Bitcoin trading performance evaluation: A comparative study of momentum and mean reversion strategies from 2020 to 2025
Ekström, Marja (2025)
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025112930915
https://urn.fi/URN:NBN:fi:amk-2025112930915
Tiivistelmä
This thesis studies the performance of two systematic Bitcoin trading strategies: momentum and mean reversion. The focus is on two widely used technical analysis tools — the moving average crossover strategy (momentum based) and the Bollinger Bands strategy (mean reversion based). Using high frequency Bitcoin price data, this thesis evaluates each approach relative to one another and to a buy-and-hold benchmark. This study also includes transaction fee modelling to estimate actual profitability of active strategies.
Annual returns, Sharpe ratio, volatility, profit factor, and maximum drawdown are used to assess the performance of different strategies, and to make the comparison of profitability and risks. The findings show that Bitcoin has mean-reverting behaviour on hourly data analysis, and the Bollinger-based strategy outperforms both the momentum strategy and passive HODLing both in terms of return and risk-weighted performance. The momentum strategy shows moderate return potential, while buy-and-hold remains effective over long term but is not optimal in high-frequency environments.
These results suggest that the effectiveness of a trading strategy strongly depends on time resolution, whether hourly or daily data is used to analyse the performance. This helps researchers to better understand how cryptocurrency markets behave and it gives useful guidance for traders and investors to choose suitable trading strategies.
Annual returns, Sharpe ratio, volatility, profit factor, and maximum drawdown are used to assess the performance of different strategies, and to make the comparison of profitability and risks. The findings show that Bitcoin has mean-reverting behaviour on hourly data analysis, and the Bollinger-based strategy outperforms both the momentum strategy and passive HODLing both in terms of return and risk-weighted performance. The momentum strategy shows moderate return potential, while buy-and-hold remains effective over long term but is not optimal in high-frequency environments.
These results suggest that the effectiveness of a trading strategy strongly depends on time resolution, whether hourly or daily data is used to analyse the performance. This helps researchers to better understand how cryptocurrency markets behave and it gives useful guidance for traders and investors to choose suitable trading strategies.