Henry Lin

Posted on Oct 2

Lesson 1: Introduction to Quantitative Trading and Freqtrade

#cryptocurrency #beginners #automation #tutorial

Lesson 1: Introduction to Quantitative Trading and Freqtrade

📚 Course Series: Complete Freqtrade Quantitative Trading Tutorial Course
📖 Part: Part 1 - Basic Introduction
⏱ Duration: 1.5 hours
🎯 Difficulty: ⭐ Beginner Level

🎯 Learning Objectives

After completing this lesson, you will be able to:

✅ Understand what quantitative trading is
✅ Know the advantages and risks of quantitative trading
✅ Recognize Freqtrade and its features
✅ Clarify your learning path and goals
✅ Prepare mentally for learning quantitative trading

📖 Part 1: What is Quantitative Trading?

1.1 Traditional Trading vs Quantitative Trading

Traditional Trading Method

Xiao Ming's Trading Day:

9:00 AM  - Check crypto prices, BTC down 5%, feeling panicked
10:00 AM - Friend says it will rise, immediately buy
12:00 PM - Up 2%, hesitate whether to sell
3:00 PM  - Dropped back, regret not selling
8:00 PM  - Can't stand it anymore, cut losses at -3%
Next day - BTC surges 10%, regret selling last night...

Problems:

😰 Emotional decisions (fear, greed)
⏰ Tiring to watch the screen (24-hour market)
🎲 High luck component (based on feelings)
📉 Often buy high and sell low

Quantitative Trading Method

A Trading Bot's Day:

while True:
    current_price = get_real_time_price()

    if EMA20_crosses_above_EMA50 and price_above_EMA20:
        buy(100 USDT)
        print("✅ Buy signal triggered")

    if profit > 5% or loss > 2%:
        sell()
        print("✅ Sell signal triggered")

    sleep(5_minutes)

Characteristics:

🤖 Rule-based decisions (strict execution)
⚡ 24/7 non-stop (capture every opportunity)
📊 Data-driven (based on historical statistics)
🎯 Strong discipline (unaffected by emotions)

1.2 Definition of Quantitative Trading

Quantitative Trading: A trading method that uses mathematical models and computer programs to automatically execute buy/sell decisions based on preset trading strategies.

Three Core Elements:

Strategy → Data → Execution
    ↓        ↓       ↓
   Rules   History  Automation

Strategy
- When to buy? When to sell?
- Example: Buy on EMA golden cross, sell on death cross
Data
- Historical prices, volumes, technical indicators
- Used for backtesting and optimization
Execution
- Automatic order placement, stop-loss, take-profit
- No manual intervention required

1.3 A Simple Example

Strategy Description: Buy when Bitcoin's 20-day moving average crosses above the 50-day moving average, sell when it crosses below.

Code Implementation (Pseudocode):

# Check once a day
for each_day in historical_data:
    MA20 = average_price_of_last_20_days
    MA50 = average_price_of_last_50_days

    if MA20 just crossed above MA50:
        buy(1000 USDT)

    if MA20 just crossed below MA50:
        sell()

Backtest Results (Assumed):

📅 Test Period: Full year 2024
💰 Initial Capital: 10,000 USDT
📈 Final Capital: 12,500 USDT
🎯 Return Rate: +25%
📊 Win Rate: 65%

Conclusion: This strategy performed well in 2024, but doesn't guarantee effectiveness in 2025!

🚀 Part 2: Advantages and Risks of Quantitative Trading

2.1 Advantages of Quantitative Trading

✅ 1. Eliminate Emotional Impact

Case: Panic Selling

Traditional Trader:
BTC down 10% → Panic → Cut losses → Next day recovers → Regret

Quantitative Trader:
BTC down 10% → Bot checks strategy → Sell conditions not met → Continue holding → Next day profit

Advantages:

Won't sell prematurely due to fear
Won't chase highs due to greed
Strictly execute preset rules

✅ 2. 24/7 Round-the-Clock Trading

Cryptocurrency Market Characteristics:

