Human Plus Machine


ivo Luhse

By Ivo Luhse


Problem 

Trading in financial markets is painfully difficult, and many traders are simply terrible at it. In fact, 82% of independent traders lose money. Why? And what can be done about it?

For many traders, they only have themselves to blame for succumbing to greed and gambler's felicity. The industry has long been playing tricks with traders and preying on the “fish" - the proverbial “sucker” at a poker table. Showing images of supercars, yachts and beachside trading in their marketing campaigns for years, brokers have filled traders with attractive dreams of easy money. With only the press of a few buttons on your laptop you can make a fortune while sipping Pina-Coladas on a Maldivian beach. 

Brokers will attract aspiring traders with 200:1 leverage, allowing you to trade heavily on margin. Come and trade like a boss with $1,000,000 buying power by depositing only $5,000: a mathematical certainty to kill your entire deposit with just a 0.5% market move.  

Once you deposit your hard-earned cash, this is where the real slithering nature of the industry comes out. Market makers will entice you into positions so that you become emotional and think irrationally. This is achieved by moving the market and tricking you into taking a position that they later reverse. They will hit your stops and clear the board, the practice of moving the market through order clusters. How many times have you seen the market hit your Stop, and then continue in your anticipated direction? They see every single one of your trades including current and past trades. They know your trading habits better than you know them yourself, which is another shortcoming for nascent traders; they ignore keeping a rigorous record of their trades.

You are actively encouraged to day-trade, follow the "trade recommendations" and copy trade. The larger the size, the better! Ultimately, you will be forced into a margin call and out of the game. 

Now all of a sudden, your trader’s dream has turned into a nightmare.  

The industry treats new traders like terminally ill patients; 90/90/90 rule is a case in point. It states that 90% of traders will lose 90% of their money within 90 days. So, their job is just to keep the traders happy until they lose all their money, typically within 90 days. 

Are you still interested in trading?

Okay, you must be a really stubborn individual – a bit like me when I first started trading Forex in 2005. As a young engineer, I was fascinated when my father, who worked as an analyst in the 1990s for Russian investment bank Menatep, showed me one of his trading platforms – the Omega 3000! It was a state-of-the-art market analytical platform complete with customisable charts, alerts and colourful indicators. 

“Wow”, I thought, “this is amazing! This is what I want to do for the rest of my life.”

So, when I finally saved enough money to open my first Forex trading account in 2005, I was full of excitement and valour. Naturally, while I had all the enthusiasm in the world, I lacked any substantial market knowledge. As such, the market quickly relieved me of my grand fortune of $10,000. 

I was the fish and the market sharks had gobbled me up in less than 90 days. I confirmed the 90/90/90 rule. 

First, I was shocked that it happened to me – How could it be when I was disciplined and followed a “system” and risked “only” 3% per trade? But of course, as it turned out, the system was a scam — a completely made-up strategy based on false beliefs, sold online for $997.

After the initial shock passed, the drive to really understand how the market worked kicked in and I wanted to know how the professionals make money trading the markets. I was hooked. Deep down, I knew I could figure this out. I would not give up without a fight – plus I really could not see myself working for someone else in a traditional nine-to-five job. 

After a few more failed attempts at trading on my own, my epiphany came when I discovered one thing in common among some of the most successful hedge funds and money managers – they used bots. 

This became my obsession and my passion – building machines to better trade the markets. I became an Algo trader.

Solution: Machine 

  • It is much harder to trick a machine. 

  • Machines do not encounter the emotional shortcomings of human traders. 

  • The machine will not get tired after 8 hours or need sleep. 

  • It will scan thousands of markets 24/7 to search for the best trade opportunities autonomously.

  • It will not get distracted or move with the herd or get influenced by others’ opinions and fake news.

  • It will not get greedy after a successful winning run. 

  • It will not panic during an inevitable losing streak.

No wonder all the corporate traders today use computers to build and execute their models, whether they are crypto-hedge funds or government-backed pension funds.

However, it is not as easy as just turning on your new Robo friend and keep sipping those Pina-Coladas. Believing that it is, is a classic trap into which every novice Algo trader falls. They confuse performance – the ability of a machine to surpass the result of a human – with the process through which those results are achieved. Add to that a strong psychological desire for humans to distance themselves from accountability and responsibility, especially if the machines appear complex and alien, and you can see why many of those who tried autonomous Algo trading have also failed.  

Machines are excellent tools that help traders make better trading decisions in the markets. But without an intelligent human, the machine is just a pile of wasted metal (or wasted code, in Algo trader’s case).  

Machines are good at repetitive, mechanical tasks such as analysing a specific market pattern with lighting-speed. But machines are bad at tasks that require abstract thinking rather than computation, and individual-specific responses rather than codified ones. For example, only you can create your trading plan with and goal of quitting your job in 5 years without risking more than 20% of your savings.

Machines cannot think for you because they have no idea what is important to you and carry no common knowledge and awareness. Unless we directly specify them, machines have no way of knowing whether some actions matter to us more than others.

Mathematician Anton Holt said it the best:

Artificial Intelligence is a good description of any intellect that can make a brilliant chess move while the room it’s in fills with smoke as the house is burning down!

This is why rather than rely solely on the intelligent systems, humans must use these machines instead as a tool to help, guide and improve decision making.

