Running a simulation
Before risking real capital, you should always test your trading strategy against historical market data. Tradecraft's backtesting simulator runs your strategy against actual past signals and price movements, showing you exactly how it would have performed.
Why backtest your strategies?
Validate your strategy works as expected
See real performance on historical data
Build confidence before going live
Identify issues in your trading logic
Compare different approaches to find what works best
What is a Backtest?
A backtest simulation:
Takes your strategy code
Replays the signals your strategy received over the time period you specify
Simulates buying and selling based on your strategy rules
Shows you the profit/loss you would have made
The simulator uses real historical prices and actual signals that your data sources sent, so results accurately reflect what would have happened if you had been trading live during that time period.
How to Run a Backtest
Step 1: Open the Development Tab
Navigate to your strategy and click the Development tab. You'll see:
Strategy Editor (left side) - Your strategy code, including the Run button to start a simulation
Start and End (right side) - Select the start and end simulation period in the signal table
Results (bottom) - The simulation results after running
Step 2: Run the Simulation
Click Run and wait for the results. The simulator will:
Load all your historical signals from the selected data sources
Run your strategy against each signal's price history
Calculate your profit/loss and performance metrics
Note: Large date ranges with many signals takes longer time to complete.
Understanding Your Results
After the simulation completes, you'll see several key metrics
Summary Metrics
Initial Wallet
The starting capital you configured (e.g., 5 SOL)
Total Remaining
How much SOL you have left in your wallet, plus the SOL value of the tokens held in the wallet but not yet sold
Unrealized
Value of tokens you're still holding at the last known price
These are open positions that haven't been sold yet
Trade Details
For each token traded, you'll see:
Token Address - Which token was traded
Entry Price - Price when you bought
Amount - The amount in SOL that was bought
Unrealized - The sum of the SOL value of all the tokens held in the wallet at that point and not yet sold.
Wallet - The SOL balance of the wallet at that point in time
Decision - Whether it's a buy or a sell
Price gain - The multiplier when selling, for instance a 100% price increase is a 2x.
Price - The USD price at the time of the buy or sell
Event time - When the buy or sell took place
Tips for Effective Backtesting
Start with Recent Data
When first testing a strategy:
Use the last 2-5 days of data
Verify your strategy logic works correctly
Check that you're seeing the trades you expect
Once validated, expand to longer time periods for more comprehensive testing.
Test Different Market Conditions
Markets behave differently at different times. Test your strategy on:
Trending markets - When prices are generally going up or down
Volatile periods - High price swings and rapid changes
Quiet periods - Low volatility with stable prices
A good strategy should work reasonably well in different conditions, not just one type of market.
Compare Trade Sizes
Try your backtest with different trade size settings:
Small fixed amounts (0.1 SOL) - Lower risk, steadier growth
Larger fixed amounts (0.5-1 SOL) - Higher risk, bigger wins and losses
Percentage-based (2-5%) - Dynamic sizing that scales with your wallet
See which approach gives you the best balance of profit and risk for your style.
Check Signal Volume
Look at how many signals your strategy received during the test period:
Too few signals (< 5 per day) - Your strategy or filters might be too restrictive
Too many signals (> 50 per day) - You may be overtrading, which increases costs in live trading
Review Individual Trades
Don't just look at overall profit - examine individual trades:
Which tokens performed well? - Look for patterns in successful trades
Which tokens lost money? - Understand what went wrong
Are you exiting too early? - Check maximum price vs. your exit price
Are you holding too long? - See if profits evaporated after peaks
Use these insights to refine your strategy's buy and sell logic.
What Backtesting Can and Cannot Tell You
Backtesting IS Good For:
✅ Validating your strategy logic works correctly ✅ Seeing how your strategy would have performed historically ✅ Comparing different strategy approaches ✅ Identifying obvious flaws or bugs in your code ✅ Building confidence in your trading approach
Backtesting CANNOT:
❌ Predict future performance with certainty ❌ Account for all real-world trading costs and slippage ❌ Guarantee you'll make the same returns live ❌ Replace careful live monitoring ❌ Tell you when market conditions will change
Golden Rule: Use backtesting as one tool in your strategy validation process, not as a guarantee of success.
Avoiding "Overfitting"
Overfitting means optimizing your strategy so specifically for historical data that it fails on new data.
Warning Signs:
Extremely high returns (>1000% ROI) that seem too good to be true
Strategy with very complex rules tailored to specific past events
Works great on training data but fails immediately when live
How to Avoid:
Keep your strategy simple - simpler is often better
Test on multiple different time periods
Don't over-optimize to squeeze out every last percentage point
If it seems too good to be true, it probably is
Last updated