How do you define a profitable trading system? | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

If you have a profitable system, will brokers sabotage you? 61 replies

How do you define risk? 32 replies

How do you Define Identify and Follow Trend? 5 replies

How do you define a trend? 20 replies

How do you define a strong trend? 8 replies

How do you define a profitable trading system?

  • #1
  • First Post: Jul 7, 2018 7:04pm Jul 7, 2018 7:04pm
I have always wanted to be able to define what a "profitable system" is. I have spent good amount of time to think about it since there are infinite ways to make money in Forex and of course lose money as well.

I came up with the following list of criteria to be present in a system for me to believe it is a profitable system and what would interest me to invest real money.

1) The system should be back testable.
2) The system should produce similar results when backtested with multiple broker data.
3) The system should be tested on at least last 3 years of historic data.
4) Depending on the Lotsize used the per year max drawdown and Return on investment should be not less than 1:4 ratio.

I am very much interested in members sharing what you your definition of profitable system is and what all would you want to know to believe enough to be comfortably invest good chunk of your own hard earned money in it?
  • #2
  • Jul 8, 2018 5:52am Jul 8, 2018 5:52am
Sounds good. Have you such a system already in mind?
  • #3
  • Jul 8, 2018 11:03am Jul 8, 2018 11:03am
Greetings,

I agree with the above...sounds good. But can you make it work?

Another trader might think the opposite i.e. no backtesting, only forward testing, risk/profit ratio 2:1. There's no right or wrong...its simply a matter of observation. One trader might have a system where he targets 5+ x the risk & another system where it's only 0.5+ x the risk & ON THE SAME INSTRUMENT! surely that goes against all norms? but if it works & you can make sense of it all then convention doesn't matter. Therefore, if it makes sense & it's working then fabulous.

Finally, if you start trading then you should only risk small, enough to lose fairly comfortably. This way you can 'forward' test your approach & generally progress if you're serious. Eventually confidence & discipline grows which is key. However this takes a lot of time & effort...years in most cases I imagine. Backtests are useful for developing but it's live trading that really matters...learn from all mistakes, create rules if possible.

Good luck
who dares wins
  • #4
  • Jul 8, 2018 1:46pm Jul 8, 2018 1:46pm
@dkrock, i was impressed about your explanation..especially on number 12. Only trade on profitable movement..I'm still hard to catch it.
  • #5
  • Jul 10, 2018 11:37am Jul 10, 2018 11:37am
Hi Yalgaar,

i agree that a system should be back tastable for me to invest money in it. The point of the multiple broker data is a very interesting thought, especially the question, if the choice of broker could be some sort of optimization. Though i have no experience with that, its an interesting thought.

The following points i would slightly disagree:

3) The system should be tested on at least last 3 years of historic data.

The amount of data stated as years is in my experience irrelevant to a degree. What i want is enough data, which means trades in my backtest. if you take the 3 years, you will have 36 bars on a monthly timeframe, but roughly 22000 bars on an hourly chart. So in my opinion it is better to set a minimum of trades than years. As for how many trades should be available, i will return to that later.

4) Depending on the Lotsize used the per year max drawdown and Return on investment should be not less than 1:4 ratio.

Lotsize is for me of no consideration. The rules of entry and exit are the parts that make the system, money management is just a way to handle it. Return on Investment (or capital) is a measure which is imo not important regarding one system. It is a tool for asset allocation.
The drawdown is important, as it got to fit your personality and risk tolerance.

The points i would add are the following:

- i want a system to have as little rules and indicators as possible. Usually i will not even look at a system with more than 4 rules for each side. (Close above price and stochastic crossing are two rules. stochastic crossing under 20 would be a third).
- i want at least 20 trades per rule. so with 3 rules per side would be 300 trades. That is the absolute minimum, the more the better.
- the backtest should be a forward test with at least 3 periods
- results of the backtest should still look good if you remove one standard deviation
- results should stay constant even if input parameters are changed slightly

with that given and an adequate drawdown i would consider trade a system.

I hope that was at least a bit interesting.
  • #6
  • Jul 10, 2018 11:30pm Jul 10, 2018 11:30pm
What works now/today, may not work in the future.

Market behaviours change over time.

It is naïve to think that a trading system could be created/found to generate lasting profitability forever.
'How many times have any typical trader/you change/modify his/her/your trading systems, in the pursuit of 'beating' the market?' Ask yourself that.

