How a simple momentum strategy can outperform a buy and hold cryptocurrency portfolio

Alexander Thomsen
3 min readDec 20, 2017

--

Summary

A simple momentum strategy with monthly rebalancing for cryptocurrencies achieves superior returns compared to CRIX index⁴. In this current market environment, it’s becoming increasingly important to have a diversified and agile cryptocurrency portfolio, in order to increase the probability for higher returns and stability (if that is your goal). If you don’t have the time or experience to build a well-thought out, rule-based portfolio, check out Cryptovesta, which provides cryptocurrency portfolio advice.

Full Version

Cryptocurrency is all the rage right now around the world, with prices at an all time high. It’s therefore more important than ever to hold a portfolio of cryptocurrencies rather than put all your eggs in one basket. In addition, I would recommend, especially when we consider current market conditions, that a buy and hold strategy would underperform a simple momentum strategy that buy/sells cryptocurrencies on a monthly basis. Before I share this simple strategy, first have a look below, how the top 10 coins by market cap has changed over the years. The cryptocurrency market is still relatively new, which means if you decide to create a portfolio based on top coins by marketcap this could reduce your probability for higher returns and stability.

December 2013¹
December 2015²
December 2017³

Introduction to strategy

The strategy uses a simple momentum approach. For people who are not familiar with momentum, it comes in many variants using Relative Strength index, Bollinger Band etc. In this example, all we will do is to measure the difference between current price versus a price in the past (e.g 10, 15, 30 days ago).

Example using 15 days comparison:

Current price: $500

Price 15 days ago: $450

Momentum: 450/500 = 0.9

Walk-through using excel

In excel, I downloaded historical data for 57 coins and calculated 2 momentum's, Mom_1 (15 days difference) and Mom_2 (30 days difference) and then gave 50% weight to each momentum (50% Mom_1 & 50% Mom_2), which gives me a daily momentum for each coin. I then rank each coin against each other, while coins with higher momentum received better rank. I then create a condition, that my model will only choose coins with a certain rank threshold (e.g top 5). The model buy and sell coins beginning of every month.

Evaluation

I backtested this strategy from December 2015 to December 2017. For convenience, I didn’t take into account commission fee, but as it’s monthly rebalancing, it shouldn’t make a large impact. Furthermore, I didn’t try to optimize the strategy, for example we could have tested which duration is better (15, 30, 45 days etc.) or what would be a better rank threshold (3,5,7 etc.). Please see below the results.

Backtesting result versus benchmark CRIX⁴

This model’s backtesting should be taken with a grain of salt. More importantly my goal is to really show that a basic rule-based strategy can achieve superior returns compared to a buy and hold strategy. If you don’t have the time or experience managing such a strategy yourself, I would highly recommend you to check out Cryptovesta, which will provide you with cryptocurrency portfolio advice.

--

--