🌏 Global market, 24/7 trading
🌙 Best trading opportunities may appear late at night
⏰ Humans can't watch the screen continuously

Quantitative Trading Solution:

Human: Asleep 😴
Bot: Detects buy signal → Automatic buy → Wake up to 3% profit next day ✅

✅ 3. Fast Execution

Speed Comparison:
| Operation | Manual | Bot |
|-----------|--------|-----|
| Detect Signal | 1-5 minutes | < 1 second |
| Place Order | 30-60 seconds | < 0.1 second |
| Monitor Simultaneously | 1-3 pairs | Unlimited |

Actual Impact:

In fast-moving markets, 0.1% price difference can determine profit/loss
Bots can monitor 50+ trading pairs simultaneously
High-frequency strategies (minute-level) can only be done by machines

✅ 4. Data-Driven Decisions

Traditional Method:

"I think this coin will rise" ← Subjective judgment
"Friend recommended, should be reliable" ← Blind following
"Chart looks like it will break out" ← Empiricism

Quantitative Method:

Backtest 1000 historical trades → Strategy has 70% win rate, average profit 2%
Statistical analysis: Strategy performs well in ranging markets, poorly in trending markets
→ Use only in ranging markets

✅ 5. Backtestable and Optimizable

Advantage Process:

Design Strategy → Backtest Validation → Find Problems → Optimize Parameters → Backtest Again → Live Test

Comparison:

Traditional: Can only test with real money 💸
Quantitative: "Free" testing with historical data 📊

2.2 Risks and Challenges of Quantitative Trading

⚠️ 1. Overfitting

What is Overfitting?

Imagine an extreme strategy:

if date == "2024-03-15" and time == "14:23":
    buy BTC
if date == "2024-03-16" and time == "09:47":
    sell BTC

Backtest Result: Perfect! Caught a historic surge 📈

Live Trading Result: Completely ineffective, because history doesn't repeat exactly 📉

How to Avoid:

✅ Test strategy in different time periods
✅ Keep strategy logic simple, not overly complex
✅ Reserve out-of-sample data for testing
✅ Test at least 1+ year of data

⚠️ 2. Technical Risks

Possible Technical Issues:

Problem	Impact	Case
Network Disconnection	Miss trading opportunities	10% drop during network outage
Program Crash	Cannot execute stop-loss	Server down, losses expand
API Limits	Cannot place orders	Exchange API limit reached, banned
Data Errors	Wrong decisions	Price data delay, buy at high point

Countermeasures:

✅ Use stable servers (VPS)
✅ Set up network monitoring and auto-restart
✅ Use multiple exchanges (backup)
✅ Set maximum loss limits

⚠️ 3. Market Changes

Reasons for Strategy Failure:

2024 Strategy: Ranging market, mean reversion strategy performs well ✅
2025 Market: Bull market, mean reversion strategy loses ❌

Reason: Market characteristics have changed

Response:

✅ Continuously monitor strategy performance
✅ Regular backtesting and optimization
✅ Prepare multiple strategies for different market conditions
✅ Set up early warning for strategy failure

⚠️ 4. Black Swan Events

What are Black Swans?

Extreme rare market events
Rarely appear in historical data
Cannot be predicted by backtesting

Cases:

March 12, 2020: BTC dropped 50% in a single day
May 2022: LUNA collapse, from $80 to $0.0001
Exchange bankruptcy (FTX incident)

Response:

✅ Set strict stop-losses
✅ Diversify investments (don't all-in one coin)
✅ Only invest money you can afford to lose
✅ Maintain appropriate cash reserves

⚠️ 5. Psychological Challenges

Even with quantitative trading, you'll encounter psychological issues:

Backtest Return: +50%
Live Trading 1 month: -5%

Your Reaction:
❌ "This strategy has problems, stop it immediately!"
✅ "Backtest was 1 year of data, 1 month drawdown is normal, continue observing"

Common Psychological Traps:

Frequently modify strategies (pursuing perfection)
Stop when seeing losses (lack of patience)
Increase position after profit (greed)
Manually intervene with the bot (distrust the strategy)