Focusing/relying on rigid methodology and expecting to generate profits in the long run is what cause traders to lose money.
'For the market to work, it needs people who think that they can beat it.'
  • #7
  • Jul 11, 2018 2:42am Jul 11, 2018 2:42am
Quoting AlastorFate
What works now/today, may not work in the future. Market behaviours change over time. It is naïve to think that a trading system could be created/found to generate lasting profitability forever. 'How many times have any typical trader/you change/modify his/her/your trading systems, in the pursuit of 'beating' the market?' Ask yourself that. Focusing/relying on rigid methodology and expecting to generate profits in the long run is what cause traders to lose money.

Can you be specific about what market behaviours change, and why, and how you recognise these changes?
  • #8
  • Jul 11, 2018 2:51am Jul 11, 2018 2:51am
One that makes money?

If you are looking to find a trading system and backtest if and then trade it profitably you'll still be asking the same questions in 2 years. Indicators repaint, markets change - backtesting is nearly useless in most cases.
  • #9
  • Jul 11, 2018 3:22am Jul 11, 2018 3:22am
Quoting Catch-22
One that makes money? If you are looking to find a trading system and backtest if and then trade it profitably you'll still be asking the same questions in 2 years. Indicators repaint, markets change - backtesting is nearly useless in most cases.

What do you see that tells you there has been a change in the market? What would have caused this?
  • #10
  • Jul 11, 2018 5:57am Jul 11, 2018 5:57am
Quoting tomorton
{quote} What do you see that tells you there has been a change in the market? What would have caused this?

Volatility, length of trends, amplitude of corrections, length of corrections are some of the usual things that change. Plus the general tendency of the market might change, a trending market can become choppy and vice versa. The cause of these changes can only be speculated in most cases, but usually they are not that important. The FX market is usually very macro economical driven, which gives a good indication.

The problem with these changes are that if you take an increase in volatility for example, what has been a reasonable stop loss before becomes now to narrow. If the volatility decreases your stops might be to far away and the profits to small to cover the losses. This obviously changes the results of the system.

Nevertheless, i would disagree that backtesting is useless. You "just" need to know what you are doing and check from time to time if your analysis is still valid.
Problems occur always when you dont know what you do and i will give you an example:

I never trade on central bank reporting days, to choppy, to wild, not my turf. The ECB always reports on Thursdays, so with an analysis of which days are performing best and worst one might come to the conclusion that on Thursdays it is better to stay out of the market.
If you dont know why, and the ECB changes their reporting day to Monday, you will be in the market during choppy days and leave 52 valid tradings out per year, which could cause your system to lose money. If you know that you dont trade ECB days, you can adapt more easily.
  • #11
  • Jul 11, 2018 7:30am Jul 11, 2018 7:30am
Quoting AlastorFate
What works now/today, may not work in the future. Market behaviours change over time. It is naïve to think that a trading system could be created/found to generate lasting profitability forever. 'How many times have any typical trader/you change/modify his/her/your trading systems, in the pursuit of 'beating' the market?' Ask yourself that. Focusing/relying on rigid methodology and expecting to generate profits in the long run is what cause traders to lose money.
A very erroneous post if I may say so, and for the benefit of those new or relatively inexperienced, I would like to challenge it. This is of course my opinion.

I have traded now for ten years, and I have never known market behaviour change in any fundamental way beyond the normal day to day swing and rhythm of the market. The system I created for my own trading worked in the past once it became part of who I am as a trader, it continues to work, and I do not expect it to suddenly stop working as long as I continue trading.

One of the reasons why this is so is because the market is driven by human emotion, namely greed and fear, and this is a constant, it never changes no matter who the market participants are.

The other reason, and just as fundamental, is that systems do not exist in a vacuum, to be picked up by anyone and used to trade the markets. My system isn't really a system at all, it is an amalgamation and fusion of the knowledge and experience that I have gained over the years. Therefore, whatever changes in market behaviour there are, such as increased volatility etc, I will have seen many times before and will be able to adapt my trading accordingly.

This notion, which I have seen a lot of traders here on ff express, that what works now may not work in the future, is complete nonsense. Learn how the markets move, determine the most probable S&R levels on any given day, and learn how to identify optimum entry and exit levels, and this knowledge you will be able to use, trading any market in the world for as long as you trade.
  • #12
  • Jul 11, 2018 1:17pm Jul 11, 2018 1:17pm
Quoting yalgaar
I have always wanted to be able to define what a "profitable system" is. I have spent good amount of time to think about it since there are infinite ways to make money in Forex and of course lose money as well. I came up with the following list of criteria to be present in a system for me to believe it is a profitable system and what would interest me to invest real money. 1) The system should be back testable. 2) The system should produce similar results when backtested with multiple broker data. 3) The system should be tested on at least last...
Your criteria might not be enough, but it is a good start. One thing you might want to think about: How history turned out is just one scenario out of infinite scenarios. It might not be enough to just backtest on how history played out to be able to create a profitable system for the future.
  • #13
  • Jul 11, 2018 2:07pm Jul 11, 2018 2:07pm
'adapt my trading'

Nicely put.

