THE CASE FOR TRADERS COLLABORATION
When Crypto-Trading Gets Social
Few traders join forces spontaneously, but industry trends may push collective trading to the forefront sooner than later
We recently celebrated Satoshi’s whitepaper’s tenth anniversary.
Crypto is slowly but steadily growing up.
The child will soon turn into a man and, thus, be expected to wear a suit and tie.
Long gone are the times when an incipient industry was driven by idealist souls sharing the values of freedom and power to the people that transpire from the libertarian interpretation of the potential of cryptocurrencies.
Nowadays, the suits are taking over.
Be it due to the barriers imposed by regulators who seem eager to make sure the cost of compliance is only affordable to deep pockets, or simply because it’s the nature of capitalism, the industry is maturing and attracting increasingly larger capitals, along with their heavyweight corporate structures.
In a sense, that should be good news for traders. Or — should I say — for trading?
Arguably, larger and more efficient markets with higher liquidity should be good news for everyone. However, not all traders are well equipped to deal with the new competition and potentially lower long-term volatility.
If there was ever a time in which amateurs could reliably increase their capital trading crypto, that time is gone. As crypto-trading professionalizes and big firms put their algorithms up to speed, the harder it is for small players to compete and have any measure of reliable performance.
The more investment firms get in the business of trading crypto, the higher is the barrier to enter or stay in the game for individuals.
The more advanced the trading algorithms firms deploy, the slimmer the opportunities left for independent traders — in particular, those who can’t code.
It is inevitable; a matter of time.
The dreams of thousands of crypto early adopters who taught themselves to trade in the hope to make a living out of trading crypto are vaporizing.
It’s ironic, as this group of enthusiasts certainly contributed their fair share to bootstrapping the industry at times when corporations were not interested in the game.
The Trading Food Chain
Most amateur traders are at the bottom of the food chain.
They tend to be the most vulnerable player because they have several disadvantages, starting with their comparatively shallow understanding of the markets and relatively lower experience.
Also, amateurs seldom have access to state-of-the-art tools available to professionals in investment firms.
With amateur traders silently walking down the path to extinction… Who’s next in the food chain? Whose carcasses are corporate robots going to feast from next?
Let’s take a look at the open grasslands… the prairie of crypto markets…
Who’s the next vulnerable herbivore?
Who is at the largest disadvantage against the newly emerging predators?
I would argue individuals doing manual trading are next in line.
Those who can’t code their trading algorithms depend on third-party solutions that seldom fulfill their needs, as they are either too expensive, clumsy, limited or overly complex to use.
Most of those non-coding traders will rely entirely on the limited resources they have as humans; the best-case scenario is one brain, a pair of hands, a pair of eyes, and the time remaining after sleeping and dealing with the rest of the mundane feats we all need to deal with daily.
Indeed, comparing what a single human can do with its limited resources against a swarm of algorithms with unlimited resources is disheartening.
Let’s save ourselves the sorrow and not go in the details.
Suffice to say that even pro traders and portfolio managers at a wide range of firms worry about the increasing participation of algorithms in the markets.
Change is the law of life. And those who look only to the past or present are certain to miss the future. — John F. Kennedy
Evolutionary Lessons
In nature, it is the outcast, the elder, the underage and the physically impaired specimens who are the most likely to end up staring death in the face.
In the wilderness, large herbivores move around in tight herds to protect the weak.
Predators team up to hunt in groups to improve efficiency.
Some insects organize in colonies where collective forms of consciousness emerge.
Collaboration is a fact of life, and humans learned the lesson early on too.
In union there is strength — Aesop
Scholars argue that one of the resulting features of the bigger brains in Homo sapiens may have been instrumental in acquiring an edge over other human species: the capacity we developed to cooperate flexibly in large groups.
In fact, modern humans build the largest and most complex social structures that have ever emerged in nature.
Each of us is simultaneously involved in a multiverse of parallel and interconnected collaborations: starting with our family structure, the interaction with our neighbors, our kids’ school organization, our city, state and national institutions, our affiliation to sports, religion, the arts… everything we do involves collaboration.
Even doing business requires collaboration between organizations and people within them too.
Indeed, all significant advances achieved by humankind have been the result of collaborations. Try to think of one single invention, conquest, discovery or enlightenment that may have been the product of the efforts of a single individual. Let me know if you find one.
Why is it then that there is so little collaboration in trading?
Trading Intelligence Evolves in Silos
Investment companies acquire significant talent in the form of data analysts and tech resources; however, they may be failing at organizing human and machine resources in a way that would yield the best possible results.
Sometimes due to the structure of incentives, workflows, organizational requirements or segmentation of incumbencies — save for a few exceptions — there is little collaboration within traders in firms.
In fact, small teams and many times individuals work in absolute secrecy.
Needless to say, there certainly is zero cooperation among firms.
Secrecy is believed to give market participants an advantage over the rest of the field. This notion is particularly entrenched among players who see the markets as a zero-sum game. Also, there is a general belief that algorithms and strategies lose their edge when used too broadly.
Despite the advantages investment companies may think they have by working in silos, the fact is that no trading intelligence has ever evolved to the point of dominating the space or achieving a remarkably superior performance — as Homo sapiens did as a species at some point in time throughout the evolutionary race with other human species.
May that fact be a consequence of the closed model itself?
May lack of collaboration explain the non-occurrence of a dominant intelligence ever emerging?
Lack of collaboration may indeed be a systemic weakness, common to most actors playing in the markets.
The Case for Collective Trading
One plus one equals two, every time.
If you agree that individual traders — especially, those who can’t code — are at a disadvantage and that working in silos — in particular when doing manual trading — may add to the existing handicap, then it should be easy to conclude that there is a worthy experiment to be made.