Response:

✅ Create trading plan and execute strictly
✅ Keep trading diary, analyze psychological changes
✅ Give strategies enough validation time (at least 3-6 months)
✅ Accept that strategies have both profits and losses

2.3 Is Quantitative Trading Suitable for You?

✅ People Suitable for Quantitative Trading

📊 Like data analysis: Willing to study historical data and statistical patterns
💻 Have basic programming skills: Can read and understand code, make simple modifications
⏰ Have patience: Accept that strategies need time to validate
🎯 Have discipline: Can strictly execute trading plans
💰 Risk tolerance: Only invest spare money

❌ People Not Suitable for Quantitative Trading

🎲 Like gambling excitement: Pursue overnight wealth
😰 Highly emotional: Panic when seeing losses
⏱ Lack patience: Expect immediate profits
💸 Use loans or essential funds: Cannot afford losses
🚫 Completely non-technical: Unwilling to learn

🤖 Part 3: What is Freqtrade?

3.1 Introduction to Freqtrade

Freqtrade is a free, open-source cryptocurrency trading bot written in Python that supports strategy development, backtesting, optimization, and automated trading.

Official Information:

🌐 Official Website: https://www.freqtrade.io/
💻 GitHub: https://github.com/freqtrade/freqtrade
⭐ Stars: 28,000+
👥 Contributors: 500+
📅 First Release: 2017

3.2 Core Features of Freqtrade

Feature 1: Strategy Backtesting 🔍

Purpose: Validate strategies with historical data

# Test strategy with one command
freqtrade backtesting --strategy MyStrategy --timerange 20240101-20241231

# Output:
# Number of trades: 150
# Win rate: 65%
# Total return: +35%
# Max drawdown: -12%

Value:

Avoid testing directly with real money
Quickly validate strategy feasibility
Discover strategy strengths and weaknesses

Feature 2: Parameter Optimization ⚡

Problem: Strategies have many parameters, how to find the best combination?

# For example, what should these parameters be set to?
RSI_threshold = ?   # Choose between 30-70
MA_period = ?       # Choose between 10-50
stop_loss = ?       # -5% or -10%?

Freqtrade's Solution:

# Automatically test 1000 parameter combinations
freqtrade hyperopt --strategy MyStrategy --epochs 1000

# Output best parameters:
# RSI_threshold = 35
# MA_period = 23
# stop_loss = -7%

Feature 3: Paper Trading (Dry-run) 🎮

Purpose: Test with virtual funds in real-time, no real money

# Start paper trading
freqtrade trade --strategy MyStrategy --dry-run

# System will:
# ✅ Get real-time market data
# ✅ Execute real buy/sell logic
# ❌ But won't actually place orders
# ✅ Record all trading results

Value:

Zero-cost strategy validation
Discover potential live trading issues
Build confidence in the strategy

Feature 4: Automated Trading 🤖

Purpose: Execute trades for real

# Start live trading
freqtrade trade --strategy MyStrategy

# Bot will automatically:
# ✅ Monitor market
# ✅ Discover buy/sell signals
# ✅ Auto place orders
# ✅ Auto stop-loss/take-profit
# ✅ Record trading logs

Value:

24/7 continuous trading
Strict execution of strategy rules
No emotional decisions

Feature 5: Visualization Analysis 📊

Purpose: Generate charts to analyze strategies

# Generate strategy charts
freqtrade plot-dataframe --strategy MyStrategy --pairs BTC/USDT

# Output: Interactive charts with buy/sell point markers

Example Chart Elements:

🟢 Green arrows = Buy signals
🔴 Red arrows = Sell signals
📈 Candlestick charts
📊 Technical indicators (EMA, RSI, etc.)

