2025 TRADING RESULTS
And so as the year is coming to an end and I’ve dedicated a full year to running my algorithmic trading strategies it is only natural to show the results for the year!
In general it has been a very bad year for crypto with the altcoin space entering a year long bear market aside from Bitcoin itself ever since the Presidential Election results of late 2024. Managing my algo portfolio has been more painful and nerve wrecking than expected simply due to the unforgiving nature of the crypto markets.
What I did right:
1) I used proven strategies that existed long before my time and improvised on them
2) I reduced the use of leverage the moment I felt my PNL taking too much of a swing and variance
3) I was super strict with my choice of strategies and backtesting to avoid overfitting, and accounted for real world friction such as slippage and fees
4) Testing with code: My backtest were often batches of excel data + some Python backtesting, often stress testing with models such as Monte Carlo
5) I automated and reduced interferences with live-trades as much as possible
All these led to the PNL result in this post :)
Now, the things I did wrong…
1) Chose the wrong market to trade in, crypto was awful in 2025 🥲 If I could rewind time I’d apply my strategies with the Mag 7, Gold or something else. Definitely not altcoins 🤢
2) Not trusting my strategy edge fully - I often doubted the validity of my entry condition and sometimes I did interfere with the trade simply because of the lack of trust of my strategies. As I look back, I’d have been better off keeping my itchy fingers away 😅
More or less these are the mistakes I made in my journey in algorithmic trading.
I’m already running new strategies on walk forward mode for 2026. Ready to deploy on 2nd Jan 2026, a new year a new start - whatever the markets bring, game on!
Reflecting on my 2025 crypto quantitative trading journey, one insight I gained is the critical importance of market selection. While I initially focused on altcoins, the persistent bear market revealed that diversifying into more stable assets like Bitcoin, gold, or major tech stocks (Mag 7) would have been wiser. This is something I plan to explore deeply in 2026. Another challenge was the psychological aspect of trusting algorithms. Often, I interfered with trades due to doubt, which impacted my performance. It's vital for anyone involved in algorithmic trading to cultivate confidence in their backtested strategies and resist impulsive decisions. Moreover, the application of backtesting techniques like Monte Carlo simulations and stress testing helped me understand potential variances and risks better. These tools are invaluable in refining your trading system and accounting for real-world issues like slippage and transaction costs. For traders interested in crypto quantitative trading, I recommend focusing on comprehensive backtesting, limiting leverage use, and maintaining strict discipline around strategy execution. Automating your trades as much as possible reduces emotional bias and improves consistency. Looking ahead, adopting a walk-forward approach to adapt strategies to live data is essential in staying ahead of market changes. My plan for 2026 includes deploying improved, tested strategies while managing expectations about market volatility. Sharing this experience emphasizes that success in crypto quant trading requires patience, adaptability, and continuous learning. Remember, no strategy is foolproof, but with the right mix of tools and mindset, profitable trading is achievable over time.