Bitcoin RSI

Among many different indicators, that can be applied to describe and predict the price of Bitcoin and other cryptocurrencies, the “Relative Strength Index” (RSI) is one of the most popular indicators. It displays the price action in a fixed range and offers a clear classification, of whether the price of an asset is in oversold or overbought range. However, as with all indicators it is questionable, if the shown data can really be used for price forecasts or if the alleged impacts cannot be proven statistically. Therefore, I will take a look at the RSI indicator in this article and analyze its predictive impact on Bitcoin’s daily price.

Let’s start with a description of the RSI. The “Relative Strength Index” (RSI) is a momentum indicator used in technical analysis, that measures the magnitude of price movements on a scale of 0-100.¹ Its main purpose is to identify whether the price of an asset is in overbought, neutral or oversold range.² An RSI value below 30 is commonly perceived as oversold, indicating a future uptrend, whereas an RSI value above 70 is perceived as overbought, indicating an upcoming downtrend.³ In general, the lower the RSI value, the higher the chance for the price bouncing upwards and the higher the RSI value, the higher the chance for the price moving downwards. However, it is not guaranteed that the price will move in the direction indicated by the RSI.

So much for the theory, now onto the practical application. As of Spring 2020, there are around 9.7 years of public trading data available for Bitcoin against the U.S. Dollar. On 58.5% of the days in that time span, Bitcoin has had a daily RSI value of above 50, while the daily RSI value has stayed below 50 for 41.5% of the time. Considering overbought and oversold range, Bitcoin has spent way more time in the overbought area, than in the oversold zone. On 18.4% of the days, Bitcoin’s daily RSI was above 70, which is generally perceived as overbought. On 8.4% of the days, its daily RSI was above 80 and on 1.8% of the days it even had an RSI above 90. Considering oversold range, Bitcoin’s daily RSI was on just 4.2% of the days below 30. A daily RSI below 20 was even rarer with just 0.8% of the days, while the daily RSI moved below 10 on just one single day.

Let’s have a closer look at how profitable certain trading strategies were in the past. One simple strategy would have been to buy Bitcoin at the first daily close below a certain RSI value and to sell it again at the first daily close above a certain higher RSI value. I identified five different occasions, where this would have been profitable on average, although the sample sizes are rather small – each pattern occurred between 10 and 30 times in the last 10 years. The five patterns and their average returns are listed in the graphic below:

We can also reverse the first trading strategy and sell Bitcoin at a certain RSI value, to buy it back when it has reached a lower RSI value. However, in contrast to trading strategy number 1, this didn’t play out as expected. If one would have sold Bitcoin at a daily closing RSI value of 70, 80 or even 90, with the goal to buy it back once it has reached a lower RSI value, it would have resulted in a loss compared to the amount of sold Bitcoin. Due to its incredibly bullish price action, Bitcoin has reached very high RSI values in the past, without necessarily slowing down at the first time it reached those values.

Let’s say our goal is to sell at a daily close in the oversold zone above 70 and buy back in the neutral area at 50. With a buy back at the first daily closing RSI below 50, a sell at the first daily closing RSI above 70 would have resulted in a price increase of 95%, while a sell at the first daily closing RSI above 80 would have resulted in an average price increase of 112% and a sell at the first daily closing above 90 would have resulted in a price increase of 119%. As a conclusion, trading strategy 2 on average would have resulted in around 50% less Bitcoin at the point of the buy back compared to the selling point.

Looking back at the relative strength index during Bitcoin’s last ten years of public trading data, we can see that the RSI has spent way more time in overbought territory than in the oversold range. In addition to that, Bitcoin’s RSI value has stayed above 50 for around 60% of the time. Of the two discussed trading strategies, only the first one had been profitable on average, in which one would have bought Bitcoin at a lower RSI value to sell it again at a higher RSI value. Selling Bitcoin at a high RSI value and buying it back at a lower RSI wouldn’t have worked on average, since one would have only been able to buy back less Bitcoin than one has sold before.