Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Winners scale winners, losers scale losers 12 replies

What methods do you use to let your winners run? 38 replies

If losers average losers, then I am a loser 10 replies

My trading journal-winners and losers shown 137 replies

Winners And Losers Trading System 138 replies

Attachments: Do you manage losers and let your winners run?
Exit Attachments

Do you manage losers and let your winners run?

Quoting skyway
The key to manage drawdowns as in to not be exposed to large drawdowns is to not open trades during unfavorable random market conditions.
....and there lies the rub. How to identify unfavourable market conditions (which you can clearly see as an outcome in your equity curve).

I am hopeless at any form of preventive technique to avoid them apart from dynamically adjusting (with lag)...however IMO this is where portfolio management using uncorrelated return streams has the ability to achieve this with all composite strategies turned on 24/5.

If you apply a single trading strategy that is either convergent or divergent in nature then you need to be able to 'filter' for market condition......For example if you run a live walk forward demo in tandem with the your strategy then you can use the equity curve of your demo to turn on and off your live trader. As drawdowns are building 'turn it off'......but once they start turning......turn your strategy back on. It still is lagging in nature and so far at least for me is less suitable than a portfolio approach using non-correlated divergent and convergent strategies......however I have not really fully examined this technique so far.
Quoting Copernicus
{quote} ....and there lies the rub. How to identify unfavourable market conditions (which you can clearly see as an outcome in your equity curve). I am hopeless at any form of preventive technique to avoid them apart from dynamically adjusting (with lag)...however IMO this is where portfolio management using uncorrelated return streams has the ability to achieve this with all composite strategies turned on 24/5. If you apply a single trading strategy that is either convergent or divergent in nature then you need to be able to 'filter' for market...
I have tried a lot of carp and at the end it simply boils down to how much time and effort one is willing to put in to trade their strategy. I know for sure trading is NOT for me if I have to put in more than 2-3 hours a day. Part of my automation is everything around this aspect as I turn 50 next month. Time and effort (and stress) has no relation to how much profit you can yield out of markets. My current discretionary style is fastest amount of profit with least amount of time but comes with a lot of stress and limited scalability. This is where my focus is going to be for the next 10 years. Its going tp be more income/cash flow driven s i can no longer assume recovery from drawdowns with the hopes and expectations of long term growth of capital. Just a $1000/week cash flow is more than enough (for me). I am alao at a stage where I know where my own personal weakness lies. What I have control over and what i just cannot. The compliance manager function of my EA is designed to be my boss forcefully correcting any actions that i take which is not part of my trading plan. (Just sharing so others can know there are many many ways to develop and impose self descipline. Example: Any bad behaviour and my EA will close one trade in loss as a punishment. Another example: if i get greedy and stack too close to my open position, it will automatically close the trade out in loss or profit to avoid over exposure. Another example: Stop Out Level is the worst form of scam by brokers designed to drain your account beyond recovery. Margin call must occur when your margin level has dropped to 100%. You have no business trading or keeping open positions if your floating losses becomes greater than. Used margin. My EA does all this sh*t for me so i don't have to worry about it...same with drawdown % from peak equity. Its a beauty to see it in action when it starts partial closing one microlot the second floating loss reaches a predefined %. I am sharing this because there are literally 100s of alternative ways to limit, control, reduce and eliminate risk outside of putting a hard stop loss on a per trade level.
Staying in my lane...
Quoting VEEFX
{quote} I have tried a lot of carp and at the end it simply boils down to how much time and effort one is willing to put in to trade their strategy. I know for sure trading is NOT for me if I have to put in more than 2-3 hours a day. Part of my automation is everything around this aspect as I turn 50 next month. Time and effort (and stress) has no relation to how much profit you can yield out of markets. My current discretionary style is fastest amount of profit with least amount of time but comes with a lot of stress and limited scalability....
We come to this stage of thinking only after we reach such length of living.

https://www.forexfactory.com/showthr...9#post10576779
https://www.forexfactory.com/showthr...9#post10577039

Another approach is to trade the longer TFs, once a week/month/year trading.
Quoting skyway
The key to manage drawdowns as in to not be exposed to large drawdowns is to not open trades during unfavorable random market conditions. This means that each individual potential trade has to be rigorously risk assessed prior entry. When a drawdown has already occurred, the drawdown management process that follows to limit damage is a little too late.

