Are you really making money in the Forex market? - page 2
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Some other strategies can work well in manual trading but will not work well in automation
I'm trying to imagine a profitable manual trading strategy that can't be profitably automated, and I can't.
What would be an example of this, please?
(I'm not asking you to disclose any specific strategy--but just to describe the general nature of it).
There are many!
This is the case because many manual strategies are discretionary, and may also depend on analysing and perceiving the impact of human behaviour of "institutional interventions", or the "herd mentality" or the well know "market sentiment".
Many traders "manual" habits have been cultivated through years of experience and acquired skill, and it has become a "gut feeling" on how they trade. It simply is not possible to automate it completely, only partially.
EDIT: They still have a trading plan, and mechanical trading rules for entry and exits, but it remains discretionary for them, on when to apply those rules and when to deviate from the plan.
Although you may not know it yet... I believe that you're on the long road to fully automated trading.
I certainly didn't know that I was on that road years ago. But still... your nonfiction story about manual trading gives me the old heebie-jeebies.🤢
A bot never sleeps, gets over-caffeinated, argues with a mate, celebrates winners, nor gets depressed by losers. It's 100% rigid.
EDIT: They still have a trading plan, and mechanical trading rules for entry and exits, but it remains discretionary for them, on when to apply those rules and when to deviate from the plan.
Got it. Thanks.
I'll call that "You might have an edge," as opposed to "You have an edge or you don't."
I lost my Ethereum coin when I added unnecessary code in auto mode. If you're not a developer and accidentally paste phishing code, your entire crypto could be gone. Be cautious when dealing with automated trading.
I'm sorry to hear that.
As a generally understood rule, all code must be thoroughly tested in demo mode before deploying live.
There's nothing wrong with copying and adapting others' code for your own use per se but yes, you need some coding knowledge. Phishing code would certainly be evident to a developer.I'm trying to imagine a profitable manual trading strategy that can't be profitably automated, and I can't.
What would be an example of this, please?
(I'm not asking you to disclose any specific strategy--but just to describe the general nature of it).
Making a profitable EA is of course still possible. But it cannot have the edge of a seasoned manual trader.
At this point, I'm willing to accept that quantitative traders and manual traders assume different underlying philosophies. While quantitative traders rely 100% on numbers, statistics, math, algorithms, optimization, and/or historic performance, manual traders also rely on an element of discretion, intuition, emotion, and/or gut feeling. This is a classic contest of science versus belief.
I also agree that quantitative traders and manual traders cannot have the same edge. A competent quantitative trader's numbers don't lie. As the famous quant, Haim Bodek, said, any problem that arises is "your code." If you don't like the statistics generated by your code, then fix your code. It sounds simple enough but in reality, it's quite challenging. Another wiseman said that you should be able to explain your trading strategy to a person who doesn't speak the same language with nothing more than a chart. The point is that the quantitative trader has clear and scientifically deduced strategy elements that either produce acceptable profit or unacceptable profit/losses. The quantitative strategy can be empirically tested prior to deployment. Again as Giovanni Gramegna said previously in this thread, "You have an edge or you don't."
In contrast, the manual trader may use some quantitative tools but is free to deploy or abandon them at any given time. I think we can all agree that this is obviously unscientific and cannot be empirically tested prior to deployment. The point here is that the manual trader has intentional gray areas that may or may not result in consistent decisions based on the same data set that the quantitative trader used. Here, let's imagine trying to explain manual trading to the aforementioned stranger--not so easy. Again as I said previously in this thread, "You might have an edge," or you might not, or you might just be on a lucky streak. We simply can't know.
Price action based tools inevitably provide a plethora of false signals.
The point here is that the manual trader has intentional gray areas that may or may not result in consistent decisions based on the same data set that the quantitative trader used
But this is the thing - It's not only tactics and strategy, it's the art of experience. Trading is not exclusively a science and it's an illusion to think it is. Profitable trading is both a science and an artform. It's kind of like a fight in a UFC ring. A bull is fighting with a bear. The bull can't simply use a proven technique and think that he will win, he has to continuously adapt, learn the opponent in the moment, expect something unexpected, and not be predictable.
If we're going to analogize the Ultimate Fighting Championship and animal fights to manual trading, we might as well analogize horse racing and Texas hold'em to manual trading too. Interestingly, all 4 of those activities are gambling events.
A running loss is one of the most biggest caveats in automated trading. The bot does not care about it, it will freely grow it, and a manual trader can be intelligent about deciding when it's best to close the position.
One of the biggest benefits of algo trading is a system of rigidly enforced stops and/or loss mitigation. The algo does not experience the stress, happiness, sadness, insomnia, tiredness, over-caffeination, greed, fright, nor emotional swings of a manual trader. I would put one of my algo's up against any manual trader, and then watch the latter succumb to gut feelings as the latter doubles down and/or jumps out early.
But the argument still holds that you cannot automate art.
AI is already making art in the form of stories, images, sound, and videos.
We rely on markets to maintain a statistical element with volatility and momentum, otherwise automation would barely be possible other than automating news impact.
Professional quants get paid big money to quantify much more than volatility, momentum, and news events. Their algo's can analyze order flow, iceberg orders, social media content, and anything else that has a data source. Codependent weighting factors are applied to each data point. One has to wonder what special perceptions a manual trader has, the data for which a quant's algo is unable to analyze... if any.
Another thing is that forward fixed periods and fixed timeframes are problematic. Most indicators are using fixed periods. It could cause one year to be profitable, but not another year.
Regarding fixed periods, have a look at some of Mladen's work in the CodeBase for just a few examples of adaptive indicators.
Regarding fixed timeframes, my algo's disregard time aside from a 21 hour per day trading time filter. As a side note, good luck to a manual trader working 21 hours per day. My algo's run on custom charts.
Finally, thank you for participating in this lively debate. I see that you joined up here in 2023, so I'm not trying to trash your level of coding knowledge. I've been around since the mql4.com days so... I certainly didn't know much about coding within my first 2 years of finding MT4. I too, traded manually for about 12 years prior to that.