algorithmic trading / backtesting / strategy evaluation
What is Backtesting? How to Test Your Trading Strategy
Course walkthrough
Algorithmic Trading Fundamentals
All lessons in this series
- 1What is Algorithmic Trading? A Beginner's Guide to Automated Strategies
- 2What is a Trading Indicator? The Building Blocks of Algorithmic Trading
- 3Technical Analysis Indicators Explained: Mean Reversion vs Momentum
- 4What Are Fundamental Indicators? Understanding a Company's Financial Health
- 5Alternative Data Sources for Trading: Satellite Imagery, Sentiment Analysis & More
- 6From Indicators to Conditions: Building Your Trading Rules
- 7What is a Trading Condition? Turning Indicators into True/False Logic
- 8What Are Trading Actions? The Final Building Block of Algo Trading
- 9What is a Trading Strategy? Indicators + Conditions + Actions
- 10How to Know if Your Trading Strategy is Actually Good?
- 11What is Backtesting? How to Test Your Trading Strategy
- 12What is Trading Strategy Optimization? (And How to Avoid Overfitting)
- 13Recap: From Building Blocks to Deployment
- 14How to Deploy a Trading Strategy: Paper Trading to Live Trading
- 15How to Backtest a Trading Strategy on NexusTrade (Beginner Guide)
A strategy might sound logical, but a backtest tells you how it actually performed.
Backtesting is a historical replay of your trading rules over a specific time period. It's one of the most important parts of evaluating a strategy - without it, you're just guessing.
But here's what most people get wrong: a backtest is not a prediction of the future.
In this video, I cover: → What backtesting actually is → Why it's essential for evaluating strategies → The non-stationarity problem: what works this year might fail next year → Bull vs. bear market performance - why robustness matters → Hindsight bias: you already knew AI stocks would blow up
A good backtest doesn't guarantee future success. But understanding its limitations makes you a smarter trader.
Watch lessons, complete hands-on activities, pass quizzes—and by the end, you'll have a real strategy deployed to the market.
Transcript
0:00Now that we know what makes a trading strategy good and what makes a trading strategy bad, let's talk about how we can test a trading strategy to make that determination. This is a process called backtesting.
0:13A backtest is simply a historical replay of how a set of trading rules performed during a specific period in the past.
0:25Backtesting is extremely powerful because it gives us insight on how good these trading strategies actually are.
0:33You know, at a high level, a trading strategy might sound good, it might sound logical, and it might sound reasonable, but a backtest tells you how that trading strategy actually performed.
0:46Without a backtest, you don't really know if you have a good or bad trading strategy. It is one of the most important parts of the testing and evaluation phases.
0:57Now, when it comes to backtests, people have this flawed misconception that it's a perfect simulation of what's definitely gonna happen in the future, and this couldn't be further from the case.
1:11Like I said in the beginning of this module, stock market returns are non-stationary. They change over time. So what makes a backtest good now might not be a good backtest in the future.
1:24Or put differently, a trading strategy is good this year, but it might be terrible next year.
1:31A trading strategy might be extremely good when the market is overall bullish. It might significantly outperform the broader market in every single way.
1:42But then the next year, some economic uncertainty happens and the market is in a bear market. That same trading strategy might be catastrophic.
1:54It might lose 70% or 80% of its value while the broader market, the S&P 500, only lost 10% to 20%.
2:03So when you're creating a trading strategy and when you're backtesting it, you want to make sure that the strategy is robust, that it's not fragile and doesn't fall apart in different market conditions.
2:18Finally, when you create a trading strategy, you have to be aware that it's a historical simulation which is inherently biased.
2:28You know, you might create a trading strategy using AI stocks with the highest revenue.
2:34And you see NVIDIA, Google, and Microsoft are on that list, and you conclude that you should always have a trading strategy that trades these AI stocks. But your perception is biased because you already knew that would happen.
2:50You know, you already knew that these AI stocks blew up. So that's why you wanted to build an AI trading strategy.
2:58But next year might not be AI. Next year, it might be nuclear energy, or green energy, or bank stocks. You know, it can be any number of things that change because the overall market conditions change.
3:13You need to be aware that a backtest is biased.
Join the conversation
Loading conversation…