Trading insights for 2018
A useful tool for a trader is perspective, and perspective can be increased by changing the perception of the trader’s reality. With the help of a friend I was recently able to gain new insights on trading and cryptocurrencies which I will share here.
Part 1: Trading, Gambling, and Fallacies
Recently I’ve been exposed to the argument of how many trading ideas are actually similar to gambling, in the sense that they are riddled with gambler’s fallacies, misconceptions, and deliberate lies.
At its core there is a fallacy: the ability to derive a complex multifactorial prediction engine based on a single factor such as price. It is simply impossible to just look at past price and be able to determine future price with any meaningful accuracy.
Let’s take an example where we combine weather prediction and the analogy for trading. It is just as impossible to predict the weather accurately in more than a few weeks (future price) just by looking at the temperature (price action), while ignoring solar radiation, clouds, wind, humidity, etc (demand, supply, news, new exchanges, margin rates, etc) which all affects the weather (price). Correlation does not equal causation, nor does it provide predictive ability, and just drawing lines in the past price action fails to account for any of the factors that led to the price action in the first place (even if price itself was one of those factors).
Traders swearing by technical analysis are plagued by the retrodiction fallacy, or postdiction.
One can build a trading model based on past price action, but considering the overwhelming amount of data required, it would take a supercomputer so massive it is simply not achievable in the foreseable future. This is also why centrally planned economies tend to fail, since they run into the economic calculation problem.
One does have at least 1/3 of a chance of guessing the market is either going up, down, or sideways, and the lower the resolution of the prediction the greater the margin of error allowed, which is why it could seem that many traders are consistent in their profits.
There are other factors at play as well; Survivorship bias is rampant in the cryptocurrency community, since stories of those that made big winnings or have an interest in promoting their winnings to sell a paid group membership, a course, or a book, get upvoted, shared and spread while stories of those that lost big amounts are less likely to be equally shared by the nature of the negative emotions associated with it.
I think in the end the main argument that should convince you is that hedge funds, a multi-million dollar industry with considerably more resources than any single trader with a simple TA approach, deliver less average returns by trying to predict the market than what the market actually moves up by. See:
Hedge funds delivered an average return of 5.6 per cent in 2016, falling well short of the 11 per cent gain of the S&P 500 last year, according to figures from HFR, the data provider.
So what actually can you trade with? It is often said that fundamentals are key to trading, but access to fundamental information is rare unles one is part of the organizations or companies in question. By the time fundamental news reaches the market and reaches any average trader, the window of action is limited, which increases the risk and limits the reward in a potential trade. In this sense, only those with insider information stand to gain from fundamentals-based trading. This is also why there are many cases of insider trading in cryptocurrencies markets, as well as false rumours to manipulate emotions. If the trader has a sufficiently large position and is looking to trade in a ridiculously illiquid asset, then he will become the market maker (whale), which allows him to manipulate price action in a certain direction. I think, however, that the majority of traders are not in this position.
In this sense, most traders are actually gambling, and each time they take a trade their chance of being right is only 33.33%.
Part 2: The paradox of a deflationary currency
I would like to make an argument that most cryptocurrencies, especially bitcoin but also all limited supply cryptocurrencies, are not actually currencies but rather stores of value, like gold.
First I would define wealth as: tradeable goods and services in the economy.
If we were to increase the quantity of currency (inflation), you would have more currency chasing the same amount of wealth, and the overall value would flow from currency to wealth.
However the opposite (increasing the amount of wealth, or deflation) would merely switch the flow direction from wealth to currency. See rent-seeking.
There is no incentive to spend bitcoins over dollars since bitcoins are limited in supply. However the amount of wealth in the world is expected to increase. By having incentives in place to encourage people to hold, this leads to less money available in the economy for producing wealth since everyone is expecting to get richer simply by holding, leading to a deflationary spiral.
Bitcoin was built as a reaction to inflation, and we know inflation has its problems as well. I believe the solution will be to develop dynamic inflationary cryptocurrencies, or cryptocurrencies whose supply somehow matches real live demand without trusting a third party.
Part 3: Bitcoin Bubble & Systemic Risks
There is an argument to be made that bitcoin behaves as it if was almost intentionally designed to act as a speculative asset bubble. The underlying cause of this is the limited supply and the mining difficulty adjustment, both affecting the supply inelasticity.
When miners mine bitcoin, there is a feedback mechanism which adjusts the amount of computation required. If many miners are attempting to mine bitcoin, the difficulty of the computation increases. If less miners are involved, the difficulty decreases, encouraging more to mine. The goal of the software is to ensure a steady production of blocks containing transactions.
However, supply and demand are connected in the real world, and the fact that there is no real connection between the demand for bitcoin and the rate at which new bitcoin is released merely exacerbates the problem. Since there is a fixed decreasing supply rate, the demand for bitcoin needs to decrease at the same rate otherwise bitcoin is undersupplied. This means theprice has to go up relative to demand from speculators, which becomes exponentially less sustainable the longer it goes on. At some point in any bitcoin bubble cycle, the demand from speculators will not be able to increase exponentially while the supply continues to decrease exponentially, trigerring the bubble burst. Even if bitcoin somehow overcame all its current scaling issues and replaced the USD, this would merely delay the problem.