All what you said can be replaced with just one word: discipline.
Attached Image
NO MATTER THE SITUATION,NEVER LET YOUR EMOTIONS OVERPOWER YOUR INTELLIGENCE
Quoting hanover
{quote} Sounds like you might be a quant trader, whereas there are institutional traders whose analysis includes other techniques, which could account for some of the differences between our conclusions.
Well, I can't call (confidently) myself a quant yet. But this is the natural evolution in the trading world. And like most people I want to evolve one way or another!

Actually, up until recently I was big supporter of discretionary trading. (And I still am to some extend)
Because despite my efforts, I still have not found a completely reliable way to eliminate the human judgment from the trading equation.
However, as time goes by I'm getting closer and closer to this goal.

Currently I'm quite deep into probability theory, time series analysis, and machine-learning. Trying to expand my knowledge by leaning new things every day...

My ultimate goal is to create a system which brings together the best of both worlds.
On one side we have human judgment (so far proven indispensable), and on the other hand we have robots which never make mistakes and do exactly as they are told. Now the billion dollar question is, how to combine these 2 things? How to reduce human judgment into models with exact formulas? What are the inputs?

We should all keep in mind that, In the end of the day, trading is math and nothing more! The computers are extremely good at math(calculations and speed) once you give them structured problem to solve.
And we the humans are quite good at pattern recognition. We are not very fast, but we have the unique ability to "connect the dots" in the sea of "random" variables. We also have the ability (intuition) to sense danger.
Quoting alphaomega
{quote} Well, I can't call (confidently) myself a quant yet. But this is the natural evolution in the trading world. And like most people I want to evolve one way or another! Actually, up until recently I was big supporter of discretionary trading. (And I still am to some extend) Because despite my efforts, I still have not found a completely reliable way to eliminate the human judgment from the trading equation. However, as time goes by I'm getting closer and closer to this goal. Currently I'm quite deep into probability theory, time series analysis,...
its a very good post

actually the robot game is no different to the human game... 95% of robots lose to the 5% better robots

one thing a robot cannot do is break away from their programming without human intervention

and humans made the robots.. isn't that a funny thing?

Inserted Video
Time hides Nothing
Quoting alphaomega
{quote} Well, I can't call (confidently) myself a quant yet. But this is the natural evolution in the trading world. And like most people I want to evolve one way or another! Actually, up until recently I was big supporter of discretionary trading. (And I still am to some extend) Because despite my efforts, I still have not found a completely reliable way to eliminate the human judgment from the trading equation. However, as time goes by I'm getting closer and closer to this goal. Currently I'm quite deep into probability theory, time series analysis,...
Nice post AO. There is a reason why I don't refer to my EA as a robot. My 'automation' is the epicenter of what you just described. The best way to introduce human elements into a robot is to design your robotic features into small building blocks that mimic human actions and decision making capabilities. Not just human action but every aspect of the trading plan. Once your automation is structured this way, then the next and crucial step becomes when to employ such building blocks (soldiers) to work in unisom in command from the mothership(strategy).

