A Love Letter: Insights for Manual Cryptocurrency Trading
This is part of Kollider’s “Long Story Short” series, where we share short and sweet thoughts and tips with our community. It is…
This is part of Kollider’s “Long Story Short” series, where we share short and sweet thoughts and tips with our community. It is educational and is not financial advice.
To date, we’ve talked quite a bit about different strategies that work for both humans and bots. But, in our recent Long Story Short, we focused only on algorithmic trading and the process of backtesting.
Today, we’ll give a little love to our manual traders and share some things that we learned during our careers in the markets. Hopefully, these will give your trading hobby or career a boost. This is especially for those who are just starting out.
You may never get started if you try to learn everything you possibly can about a certain product, but some foundational knowledge does go a long way even at the early stages. So how do you take action while still learning?
One way to do this organically is to focus on less complex products first. Delta one products like spot, futures, and perpetual swaps theoretically have simpler price dynamics. If you find yourself learning “the greeks”, you might already be encroaching on more advanced territories.
Let Time Give You Structure
Time is crucial for a trader. It impacts many aspects of trading, such as execution speed, data latencies, frequency, hold times, etc. It is an important variable to consider when analysing any market, which is why the “time frame” at which you view it is particularly helpful, especially for beginners.
When analysing a market, it is easy to lose yourself if you don’t hold time constant. By holding it constant, you can derive specific types of useful information. For example, let’s say that you choose to look at a market in 1-hour intervals. You might then look at the following pieces of information:
- Volume. What are the different amounts of total volume that go through each interval? How did these change over the last 2 weeks? How much volume in that 1-hour window is comparatively low volume? What about medium? High? Are there certain volume patterns in past hours that tend to correlate with their subsequent hourly volumes?
- Prices. How does the price change from 1 hour to the next? What is a small movement? What is a big one? Does low volume show a different pattern of price changes from a larger volume? Are there certain price patterns in past hours that tend to correlate with their subsequent hourly price changes?
- Seasonality. Does anything repeat over and over? Do they tend to repeat the same time of day or week? How much does the repetition hold? Is it fading?
- Pattern Changes. If you look at another period of 2 weeks, say one from 3 months ago, do the 1-hour intervals show different patterns in volume, prices, seasonality, or anomaly?
- Anomalies. Are there times when the patterns that often repeat with a high probability just break? Can you find other pieces of information outside of prices that might explain that (e.g., global events, some piece of news)? Could those anomalies give you insight into external factors that may affect your product?
It helps to keep to a single time window during an analysis. And you can certainly repeat the analysis for different intervals to see how those variables above will play out differently.
Balance Your Expected Gains and Losses
The time analyses above provide you with many benefits. Let’s highlight an important one. Knowing the time interval you want to trade and how the prices move within that interval will help you figure out when to cut your losses.
For example, say that you’re looking at hourly windows. You discover that you can reasonably predict up to 3 hours ahead. You also determine that the price moves an average of less than 1%. In this scenario, you’ll likely want to cut bad positions before they reach a 1% loss. After all, why would you take losses that are bigger than the usual 1% gains that you’re able to predict? But was 1% loss the right threshold for getting rid of them?
Consider this open-ended exercise: imagine that you find a price pattern that you believe could bring you a 1% winner three out of ten times. At what loss percentage should you cut your positions the other seven times? And let’s even throw a wrench in there: do the expected winning trades tend to go through that loss percentage first before becoming a winner (i.e., would you be able to tell the difference between a position that will become a winner or one that you should exit early?).
Think In Maybes, Not Definitelies
Aside from understanding the product and structure that you’re trading, a lot of the practice is about discovering events or conditions that correlate to price movement and being able to act on it.
But the markets are dynamic. They move due to the aggregate behaviours of all their participants, and those themselves could change. What patterns you see today may not exist tomorrow. This month’s causation may barely look like correlation the following month. If you hear someone say a market has to move in that direction when their fast moving average price crosses a slower one, be a little cautious. It may have been true when they observed it before, but it might not be true now.
And that’s OK!
We don’t know for sure that event A will always cause B or to what extent, but we do know that B tends to follow A right now. But we can call it a maybe-causal relationship and just simply think that B could happen after A. Our task then is to find as many of these “maybe-causal relationships” and estimate our market’s next moves from them.
Balance Probability With Action
But even though you’re stacking those “maybe-causal relationships” and increasing your chance of being correct, you can’t often wait for absolute certainty. The higher the probability of a move, the more likely other traders will be thinking the same. And there may only be so much volume or liquidity that that pattern can absorb.
Although it doesn’t make sense to just act with 0% probability, it is also tricky to wait until it’s closer to 100% (for one, it will be very rare for it to even get near-certainty), so you need to find the right balance for your risk tolerance. Hint: when you’re getting better at guessing higher probability moves, you can tailor the size of your orders to start reflecting that.
Trade for as Long as Possible
There’s a temptation to “go big or go home“ when you’re just starting out. There are a lot of tales of YOLO traders making huge profits after taking bold risks. For each success story you hear, know that there are probably dozens of others that went big and went broke. Consider that some of the best ones that you hear about have probably spent many years honing their craft before taking larger positions.
Trading is a skill of pattern recognition and capture. To become better, you need time to view correlated events in the market and their probability of occurring. So, how do you give yourself more time for improvement? There are two ways that we use.
First, trade smaller and only increase your sizes as you level up. Many crypto exchanges will let you trade very small positions. Kollider, for example, currently allows $1 bitcoin perpetual swap order sizes. If you find venues that allow you to “paper trade” or simulate trading, try those.
Second, “backtest” your manual trade ideas. We talked about how backtesting is helpful for bots, but you can apply the same practices manually. You can mentally and visually validate your trade ideas also by going through old charts and discerning how they would have played out historically. Like we described earlier, it helps to isolate this analysis to a certain time interval.
Embrace What Makes Crypto Markets Special
As most of our readers are cryptocurrency fans, here’s one that’s specific to crypto products. In addition to allowing you to trade small sizes and get granular market data easily, crypto markets are also volatile and fragmented. This can actually be a great thing for traders. It opens up a variety of strategies and time horizons that would not otherwise be available. We talked about the spread trade before, and those tend to show up as more fragmentation shows up.
Be Methodical and Improve
In our article on human vs bot trading, we laid out some steps for manual traders to loop through:
- Check relevant information.
- Predict the future direction.
- Look at their existing positions.
You should use the “repeat” step above to improve what you did in the last loop. Set smaller, achievable goals and keep iterating.
Lastly, if you take anything away from this article, it’s that trading can be fun, but also be challenging. It’s your job to figure out the quirks of a new market. So, get started when you’re feeling ready, but be patient. Give yourself time to enjoy and learn.
Know anyone else interested in learning about bitcoin, Lightning, and trading? Follow us on Twitter and help us share the word.