Feature 6: Multi-Exchange Support 🌍

Supported Major Exchanges:

Exchange	Spot	Futures	Recommendation
Binance	✅	✅	⭐⭐⭐⭐⭐
OKX	✅	✅	⭐⭐⭐⭐
Bybit	✅	✅	⭐⭐⭐⭐
Kraken	✅	❌	⭐⭐⭐
Gate.io	✅	✅	⭐⭐⭐
KuCoin	✅	✅	⭐⭐⭐

Switch Exchanges by Just Modifying Configuration:

{
  "exchange": {
    "name": "binance"  // Change to "okx" or "bybit"
  }
}

3.3 Freqtrade vs Other Trading Bots

Comparison Table

Feature	Freqtrade	TradingView	3Commas	Write Your Own Code
Open Source	✅ Free	❌ Paid	❌ Paid	✅ Free
Flexibility	⭐⭐⭐⭐⭐	⭐⭐⭐	⭐⭐	⭐⭐⭐⭐⭐
Ease of Use	⭐⭐⭐	⭐⭐⭐⭐⭐	⭐⭐⭐⭐	⭐
Backtesting	✅ Powerful	✅ Basic	❌ None	❌ Need to implement yourself
Community	✅ Active	✅ Active	⚠️ Average	❌ None
Learning Curve	⭐⭐⭐	⭐⭐	⭐⭐	⭐⭐⭐⭐⭐

Why Choose Freqtrade?

✅ Advantages:

Completely Free: No subscription fees
Open Source & Transparent: Code is public, secure and reliable
Powerful Features: Backtesting, optimization, and live trading all-in-one
Active Community: Quick answers to questions
Customizable: Full control over strategy logic

⚠️ Disadvantages:

Requires Programming Basics: At least able to read Python
Requires Self-Configuration: Not just a few clicks to use
Steeper Learning Curve: Takes time to master

3.4 What Can Freqtrade Do? What Can't It Do?

✅ What Freqtrade Can Do

✅ Spot trading (buy/sell cryptocurrencies)
✅ Futures trading (long/short positions)
✅ Grid trading
✅ DCA (Dollar Cost Averaging) strategies
✅ Multiple pairs running simultaneously
✅ Stop-loss and take-profit
✅ Risk management
✅ Telegram notifications
✅ Web UI monitoring
✅ Data visualization

❌ What Freqtrade Cannot Do

❌ Predict future prices (no crystal ball)
❌ Guarantee profits (all strategies have risks)
❌ Complete hands-off (needs your monitoring and maintenance)
❌ Auto-generate perfect strategies (needs your design and optimization)
❌ Handle all market conditions (black swan events are hard to handle)

📚 Part 4: Course Learning Path

4.1 Course Overview

Complete 30-Lesson System:

Part 1: Basic Introduction (Lessons 1-4)
└─ Environment setup, core concepts, data download

Part 2: Backtesting Practice (Lessons 5-10)
└─ Backtesting commands, performance analysis, strategy comparison

Part 3: Strategy Optimization (Lessons 11-15)
└─ Hyperopt, scoring systems, risk management

Part 4: Real-time Signals (Lessons 16-20)
└─ Dry-run, Telegram, visualization

Part 5: Live Trading (Lessons 21-25)
└─ API configuration, small capital testing, monitoring

Part 6: Advanced Topics (Lessons 26-30)
└─ Custom strategies, AI, multi-timeframe

4.2 Learning Recommendations

📅 Learning Plan

2-3 lessons per week, complete in 8-12 weeks

Weeks 1-2: Basic Introduction (Lessons 1-4)
Weeks 3-4: Backtesting Practice (Lessons 5-10)
Weeks 5-6: Strategy Optimization (Lessons 11-15)
Weeks 7-8: Real-time Signals (Lessons 16-20)
Weeks 9-10: Paper Trading Validation
Weeks 11-12: Small Capital Live Testing (Optional)

📖 Learning Methods

Theory + Practice Combination
- Each lesson has practical tasks
- Must complete at least 80% of tasks to proceed to next lesson
Keep Learning Notes
- Record problems encountered and solutions
- Create your own command cheat sheet
Join Community
- Freqtrade Discord
- GitHub Discussions
Don't Rush to Live Trading
- Complete at least lessons 1-20
- Run paper trading for at least 2 weeks
- Consider live trading only after stable strategy performance