Most robots fail because they expect the markets go fuction one way in a repeatable form. Once you figure out how to make the robot more adaptive within the context, it becomes the best of both automated and discretionary trading world.How I do it is to parameterize everything making me in command of my automation via frequent compilation (replicating AI and machine learning abilities which is currently too advanced for my coding abilities but i will get there some day)

Pattern matching is done without intervention but motion behaviour management within the context of Newton's law of motion is very hard to automate via fixed or variable sets of rules. I knew that from day one before delving into learning to code and start on my automation.
Staying in my lane...
Quoting Sis.yphus
{quote} This isn't about the R:R. The 1:1 RR was used to make the example simple. The question is do you manage the losers and let the winners run. This applies to any R:R.
Well, if your tp was nearing and you lifted your tp, wouldn't that mean that you aren't following your original setup?
Quoting VEEFX
{quote} Nice post AO. There is a reason why I don't refer to my EA as a robot. My 'automation' is the epicenter of what you just described. The best way to introduce human elements into a robot is to design your robotic features into small building blocks that mimic human actions and decision making capabilities. Not just human action but every aspect of the trading plan. Once your automation is structured this way, then the next and crucial step becomes when to employ such building blocks (soldiers) to work in unisom in command from the mothership(strategy)....
Would it also be fair to say that "most EA's fail" because the programmer / trader didn't account for "variations" of his/her system. Variations would also include High impact news.
Quoting alphaomega
{quote} trading is math and nothing more! The computers are extremely good at math(calculations and speed) once you give them structured problem to solve. And we the humans are quite good at pattern recognition. We are not very fast, but we have the unique ability to "connect the dots" in the sea of "random" variables. We also have the ability (intuition) to sense danger.
I disagree with what you have written.

Trading is more than a maths problem

Humans see patterns in everything and most patterns do not repeat themselves in the long run

The ability to sense danger prevents noobs from being profitable

You came here for the truth and let me unveil it for you
Quoting RondaRousey
Trading is more than a maths problem
In Jack Schwager's Market Wizards books, some of the interviewees (quants) see trading as more of an exercise in crunching data using statistics, and expectancy math; while others see it as more of a nuanced art. Hence I believe that you could argue a case either way. Of course you still need the personal qualities (patience, nerve, discipline, self-belief, etc etc) also.
Quoting RondaRousey
{quote}Trading is more than a maths problem
Daily price movements cannot be fully reduced to some quant math models.

Quoting RondaRousey
{quote}The ability to sense danger prevents noobs from being profitable
That's true for noobs and veterans as well.
Forex Trading is the best way to make some good money online, but we need to focus on both losses and profits and not only on the profits.
Actually it depends on any particular entry. There is a particular reason to take an entry. And i consider many things to decide whether to run or close. Market condition, risk reward ratio, my target, why did I take the entry, by considering all these I take decision.
Had similar line of thought... Seeing how the tiny small losses add up, when keeping other R:R than 1:1, is not a pretty sight. Also, leaving trades to run, often ends up with leaving money on the table. So even if there is "good" R:R with a 30/40 win rate, many trades end up BE, so it's not actually 30/40% win rate. Also a reasonable trailing SL, in most of the cases, is leaving half of the gain there. Depends on the trend.
So instead of trailing SL, or fixed R:R, I think it's better to define an exit strategy. It has the same importance as the entry. So yes, managing the winners.
Letting losers run... did that... It's great when it works out, but the one time when it doesn't... So, no.
There is a point, where the trade breaks out on the other side, even if it is trend strategy, or reversal. So there is no point in leaving the looser run.
Yet this does add up... It's not easy.
I'm always fascinated how the FX machinery works. Very well designed. Leaves almost no chance, and makes it look like there is plenty. And there is... those moves are there, but...
We are where we are. Not up, or down. Past Equity is not a reason to trade.
  • Post #56
  • Quote
  • Edited 1:38am Oct 4, 2020 1:17am | Edited 1:38am
Quoting CPxiom
Had similar line of thought... Seeing how the tiny small losses add up, when keeping other R:R than 1:1, is not a pretty sight. Also, leaving trades to run, often ends up with leaving money on the table. So even if there is "good" R:R with a 30/40 win rate, many trades end up BE, so it's not actually 30/40% win rate. Also a reasonable trailing SL, in most of the cases, is leaving half of the gain there. Depends on the trend. So instead of trailing SL, or fixed R:R, I think it's better to define an exit strategy. It has the same importance as the...
"I'm always fascinated how the FX machinery works. Very well designed. Leaves almost no chance, and makes it look like there is plenty."

