EA using a MA is a good example.

 

[NickF]

Hi Nickf,

Do live trades using the script match the simulation trades?

I have coded plenty of EA's that work well on simulation but only three so far match live trades placed.

Any EA using a MA is a good example.

Hi Cogs,

I just completed this script after Forex market has closed, so I'm waiting for tomorrow to check what's going on. Even so, it will only be with a demo account.

I made another script that I run for a while and that one is on positive balance (the balance is now $5542 from $5000, still on a demo account). For a few days I had the program closed on that computer and I didn't notice. It doesn't do many trades and it is triggered by major price and volume changes. It is quite primitive. I started the script on 03.06.2016. I am trading with 0.1 lots.

The version I am working now on is more dynamic and more customable. The odd thing is that the maximum profit in simulation comes for the highest number of trades and that can't be right. Plus, the success rate is ridiculously high and that again is a question mark. Plus, once I optimize over a period, I go out of sample and I still get excellent results - too good to be true. Anyway, if the demo account will show positive results, I could run it on a real account, trading 0.01 lots. Can't do a lot worse than me, manually trading

So, are you running any of the three scripts on a real account? If yes, do you prefer to manually trade, or let the system trade?

Nick

 

[cogs]

So, are you running any of the three scripts on a real account? If yes, do you prefer to manually trade, or let the system trade?

Yes two of them, but on and off when finances permit. I, like many, lack capital due to external reasons, I also think this is what generates creativity and good ideas.

I enjoy manual trading, and have for many years, but find these days manually trading against big dollar algos is mostly a lost cause, so my EA out performs me in a win/loss ratio. What used to apply manually is now long gone. Retail traders have such limited information to make decisions with, leaving us with longer time frames, patience and large capital.

 

[NickF]

Yes two of them, but on and off when finances permit. I, like many, lack capital due to external reasons, I also think this is what generates creativity and good ideas.

I enjoy manual trading, and have for many years, but find these days manually trading against big dollar algos is mostly a lost cause, so my EA out performs me in a win/loss ratio. What used to apply manually is now long gone. Retail traders have such limited information to make decisions with, leaving us with longer time frames, patience and large capital.

Hi Cogs,

Thanks for sharing your experience. In my case, I didn't trade long, so I have no idea if I would have been more successful before the proliferation of the algos. Probably not, at least in the beginning. In my case, I can leave my funds ($1737 at the moment) in Forex, I don't need to take them out. But despite they are not big, I still fight quite strongly to bring the balance back to initial $3000.

Today I lost in my manual trading over $130. I am a bit pissed off, not for the actual loss, but for my ignorance and inaction. The biggest issue was that I forgot the Forex market opens on Monday around 7:55. I had 3 S&P500 positions and some others left over the weekend, in a belief they will keep going down. So, I checked the markets around 7:15, nothing bad was happening, did my morning schedule and when I checked again around 8:30 it was already pretty bad. Had to leave for work and left my transactions open, thinking it will go down again later. Over the weekend I was checking the news and found out that the bookies are betting quite strongly in favour of a UK stay in EU (around 70:30), while some other news were saying the opinion polls show the odds for a Brexit are increasing. Quite contradictory information, so I didn't know which one to trust. Apparently people trust more the betting odds than opinion polls, although it may be just a myth they are more accurate.

Anyway, I tend to believe now that UK will remain in EU. What I think is that over the weekend a lot of people came to be of a similar opinion and as a result the markets are going up quite strongly. So, I had the knowledge (as little as it was), but I didn't evaluate if there is any significant risk in losing money on my trades - and I paid the price. At lunch (around 1:45 pm) I came across an article from AFR saying that after that PM was killed in UK, the last polls show a strong increase in favour of stay. If I had read that article earlier in the day, I would have not waited so long to close my positions...

I will probably move the real money into my FXCM account where I can run MT4 scripts. And probably I will slow down with the manual trading. I should give a chance to my scripts to prove me they are worthwhile of my time and effort

My scripts which I left running at home didn't send me any emails, so most likely they are not doing any trades. I'll check tonight what's going on.

Nick

 

[cogs]

I discovered some time ago news events are simply a catalyst to move the market for orders to be filled. I think this is a common mistake many make in trying to work out the direction of a market based on good or bad news, when markets actually move to the point of least resistance, or 'order flow' as some like to put it, and using the pump and dump method.

Market open today was a good example squeeze the price in one direction, get everyone suckered in, then blast it back the opposite direction. Always market movements are justified in media and such, after the fact, you see that this morning with Bloomberg and such talking about safe havens, but who would have predicted the opening gap direction this morning? Crazy at the mo.

I spent so much time on news proofing my EA, as such I am quite confident it would survive another SNB event. I am going to run all my EA's all this week on demo to just see how they handle the Brexit noise, and switch the live ones off Wednesday which are currently only trading Aussie pairs.

I opened a trade left it open once, then drove to work, twenty minutes later I was $2740 up. Closed it out and yep you guessed it, traded it all away the following week. Same old sucker move.

If these opening gaps don't get filled in the short term, we will have some gift horses coming up once things settle then price turns around to fill them.

What language are you using for your scripts Nickf?

 

[NickF]

I discovered some time ago news events are simply a catalyst to move the market for orders to be filled. I think this is a common mistake many make in trying to work out the direction of a market based on good or bad news, when markets actually move to the point of least resistance, or 'order flow' as some like to put it, and using the pump and dump method.

Market open today was a good example squeeze the price in one direction, get everyone suckered in, then blast it back the opposite direction. Always market movements are justified in media and such, after the fact, you see that this morning with Bloomberg and such talking about safe havens, but who would have predicted the opening gap direction this morning? Crazy at the mo.

I spent so much time on news proofing my EA, as such I am quite confident it would survive another SNB event. I am going to run all my EA's all this week on demo to just see how they handle the Brexit noise, and switch the live ones off Wednesday which are currently only trading Aussie pairs.

I opened a trade left it open once, then drove to work, twenty minutes later I was $2740 up. Closed it out and yep you guessed it, traded it all away the following week. Same old sucker move.

If these opening gaps don't get filled in the short term, we will have some gift horses coming up once things settle then price turns around to fill them.

What language are you using for your scripts Nickf?

I am also laughing at how at the end of the day the news explain why the market went up or down but don't have much to say about what will happen tomorrow. If today was a bad day, the news are grim. If it was a good day, they are full of hope for the future.

I've seen somewhere some indicator which places horizontla bars on the vertical axis and it shows the price points where most of the trading takes place. There could be a few price points that act like magnets. I could not find such an indicator available for free for Forex.

At the moment I use MQL from Metatrader. It is close to C/C++/C# and I know a bit of all of them, so I find simple to make the scripts the way I want. There are nice functions available and I can develop quite fast. Nothing too fancy, not reading external websites, files or other things. Most of the time I get caught in bad logic or get tricked by my own tricks. The step by step debugging is not available (or at least I'm not aware of it) and this makes it harder for me, since I have to enter additional lines to print variable values, then delete them afterwards.

I also compiled two sample C# projects from FXCM, they are written by professionals and the code is of excellent quality (at least it looks like that for me, a beginner). In the future I may change one of the projects to do what I want, but it's a bigger learning curve for me. I already had a go at it and found it was not so easy as I imagined

Nick