4.3 Required Prerequisites

✅ Must Have

Basic Computer Skills: Can use computer, browser
Command Line Basics: Know how to open terminal, execute commands
English Reading Ability: Most documentation is in English
Basic Investment Knowledge: Know what buy, sell, stop-loss mean

⚠️ Recommended (Not Required)

Basic Python: Can read simple Python code
Technical Analysis Basics: Know what moving averages, RSI, MACD are
Trading Experience: Have traded cryptocurrencies on exchanges

❌ Not Required

Advanced Programming: Don't need to be a programmer
Complex Mathematics: Don't need advanced math
Extensive Experience: Complete beginners can learn

4.4 Required Learning Resources

💻 Hardware Requirements

Minimum Configuration:
- CPU: Dual-core
- Memory: 4GB
- Hard Drive: 20GB free space
- Network: Stable internet connection

Recommended Configuration:
- CPU: Quad-core
- Memory: 8GB
- Hard Drive: 50GB SSD
- Network: 100Mbps+

💰 Capital Requirements

Learning Phase (Lessons 1-20):

✅ $0 (Use demo accounts)

Live Testing (Lessons 21-25):

⚠️ Recommended $100-500 (Money you can afford to lose)

Formal Trading:

Depends on personal situation
Beginners recommend not exceeding 10% of total assets

📚 Learning Resources

Supporting Documentation:

📄 TESTING_GUIDE.md - Testing Command Guide
📄 STRATEGY_SELECTION_GUIDE.md - Strategy Selection Guide
📄 CONFIG_EXPLANATION.md - Configuration File Details
📄 TROUBLESHOOTING.md - Troubleshooting Guide

Official Resources:

⚠️ Important Risk Warning

Before starting to learn, please understand:

Quantitative trading has risks and may lead to capital loss
- Even the best strategies can lose money
- Past performance does not represent the future
This course is for educational purposes only
- Does not constitute investment advice
- You are responsible for all trading decisions
Always start with paper trading
- Complete at least 2 weeks of dry-run testing
- Consider live trading only after stable strategy performance
Only invest money you can afford to lose
- Don't trade with borrowed money
- Don't use essential living funds
- Beginners recommend not exceeding 5-10% of total assets
Keep learning and be cautious
- Markets are constantly changing
- Need to continuously optimize strategies
- Don't expect overnight wealth

📝 Post-Lesson Tasks

Task 1: Self-Assessment (Required)

Answer the following questions to assess if you're suitable for learning quantitative trading:

What are your learning goals?
- [ ] Learn quantitative trading knowledge
- [ ] Develop automated trading strategies
- [ ] Make profits from live trading
- [ ] Other: ___________
How many hours per week can you dedicate to learning?
- [ ] < 3 hours
- [ ] 3-5 hours
- [ ] 5-10 hours
- [ ] > 10 hours
Your risk tolerance?
- [ ] Conservative (cannot accept any losses)
- [ ] Moderate (can accept small losses)
- [ ] Balanced (can accept some drawdown)
- [ ] Aggressive (pursue high returns, accept high risks)
Your expected returns from quantitative trading?
- [ ] 5-10% annual (conservative)
- [ ] 10-20% annual (reasonable)
- [ ] 20-50% annual (optimistic)
- [ ] > 50% annual (unrealistic)
Do you have programming experience?
- [ ] None (but willing to learn)
- [ ] Some Python
- [ ] Can program but not familiar with Python
- [ ] Proficient in Python

Recommendations:

If you chose "Conservative" and expect "> 50%", reconsider
If you have < 3 hours per week, extend your learning period
If you have no programming experience, learn Python basics first

Task 2: Register Exchange Account (Required)

Goal: Prepare for future learning

Steps:

Choose an exchange (recommend Binance or OKX)
Register account (complete identity verification)
Familiarize yourself with trading interface
Do not deposit funds yet (not needed for learning phase)

Recommended Exchanges:

Binance
- Pros: Largest global, good liquidity, low fees
- Cons: Restricted in some regions
- Link: https://www.binance.com/
OKX
- Pros: Good Chinese support, powerful features
- Cons: Slightly smaller than Binance
- Link: https://www.okx.com/

⚠️ Note: This course uses demo accounts, real accounts only for learning the interface.