This is very good observation!

Buy low, sell high, adding (averaging) to losing trades and letting losers run until profitable is what works in the Forex market most of the time.
With this strategy you can make profits almost every day and stay profitable for months and even years. It's also very easy to automate.
But eventually you will lose it all. Only one really big and smooth move against you is enough and all profit is gone + additional "interest". The interest is usually exactly equal to the total amount of paid spread and commissions. You can replicate exactly the same result with random data set with 50% probability. So it turns out the market is random 99% of the time. Sort of.....

The alternative is to invert the strategy and to add aggressively to your winning trades waiting for that big and smooth move. But the end result is the same.....

What I find most fascinating is that so many people think they can somehow outsmart the mechanics of the market by clever discretionary timing of the entries and exits. They just don't understand how this scheme works. You just can't outsmart the law of large numbers.

The LLN is important because it guarantees stable long-term results for the averages of some random events. For example, while a casino may lose money in a single spin of the roulette wheel, its earnings will tend towards a predictable percentage over a large number of spins. Any winning streak by a player will eventually be overcome by the parameters of the game. It is important to remember that the law only applies (as the name indicates) when a large number of observations is considered. There is no principle that a small number of observations will coincide with the expected value or that a streak of one value will immediately be "balanced" by the others
Quoting alphaomega
{quote} "I'm always fascinated how the FX machinery works. Very well designed. Leaves almost no chance, and makes it look like there is plenty." This is very good observation! Buy low, sell high, adding (averaging) to losing trades and letting losers run until profitable is what works in the Forex market most of the time. With this strategy you can make profits almost every day and stay profitable for months and even years. It's also very easy to automate. But eventually you will lose it all. Only one really big and smooth move against you is enough...
Money is the raw material of the game. That is for sure.
But major issue is your mindset.
Most of the time it is not the fear of losing money, it is the fear of losing your own face.
If you devise a system that can force you to be disciplined in exits, you are a winner.
Fear and greed influence exits.
Decision to enter is not tough one if you are watching the screen. Scalping with small hits of 100 is not difficult
What is difficult of calling it a day even after hitting 100 in a row.
Programme close all with profit or close all with loss as greed controllers and stick to it. Will you not change to another figure after hitting 100 in a row?

Or for positional trading, just take a toss whether it will go up or down.

Main problem for all losers is not quick hit, it is big hits.
Practice makes a person perfect
Quoting alphaomega
What I find most fascinating is that so many people think they can somehow outsmart the mechanics of the market by clever discretionary timing of the entries and exits.
Agree on - thats not very "important" in the longterm. Look at any strategy for entry/exit as a fluctuating process. On average (for "normal" entry/exit strategy) risk/reward 1:1 will give 50% +/- profitable trades. Risk/reward 4:1 will give 80% +/- profitable trades. And profits are locked in those "+/-"... With r/r 1:1 if 51% trades profitable then trading is profitable. But with 49% - loosing.

In longterm, there are fluctuations in any strategy. If in longterm average profit trades are 50% then there will be extreme periods when that % can be 30% and periods when it can be 70% - so on average 50%. But mostly it fluctuates around +/- 1-5% of the median.

Those fluctuation periods can be months or even years... And you never know when and how a particular strategy will fluctuate with performance.

BUT... if you take a bunch of strategies, for example, have ~100 EAs running with various r/r proportions then the whole package of randomness now behaves differently.

It's not possible that such a system of EAs go in 1 straight way very long time... the whole system will still fluctuate... but that fluctuation will be more stable as if taken 1 particular EA.

So run your ~100 EAs package on demo account and trade balance line - when drawdown -x% (which you observed in 10 years history of entire EAs package) then you turn on trades copy to real account and turn off copy when y% profit.

100 EAs not allways 100... just example. It could be 10 or 20 or any number EAs you find fluctuating stable.
  • Post #59
  • Quote
  • Edited 7:04am Oct 4, 2020 4:44am | Edited 7:04am
Truth is this FX game is complicated and doesn't lend itself to absolute statements, like you see here all the time spewed out with a large degree of attitude.

