Recently, I’ve been pondering a question: how many traders actually spend time verifying whether their trading strategies are reliable? I think most people don’t. That’s why backtesting is so important.

Simply put, backtesting is using historical data to test whether your trading ideas can really make money. It sounds straightforward, but there are many details to pay attention to in practice. I recently looked at a classic case using a Bitcoin 20-week moving average strategy—buy when the weekly price breaks above the 20-week moving average, sell when it falls below. Backtesting from 2019, this strategy generated five signals, starting with about $4,000, and selling at a high of $8,500. Sounds pretty good, right?

But here’s a key issue: making money in the past doesn’t guarantee making money in the future. Market conditions change, and the same strategy might become ineffective. So, the core of backtesting isn’t predicting the future, but helping you understand how the strategy performs under specific market conditions.

When doing backtesting, there are a few points that are often overlooked. First, consider trading costs, withdrawal fees, and other expenses. Many people only look at returns during testing, ignoring costs, and end up finding that the strategy isn’t profitable after all. Second, the choice of historical data is crucial. If the data doesn’t reflect current market conditions, the test results have little reference value. That’s also why some people’s backtesting results look perfect, but they lose money in actual trading.

I’ve noticed many fall into the “cherry-picking” trap—only selecting data segments that favor their hypotheses. Doing so completely undermines the meaning of backtesting. True validation should be done in real-time market environments, but without risking real money. This is called simulated trading or paper trading. Many mainstream trading platforms offer simulated environments where you can test strategies in real market conditions, but with virtual accounts.

Regarding how to perform backtesting, there are manual and automated methods. Manual means analyzing charts, data, and placing orders by hand. Automated involves using code (like Python) or dedicated backtesting software to execute trades. Many traders record test results in Excel or Google Sheets, including metrics like number of trades, profit trades, loss trades, Sharpe ratio, and maximum drawdown. A higher Sharpe ratio indicates better risk-adjusted returns. Maximum drawdown measures the largest decline from a peak to a trough, reflecting the worst-case loss.

Honestly, backtesting isn’t万能. It can only tell you how a strategy performed historically, not guarantee future success. But if you want to systematically optimize your trading methods, backtesting is an essential step. Many professional traders and quantitative analysts rely on this tool. The key is to correctly interpret backtesting results, avoid personal biases, and continue validating your ideas in real-time markets.