From a trading perspective this event would likely create a big price contraction which would slow down investment in both bitcoin and other cryptocurrencies for an extended period of time as confidence decreased. However, there is another event which could equally affect overall cryptocurrency prices: the potential fractional nature of tether cash reserves.
Tether (sister company of the Bitfinex exchange) is a central party which issues blockchain currencies known as tethers, which in principle should exist as a ratio of 1:1 to real USD. The way tether operates, however, makes it hard to distinguish if real money is fueling the new issuing of tethers, or if its acting as a fractional reserve, or if the whole set-up is merely an elaborate money laundering scheme. Since no reasonable transparent audits have been performed to reassure investors, and since tether is accepted in many exchanges and tradeable for bitcoin and other currencies, which can themselves be traded for any possible pair, this creates a systemic risk of the existing price valuations for all crypto assets to be overvalued, and at risk of a severe pullback, potentially 70–90% of their peak value.
See the following articles for more information on the subject:
On an additional note, it’s worth mentioning to traders that market caps and price itself are a misleading way to value investment into the cryptocurrency industry. Market caps are calculated by multiplying the latest traded price with the existing supply. If I made an asset with only 1 000 000 units and I sold the first one to someone by $100, the market cap would read $100 000 000, but that of course doesn’t mean that $100 000 000 exist in the asset, only that $100 000 000 could potentially exist if the supply/demand was equal (hint: it never is).
Similarly, price itself is a dynamic value. The asset is only worth $100 for me if I can sell all my assets at that price. But if there is no one to buy them at $100, I am unable to do so, and must lower the price to meet demand.
Part 4: Future cryptocurrencies
Despite all of the above problems, I still think cryptocurrencies hold an underlying promise, since there is real value in having a digital currency without a central issuing authority. However, I think we are still far away from this both in terms of the technology as well in terms of the consensus model and distribution model.
Bitcoin and most cryptocurrencies have had three basic power players locked in a game theory consensus mechanism to make the protocol work. These are, of course, the miners, the developers, and the users.
These players struggle for power and its balance seems too delicate to be able to survive a split given the open-source nature of bitcoin. One could argue this split has already happened, in the form of bitcoin cash, a form of bitcoin which changes the balance in all three of the player groups: the currency is more usable for users who desire to transact in smaller payments due to reduced fees, miner operations are currently more centralized, and code development is also currently less distributed.
I would argue all current cryptocurrencies except a few are still first generation, since they rely on the basic blockchain technology and consensus mechanism between users, developers and miners of some sort. We have glimpses of the beginning of second generation cryptocurrencies with the examples of RaiBlocks, which uses a block-lattice structure, enabling a significant scaling increase, as well as Decred, which adopts a peer-to-peer governance model enabling users and developers to form a distributed consensus on future code changes. (Disclaimer: At the date of posting, I still own some RaiBlocks).
The very nature of development, unfortunately, requires that current generation of cryptocurrencies fail to scale or to achieve their goal in order for failures to become apparent and solutions to be developed. This means that bitcoin will fail to scale, so will ethereum, so will raiblocks, so will decred, so will other existing ones, prompting something new to come.
One could argue that existing coins can simply change their code to meet the problems as they arrive, however I believe this is wrong due to the following three assumptions:
1) Speed of development will never exceed speed of adoption, as at some point adoption will catch up to underlying limitations in code;
2) It’s impossible to design for all knowable and unknowable limitations a priori and;
3)As soon as limitations are reached in a cryptocurrency project, there’s an incentive for users to switch from that project to a more scalable one, rather than to wait for changes to be implemented.
Furthermore, the very nature of decentralized consensus and protocol development limits how much code can be changed without pushback. As an example, consider the following:
Anything about $XRP can be changed at the whims of Ripple the company. Anything about $IOTA can be changed by IOTA Foundation. None of their properties, whatever they may be, are real. They are not rigid, because they do not rest on a decentralized foundation.
While easy to make changes in centralized cryptocurrencies, the opposite is true for decentralized ones. For example, if there was some fundamental need to change the supply of bitcoin, the pushback from all player groups would likely produce a lot of forks.
Even with more scalable distributed technology and a more distributed consensus mechanism, there still remains the problem of distribution.
If you’ll recall, one of the reasons bitcoin was created in the first place was to prevent the control of new supply to be in the hands of a few. If new cryptocurrencies are developed, but the distribution ends up being terribly inefficient, then a few key players end up controlling the currency, thus opening it for manipulation and perpetuating the original problem that cryptocurrencies were meant to solve. The overwhelming majority of cryptocurrencies if not all currently have terrible gini coefficients (measure of inequality in distribution) when compared to the USD. There needs to be some way of connecting the users of the currency via identity without requiring trust in a third party. I think something like a second or third generation Civic-like project could potentially fulfill this role.
According to the arguments laid above, it seems clear only the equivalent of third generation cryptocurrencies would be able to fulfill the promise of a proper decentralized scalable cryptocurrency with a fair distribution. And a proper currency will not make money for its holders, since it would require a dynamic inflation rate in order to balance supply and demand.
If you found this article useful or interesting, consider donating ethereum to the following address:
0x1b474dcfa25109044e54De9af5CbDEBd9aB78eF5