Humans are good planners, but at the same time, we are quite bad at executing our plans. Think about all the plans you have made throughout your life that never came to fruition… This is known as the Moravec's paradox: what machines are good at, humans do poorly; but in what humans excel at, machines perform very poorly.

What if we combine human and machine strengths and work as a team? Human Plus Machine

Better Solution: Human plus Machine…

Using this philosophy, I developed DARA, short for Decision-Aiding-Robo-Advisor. DARA is an intelligence-augmentation robot that helps traders finding the optimal trading patterns in the markets.

However, DARA is not a simple brute force robot that aimlessly searches for any patterns that fit the past data and it does not operate assuming that the future will remain the same. This type of black-box system is only good for producing perfectly-curved backtests. They over-fit the data – essentially finding spurious patterns where there are none – these systems have little-to-no value in real-life trading. 

Profitable trading systems arise from careful research and an ideas-first approach. Done by thoughtful and knowledgeable people who seek to better understand the origins of their profits.

The loss-making trading systems, on the other hand, typically result from the use of random data mining. These systems promote the practice of analysing data with the hope of stumbling on occasional profitable rules, without considering the reasons why the rule might have appeared profitable in the first place or whether it will be profitable in the future. 

My experience, combined with my preference for trading signals that I can trust and understand, means DARA is an Ideas-First robot. DARA is tasked to search the markets for simple technical trading patterns and human emotional patterns that are repeating over and over again. These patterns are then run through DARA algorithms, and only the patterns that are historically profitable and have positive expectancy are sent to the trader. This means that the trading signals are not only rigorously verified for profitability, they are also intuitive, simple and easy to understand. It is crucial for traders’ long-term success to not only know what is the best trading pattern in the markets, but also why it is the best pattern.

DARA structure

DARA Structure

  • Layer 1 - Inputs

    24/7 new information is collected about the state of the market - Open, High, Low, Close prices, gaps, volume and various technical indicator values (Moving averages, ATR, Bollinger bands, RSI, Strength/Weakness to name a few).

  • Layer 2 - Labelling and pattern recognition

    The incoming information is carefully structured, and various patterns are labelled. A large number of technical patterns can be identified - Candlestick trading patterns (engulfing patterns, pin bars, outside bars, doji etc.); Chart structure patterns (volatile trend, slow trend, rangebound, double tops, double bottoms, support/resistance etc.); Technical indicator patterns (Moving average cross, Overbought/oversold etc.). The pattern list can contain tens of thousand label combinations and is always evolving.

  • Layer 3 - Performance testing

    Pattern performance testing is done 24/7 in parallel mode with all the incoming information. Each new incoming pattern is compared with similar patterns historical performance (success rate of the pattern).

  • Layer 4 - Expectancy calculation

    Projected expectancy formula is used to find the patterns that have high enough expectancy worth considering to trade.

  • Layer 5 - Output signal message

    Finally, patterns that pass individuals expectancy threshold are sent to the trader in a simple dialogue format.

DARA Features

  • DARA will search for these patterns 24/7 in your specified markets and will send you a message only when something worth trading comes up. You no longer need to stare at a computer screen all day long. 

    DARA will provide all the necessary data in a simple dialogue format as if you were speaking with your trading buddy. It will offer information such as trade expectancy, reward to risk ratio, and probability of success; all the information you need to make an objective decision.

  • DARA can even execute your trades through your brokerage account and manage them for you. (Requires API connection to your broker) 

  • DARA will also help you control and manage the risk like a professional.

  • DARA will keep a journal of your trades and analyse them so that you can improve your decision-making process.  

  • You can backtest the strategies to gain valuable insight and trust for the strategies.

    The machines do all the heavy work while a human controls the machines from a simple and intuitive Web APP.

This type of mental muscle-building exercise combined with strong discipline and the guidance of a mentor allows you to gain the necessary experience and actively build your confidence. Only then you will enter "the zone" - make trading decisions intuitively and effortlessly as one Human plus Machine, Just as easy as you thought it would be when you first discovered trading.

Thanks to DARA, trading really has become as easy as simply pressing a few buttons on your phone while sipping Pina-Coladas on the beach in the Maldives. 😎

Triumph of Process

DARA Triumph of Process

Human intelligence augmentation (IA) devices are nothing new, and the practice is as old as the machines themselves. 

Many will remember Chess World Champions Garry Kasparov's defeat to IBM’s $10 million supercomputer in 1997, but not many know that after the defeat, Garry created a human+machine programme called "Advanced chess" that paired humans and machines on teams. The surprise came at the conclusion of the event. The winner was revealed to be not a grandmaster with a state-of-the-art PC but a pair of amateur American chess players using three computers at the same time.

Here is what Garry Kasparov said about Advanced chess:

It was a triumph of process. A clever process beat superior knowledge and superior technology. It did’ t render knowledge and technology obsolete, of course, but it illustrated the power of efficiency of coordination to dramatically improve the results. I represented my conclusion like this: weak human + machine + better process was superior to a strong computer alone and, more remarkably, superior to a strong human + machine + inferior process.

Today, more people than ever play chess, and all Grandmasters use strong engines to train and practice their skill. But this new reality is still the result of human labour. It takes 10,000 hours to become an expert in a field such as chess or trading the markets. 

Technology, used correctly, dramatically reduces that time, making training far more efficient.

You can start training with DARA today for FREE.

Our mission at DARA is to focus on IA, intelligence amplification, to better use information technology as a tool to enhance human decision-making instead of replacing humans with autonomous AI systems.