---

Maybe I should just say most traders who think they have profitable systems, do not have profitable trading systems at all. Proven by losing accounts.

---

What I meant by what works today/this month might not work another day/month, means a trading a setup that makes one money today/this month doesn't mean trading the same setup another day/month will reap the same positive result. In worse cases, there could be more losing days/months than winning days/months. Good R:R becomes crucial when there are more losing days/months than winning days/months. And as we know most traders struggle to get good R:R or do not aim for good R:R at all.

A clearer example, is a range bound trading strategy will usually lose money in a strong trending market. (vice-versa)

---

Correct, market movement is for the most part driven by fear and greed, buying and selling.

However there is one issue.
Fear opposes greed (vice versa)
Buying opposes selling (vice versa)

When 'greed'/buying pushes price to the strong support/resistance level. So many different scenarios could happen.
1. Price reacts(small pullback) a little and then cuts through the level.
2. Or price reacts strongly (big pullback or major reversal)
3. Or price could slice through it significantly.

Do we trade along with the market momentum or do we take a shot at shorting at a resistance level? These two are completely opposing trading decisions, and could very well be right or wrong.

---

Don't get me wrong. I do agree that having a trading system is important. However the real test is whether it can stand the test of time, the 'worst' of market movements.

Personally I have not gotten any trading system to work (as in profitably) for more than 6 months. The best I have gotten so far is 3 or 4 months of profitable success with trading system, and then my system start 'failing' or losing me money.
'For the market to work, it needs people who think that they can beat it.'
  • #14
  • Edited 2:50pm Jul 11, 2018 2:37pm | Edited 2:50pm
The answer to Your question seems obvious. Since you started using this strategy, you have made a profit, and at the time of this question if it brings you a profit - then it is profitable, otherwise it can not be.

There may be a time difference before you started and when you asked a question, but I think it doesn't matter.

Honestly, I do not think that there are many profitable systems that work on a fundamental basis, anyway, they will be formed approximately the same, everywhere will be the same essence, but will differ depending on the experience of the trader, statistics, capitalization, application markets. I have been trading on X years, and my practice and experience shows that the strategy I use give a positive result in all the markets that I tried to apply, the only difference in the performance of the price movements ( I think this is because of the different capitalization, I don't know the answer to this question, but I think the price is to manipulate in any market) the Forex market, my trading strategy gives more maneuver in contrast to indices, stocks and bonds, I am only allowed to make a few short movements, which quickly closed with a few exceptions.

Just sharing my experience...

In conclusion, the answer to your question is - you have earned money over a X period of time.
Excuse me, my English is very bad...
  • #15
  • Jul 11, 2018 3:03pm Jul 11, 2018 3:03pm
Quoting AlastorFate
A clearer example, is a range bound trading strategy will usually lose money in a strong trending market. (vice-versa)
Ranging and trending are not "a change in the markets". At least not a fundamental change in the market behaviour. Every market ranges and trends. Trends are followed by ranges and ranges by trends. The duration and amplitude might change, but the fact that a market ranges or trends is just common sense. If you did not take ranging markets or trending in consideration in your system design and backtesting, than the problem is not changing markets...
Another point might be overoptimization. Certainly a curve fitted system gets blown by slight changes in the market.

I dont mean that to be offensive, but maybe your system development is not adequate?

Quoting AlastorFate
Correct, market movement is for the most part driven by fear and greed, buying and selling. However there is one issue. Fear opposes greed (vice versa) Buying opposes selling (vice versa)
That is just obviously wrong if you think a moment about it. Buying does not oppose selling, they belong together. How do you buy something that no one sells? Ist also not more selling than buying or the other way around. The amount of bought and sold cobtracts has to be equal. Always, at all time. Doesnt work another way.
  • #16
  • Jul 11, 2018 5:10pm Jul 11, 2018 5:10pm
Market behaviour

Analogy.

A person can be happy now and be sad/depressed later.

Same person. But different behaviour/emotion at different times.

Market ranges; and trends (the strong, big type).

But the issue is what strategies you want to use
Range bound strategy - Short at Resistance, Buy at Support
Trend strategy - Breakout

You have to pick ONE strategies to trade with.
One is the opposite of the other. It's like betting up or down, black or white.