Such an experiment would involve getting a few traders to collaborate and — together — overcome the basic hindrance of working in silos. Adding developers in the collaboration would solve the technology issue as well.
Piece of cake, right?
Well… not really. That’s a start, but…
To build a system, you first have to imagine all the ways to destroy it. — Daniel Jeffries
Indeed. If we are serious about exploring the idea of collective trading, we need to first realize that nothing is as simple as it seems at a glance. There are dozens of factors that could make such simplistic arrangements fail: greed, poor communication, an inappropriate form of organization, an incorrect set of incentives — to name a few.
So, before we pull out the drawing board and start drafting a plan, we should probably get to the core and explore the fundamental aspects of collaboration.
Intelligence & Collective Intelligence
Let’s start with a few questions…
What makes a group more intelligent than any of its members?
What would it take for a group of traders and developers to collaborate effectively?
May a system be purposefully designed to maximize the group’s collective intelligence?
It seems sensible to start with a few fundamental definitions…
Intelligence has been defined in many ways and through many lenses. In her paper Mainstream Science on Intelligence: An Editorial With 52 Signatories, History, and Bibliography, Professor Linda Gottfredson (University of Delaware) defined intelligence as:
… a very general mental capability that, among other things, involves the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience.
This is the kind of intelligence that standardized intelligence tests measure.
It is generally accepted that such tests predict the individual’s ability to perform well on a wide range of tasks beyond what the specific test is measuring.
For instance, people doing well in standardized tests are expected to have a reasonable performance in reading, comprehending texts, solving problems and thinking abstractly.
It is worth differentiating two related but seemingly different concepts, or what Prof. Thomas W. Malone (Director, MIT Center for Collective Intelligence) refers to as general intelligence and specialized intelligence.
General intelligence is the ability to achieve a wide range of different goals effectively in different environments while specialized intelligence is the ability to achieve specific goals effectively in a given environment.
Notice how the definition of general intelligence by Prof. Malone is similar to the one offered originally for intelligence. On the other hand, specialized intelligence refers to the notion that an individual may acquire very specific abilities to succeed in very specific tasks.
In his book Superminds: The Surprising Power of People and Computers Thinking Together, Prof. Malone proposes that:
Collective intelligence is the result of groups of individuals acting together in ways that seem intelligent.
The book goes through extensive research conducted by the MIT CCI, which concluded — among other things — that:
- It is possible to measure the intelligence of groups.
- The measured intelligence of groups is a predictor of group performance in more complex tasks as well as in real-life situations.
- Like with individuals, groups with higher collective intelligence can learn faster.
- The average and maximum intelligence of group members are correlated with the group’s collective intelligence, but this correlation is only moderately strong.
There are three factors significantly correlated to the group’s collective intelligence:
1. The average social perceptiveness of group members. Social perceptiveness can be seen as a measure of a person’s social intelligence or the capacity to read emotions in others. The more members with high social perceptiveness, the more intelligent the group is.
2. The degree to which group members participated about equally in conversation. When few people dominate conversations the group tends to be less intelligent.
3. The proportion of women in the group. Groups with a higher proportion of women tend to be more intelligent. It has been established that women — on average — score higher than men in social perceptiveness.
From the three factors above, the only factor that was statistically significant while trying to predict collective intelligence was social perceptiveness.
This may show that the underlying mechanism in the two remaining factors may be social perceptiveness (people scoring higher in social perceptiveness may be more inclined to respect turns at speaking and women tend to score higher in social perceptiveness).
The social perceptiveness factor was a significant predictor of collective intelligence both in groups working face to face as well as in groups working remotely, communicating via text messages.
If there is one key takeaway from the work of the MIT Center for Collective Intelligence is that current science offers a clear roadmap for organizing our group of traders and developers so that the group’s collective intelligence may be maximized.
What’s Next?
We’ve explored the natural drive of humans to come together, the status quo within investment companies, and some of the essentials of intelligent teamwork…
In a follow-up piece, I will explore the next set of questions that should point us in the right direction to understand what an effective trading collaboration would look like:
What kind of incentives are required for group members to contribute their knowledge to the common cause?
How may contributions to the common cause be measured and valued?
How may greed be turned into a positive driver for collaboration?
An Ongoing Experiment
I’m a co-founder at the Superalgos Project, an open-source project building a Collective Trading Intelligence — or, as the project vision states — “a superior trading intelligence evolving within an open global collaboration of humans and machines, at the service of all people”.
After two years of exploration and prototyping, the project is finally taking the first step towards building such collaboration by releasing an early-stage version of the Superalgos software.
The alpha-stage piece of open-source software helps traders with virtually zero coding skills to develop and automate their trading systems via a graphic user interface (GUI).
Traders may build strategies and test their performance in real-time through a simulation engine with historical and up-to-date market data. Once happy with the simulation output, traders may deploy their strategies as fully automated trading bots, and trade live.
The platform implements the Superalgos Protocol, a powerful trading protocol that enables the standardization of the description of trading strategies.
Because all strategies are based on the same protocol, traders may seamlessly share and exchange strategies or even specific parts or components, enabling collaboration among trading partners, groups of friends and even large communities — a first step in the direction of building a Collective Trading Intelligence.
Preaching by example, the project has also released two open-source, fully-automated trading strategies: the Bull-run Rider and the Weak-hands buster, both of which may be found on Github.
The live, open experiment is just starting. You may join the community to use its existing trading intelligence free of charge, engage with other users and talk to the innovators behind the project.
Disclaimer: I am not a financial advisor and this is not financial advice. I share strategies and my own trading experiences as a means to disseminate knowledge. You are responsible for whatever you choose to do with the information I share.
Disclosure: The author is a core team member of the Superalgos Project.