Task 3: Join Freqtrade Community (Recommended)

Goal: Get help and communicate

Steps:

Join Freqtrade Discord
Briefly introduce yourself in #introductions channel
Browse #general and #strategy channels
Bookmark official documentation

Task 4: Reflection and Discussion (Optional)

Thought Questions:

What do you think is the biggest advantage of quantitative trading?
If a strategy has 100% backtest returns, would you use real money directly? Why?
Assuming you have 10,000 USDT, how much would you invest in quantitative trading?
What other questions do you have about Freqtrade?

Welcome to share your thoughts in the community!

🎓 Post-Lesson Quiz

Multiple Choice

What is the biggest advantage of quantitative trading?
- A. Can predict future prices
- B. Guaranteed profits
- C. Eliminate emotional impact
- D. No knowledge required
What is overfitting?
- A. Strategy is too simple
- B. Strategy performs well in historical data but ineffective in live trading
- C. Strategy loses money
- D. Strategy is too complex
What is Freqtrade?
- A. Paid trading software
- B. Backtesting tool only
- C. Open source trading bot
- D. An exchange
What is the most important mindset for learning quantitative trading?
- A. Pursue huge profits
- B. Believe in overnight wealth
- C. Caution, patience, continuous learning
- D. Complete reliance on bots
Which of the following is NOT a Freqtrade feature?
- A. Strategy backtesting
- B. Parameter optimization
- C. Predict future prices
- D. Automated trading

True/False

Strategies with good backtest performance will definitely make money in live trading. ( )
Quantitative trading requires no human intervention at all. ( )
Beginners should start with demo accounts for testing. ( )
If the strategy is good, you can invest all your capital. ( )
When markets change, strategies need adjustment and optimization. ( )

Answers at the end ⬇️

✅ Learning Checklist

Before proceeding to Lesson 2, ensure you have:

[ ] Understood what quantitative trading is
[ ] Learned the advantages and risks of quantitative trading
[ ] Recognized Freqtrade and its features
[ ] Clarified your learning goals
[ ] Assessed if you're suitable for learning quantitative trading
[ ] Registered exchange account (no deposit yet)
[ ] Joined Freqtrade community
[ ] Completed post-lesson quiz (accuracy > 80%)

If you have any questions, review the relevant sections or ask in the community.

🎯 Next Lesson Preview

Lesson 2: Freqtrade Environment Setup

In Lesson 2, we will:

Install Python and Conda environment
Install Freqtrade
Configure the first config.json file
Run the first Freqtrade command
Verify the environment is correct

Preparation:

Ensure computer has 20GB free space
Prepare 1-2 hours of complete time
Keep network connection stable

📌 Quiz Answers

C - Eliminate emotional impact
B - Strategy performs well in historical data but ineffective in live trading
C - Open source trading bot
C - Caution, patience, continuous learning
C - Predict future prices
❌ False (history doesn't represent the future)
❌ False (requires monitoring and maintenance)
✅ True
❌ False (only invest money you can afford to lose)
✅ True

💬 Post-Lesson Discussion

Discussion Topics

Share your quantitative trading goals
- What goals do you hope to achieve through this course?
- How much time and capital do you plan to invest?
Your concerns about quantitative trading
- What risks worry you most?
- Which part interests you most?
Learning Partner Recruitment
- Find partners to learn together
- Form study groups to supervise each other

Welcome to discuss in the community's #course-discussion channel!

Congratulations on completing Lesson 1! Ready to move on to practical sessions? 🚀

Next lesson we will officially install Freqtrade and begin your quantitative trading journey!

📝 Course Feedback: If any part of this lesson is unclear, feel free to provide suggestions!