Most traders stick to one style of trading, and comes the obvious problem one might be wrong sometimes or a lot of times. The market might trend strongly when one trades range bound; might ranges when one trades breakout.

Buying opposes selling. You win, while the person who bought from you loses. (vice versa)
+ vs -
Balance does not equate the equality/identical of opposing sides
Plus is not Minus

By oppose, I mean 'the opposite of'
North vs South
East vs West
Perhaps 'oppose' is the wrong word to use. What I mean to say is 'the opposite of'
'For the market to work, it needs people who think that they can beat it.'
  • #17
  • Jul 11, 2018 5:14pm Jul 11, 2018 5:14pm
once you predict direction correct EVERYDAY. then you are a trader. dont have to predict every move but the direction of the timeframe you trade you should never get wrong. no excuse. info is out there.
  • #18
  • Jul 12, 2018 12:48am Jul 12, 2018 12:48am
Quoting AlastorFate
Market behaviour Analogy. A person can be happy now and be sad/depressed later. Same person.
OT: to be sad is not even close to suffer from depression. Its statements like this that hold up the believe that a depression is just sadness lasting a bit longer. Depression is a serious condition and we should take it seriously!


Quoting AlastorFate
Market ranges; and trends (the strong, big type). But the issue is what strategies you want to use Range bound strategy - Short at Resistance, Buy at Support Trend strategy - Breakout You have to pick ONE strategies to trade with. One is the opposite of the other. It's like betting up or down, black or white. Most traders stick to one style of trading, and comes the obvious problem one might be wrong sometimes or a lot of times.
The point is, a change from a trend to a range is nothing new, its not a fundamental change in the sense of never seen before. Therefore we talk about two different things. If your strategy requires a market to stay in trend or range to be successful, sure, you are doomed to fail.

But...

You can survive times where the conditions dont meet your systems requirements. That is, at least in my opinion, the whole point of designing, optimizing and back testing a system. Done correctly you can trade any system approach long term, range or trend.
  • #19
  • Jul 12, 2018 7:10am Jul 12, 2018 7:10am
The change (usually very sudden) from trending to ranging market is enough to make trading difficult.

It is very difficult to predict how long a market will trend or range in any given time.

Optimal trading probably requires one to be right in the all the phases in any typical trade.
1. Pre market analysis
2. Entry (Difficulty: 3/5 imo)
3. Stop Loss (Difficulty: 4/5 or 5/5 imo)
4. Profit targets (Difficulty: 2/5 or 3/5 imo)
4. Live market/trading analysis decision (Difficulty: 4/5 or 5/5 imo)
5. Reaction to market moves - hold trades or close trades for profit or loss (Difficulty: 4/5 or 5/5 imo)
One has to be reasonably 'right' in every single phase of a given trade to make the most out of a trade i.e. maximizing profits, minimizing losses.

The problem with using mechanical system is over optimization,
with the intentional or unintentional assumption that history patterns repeat itself in the future.

My question to you...
1. Do you have a system that have been generating you consistent net profit for years?
'For the market to work, it needs people who think that they can beat it.'
  • #20
  • Jul 12, 2018 8:55am Jul 12, 2018 8:55am
Quoting AlastorFate
The change (usually very sudden) from trending to ranging market is enough to make trading difficult. [...]
Agreed and i never said trading is easy.


Quoting AlastorFate
One has to be reasonably 'right' in every single phase of a given trade to make the most out of a trade i.e. maximizing profits, minimizing losses.
It is not about "being right", its about doing the right thing, that is a difference.


Quoting AlastorFate
The problem with using mechanical system is over optimization, with the intentional or unintentional assumption that history patterns repeat itself in the future. My question to you... 1. Do you have a system that have been generating you consistent net profit for years?
I kind of stated that earlier, though some things will never change! low volatility is followed by high volatility, trends retrace, range markets occur! These things repeat itself and that will not change, or at least not easily.
Will there be a move of 50 pips after the crossed over 50 and the XY Indicator showed bullish? Probably not!

It doesnt matter if I have or if I dont you can take everything i said and forget it if you want to, i couldnt care less.

And once more, i didnt mean anything i said to be offensive.
My point only was, that by assuming, that it is the market changes that kill your else good systems you are looking in the wrong place.
With adequate testing you could have known before that they will not work.
If you know it before you can work on these issues of the system, instead assuming that a good system stopped working cause the market changed.
I would guess that it is not a very joyful way to trade or spend your time!

If it is, keep going and do what you did, if not, consider taking a step back and instead of working on developing systems, develop a process to develop systems.
0 traders viewing now