Day Trading Cryptocurrency is all about the math behind the numbers. It's all about picking the right technical indicators and sticking to them. Do your research ahead of time, pick your indicators, then emotionlessly follow the numbers!
Welcome to another installment of 'Musings of Random Thoughts'! Before we get started today, a disclaimer:
This post is not financial advice, I am not a financial advisor. Please make your own decisions with your money, and do your own research before jumping into any investment.
Now that the fun stuff is out of the way - to the real stuff! Day trading cryptocurrency! This is an alternative approach to the standard of buying and holding crypto, or, HODLing, but it is becoming a viable strategy as these major cryptocurrency exchanges are gaining traction, which most importantly provides liquidity. Which brings us to the first important box to check before you day trade a coin. Pick one with decent liquidity. My rule of thumb is to prefer coins that are moving at least ~$1M a day from a liquidity standpoint, as this means there is enough interest in the coin to flip it.
Day trading is all about the technicals of the cryptocurrency market. We are attempting to make some money off of short term volatility in the markets - you know, buy low, sell high. Except, what is low, and what is high? Some math can help us decide. For my day trading, I use 3 specific indicators, to help me time the market cycles. I wish I had a product to share, but for now suffice it to say that I have written my own program capable of ingesting market data, which I pump to InfluxDB where I run all my calculations and create all my charts. The screenshot in this blog, and all the math being done is derived from InfluxDB. For each indicator, I'll explain how each one works, why it's good for day trading, and at the end, my complete strategy leveraging these 3 indicators. When I'm on it, this strategy has netted me 2% profit a day, which isn't a bad rate of return in the slightest!
Chande Momentum Oscillator - The Fast Indicator
The first indicator is the Chande Momentum Oscillator (CMO). The CMO is a faster moving momentum indicator, that fluctuates between -100 and 100. The closer to -100 this value is, the more oversold or undervalued an investment is compared to its recent historical price. The closer to 100 this value is, the more overbought or overvalued an investment is compared to its recent historical price. The points of interest for this is when the value moves below -50, it's an early indicator that we should buy. When this indicator crosses above 50, it's an early indicator that we should sell. This indicator is our early indicator since it is very sensitive to emerging trends, and thus fluctuates or oscillates quickly as conditions change. Trading off this indicator alone is ill-advised due to the sheer number of false positives it ends up declaring. To read more on this, check out the official Investopedia definition.
Relative Strength Index - The Slow Indicator
The second indicator is the Relative Strength Index (RSI). The RSI is a slower moving momentum indicator, that fluctuates between 0 and 100. The closer to 0 this value is, the more oversold or undervalued an investment is compared to its recent historical price. The closer to 100 this value is, the more overbought or overvalued an investment is compared to its recent historical price. The points of interest for is that when this value moves below 30, it's a late indicator for us that we should buy. When this indicator crosses above 70, it's a late indicator that we should sell. The reason that this is considered the late indicator is because it moves slower than the Chande Momentum Oscillator. To read more on this, check out the official Investopedia definition.
Kaufman's Efficiency Ratio - The Sanity Check Indicator
The last indicator is Kaufman's Efficiency Ratio (KER). The KER helps us decide what is noise and what is an actionable trend. There are varying ways to compute the KER, but the one in InfluxDB fluctuates between 0 and 1. This implementation is a direct test on the Chande Momentum Oscillator. The closer to 0 this is, the less correlation there is between the market and the Chande Momentum Oscillator. Meaning, this ratio would be near 0 if the CMO was saying an investment was oversold yet the actual price was trending up (a reverse correlation). This ratio is near 1 if the CMO says an investment is oversold and trending down (1:1 correlation). To read more on this, check out the official Tradingpedia definition, but do note that the number are slightly different than influx for this calculation.
How To Use These?
Our primary indicator is the Chande Momentum Oscillator, since we are trying to time short term momentum to buy low and sell high. The ideal world has us in a trade for less than a few hours before bouncing back to our base currency, and hopefully taking some profit. If the CMO is not under -50, we won't even consider buying. The other 2 indicators are sanity checks on the CMO. If the Relative Strength Index falls below 30, this suggests there has been some confirmed recent downward price action. This ensure we don't buy in before the trend has been established, by waiting until the RSI has also detected over sold conditions. Lastly, we use the Kaufman Efficiency Ratio as the stoplight 🚦. If the efficiency ratio is close to 1, it's our green light to trust the buy indicators. If this ratio is not, we should wait for a better opportunity to jump into the market.
Additionally, to use short term indicators, how they are configured is everything. All 3 of these indicators depend on a number of periods. For these example, we're running each calculation on a moving 7 period basis. The value of 1 period is equivalent to the average price over a 5 minute window, meaning that each point in our technical graphs is considering 35 minutes worth of data.
Times To Buy
The ORANGE arrows indicate the first time the indicators said we should buy. The CYAN arrows indicate the second time we should buy. The top chart is the actual price, with a few moving averages overlaid. The most sporadic line is the actual price. Notice how both times to buy checked all the boxes for the rules on when to buy - this is emotionless, technical buying. Let the data say when to buy, and do it. No guess-work involved. When it comes to selling, I could use the technical indicators for this as well, as you can see, the indicators also show selling recommendations, but I make selling emotionless as well. I compute the price I need to sell to earn 0.5% profit (after fees - don't forget the fees. Mine are 0.1% per trade), set that price, and won't touch it until we've sold. The goal is to get in and out rapidly, multiple times a day.
In this 6 hour view shown above, you can see that following this strategy, we would have made 1% profit ( 🎉 📈 ), even though the coin itself has fallen in value 📉. That's pretty awesome! Now, this strategy could fail on strongly trending markets over long durations (up or down), since the indicators would be stuck at buy or sell depending on which way the market is trending. We'd inevitably get in or out too soon and miss a run. Do your research and pick a coin that is more immune to hype. I have not yet tried this strategy with Bitcoin or Ethereum. The ideal coin is one that makes sine curves with its price (regular ups and downs), allowing us to buy low, sell high, and most importantly take profits along the way even if the 30d price of the coin is largely stable.
I hope this post has been useful for you, be sure to leave a tip or drop a like if I've given you some ideas for your own crypto day-trading adventure!
Just to reiterate, this post is not financial advice. It only attempts to explain the math behind an investment strategy that I personally am using. You should do your own research, and come up with your own strategies before risking real money. Consult a real financial advisor before doing any of this yourself. I am happy to discuss the math behind this post, but I am not a financial advisor, and this is not financial advice.
If you are interested in the actual Influx Flux queries used for the chart, just drop me a comment 👇 - I am happy to share these, they are just too dry for this blog post. 😜