I am entering heavier when very confident, using multiple orders. I add with size if it spikes against. If it retraces back in my favor, I take out the initial entry (losing) offsetting with the gains from the add (in part, to maintain directional position, at even better price, or in all, to exit). If it shows too much momentum, opposite or my position risks getting to big, I may SL or hedge to freeze the loss and re-examine.

If it moves in my favor I may close one or more in profit, or open opposite order, partial hedging to lock in profits. If it moves against I, again, increase on unfavorable impulses and as it slides again in favor, I take out heaviest losers with gains from latest gainers (not the largest gainers in the initial entry; those I try to keep or close on favorable impulses on new lows/highs). Naturally, if the movements against don't retrace, then I may have to take net losses or try to compensate with gains from the (partial) hedge (if I have set it up). If it retraces to my average price in the original direction, I may take everything out.

So, the adding to losers, as long as I am confident in my directional prediction, is done, but I try to maintain proper overall position size (take out heaviest losers with winners) When the unfavorable retracement ends, I always have my add right at the top of the retr. (natural consequence of the method) as price resumes the predicted direction. The average entry price is kept far from current price. On intraday it may get hectic and certainly the broker is very happy, as the operations can easily stack up to dozens in minutes.

How does this square up to the dogmas about the idiocy of hedging or setting SL and TP or not adding to losers? I, of course, don't care about people saying that. As for results, I don't have a large enough sample to draw conclusions.

A lot of judgement and intuition goes into the exact specifics when and how heavy to increase, reduce, take losses, take profits, close everything. One thing is for sure; in the end being right about a direction is the key. Wouldn't it be better to just enter and SL TP? Maybe... but I don't know what's going to happen and this way I am more engaged with price action.

Edit. Too timid or missing an impulse in the averaging initial entry or adds to winner in retrs. and I may have to take losses to preserve position size. Too aggressive and position gets too big for comfort, and average position price gets overtaken by price and I'm forced out, or hedge all...

Edit 2. The core concept is the active management of a position, its average price, its size, the floating gain, to never go under water and never oversize at least while there is risk of position being overrun. I am thinking I should also manage a smaller opposite position at the same time, carrying it at constant limited floating loss as price moves in my favor, to use it in retrs. offsetting the premature adds as long as the retr. continues and the adds get overrun.
I view this as akin to when I learned to drive by managing clutch, accelerator, handbrake, gears in hill starts. It gives confidence and makes all senses more acute. The focus required to this must not however go to the detriment of seeing the chart and other relevant information.
Temperance (restraint in action, thought or feeling) is a virtue.
Quoting alphaomega
{quote} "You can replicate exactly the same result with random data set with 50% probability. So it turns out the market is random 99% of the time. Sort of..... The alternative is to invert the strategy and to add aggressively to your winning trades waiting for that big and smooth move. But the end result is the same... Any winning streak by a player will eventually be overcome by the parameters of the game."
Yes, it looks like it is random, but I guess it is a design. If we can replicate the same result with random data set of 50% probability, then that means that it gives no edge to trend systems or countertrend systems, or variations of these two. Actually it can't behave differently, if we think about it. (To the extent that it's not even allowed, we could say). Up, down, trend, range, all currencies. This has to be this way... because it's something like a mathematical law, and comes naturally, or because it is controlled... designed, hardcoded. But this is a longer discussion. (By the way I have read your Revert to mean thread and it resonated a lot).

So for us traders, if all winnings of a system are cancelled, as you also say, (whether we keep the system running 24/7 or not, because in the long run, you save some trades, but also miss some trades, 50/50 is a good guess I think... ) then this means that the answer is not a "good" system.
Maybe like in nature, there is a time when the fruits are ripen, so we harvest. If the fruits are left there, then everything goes back to the ground.
We are where we are. Not up, or down. Past Equity is not a reason to trade.
0 traders viewing now