🃏 Intelligent Blackjack Game

🌐 Language / 语言选择: English | 中文

A fully-featured blackjack game written in Go, integrated with Monte Carlo probability analysis, Kelly Criterion bankroll management, and intelligent decision recommendation systems.

✨ Core Features

🎯 Intelligent Probability Analysis

• Real-time probability calculation: Based on Monte Carlo simulation (10,000 trials) to calculate winning probabilities
• Action win rate comparison: Analyzes expected win rates for hit, stand, double down, and other actions
• Optimal strategy recommendation: Automatically recommends the best decision based on basic strategy

💰 Kelly Criterion Bankroll Management

• Intelligent betting suggestions: Provides scientific betting amount recommendations based on bankroll status
• Risk assessment: Real-time evaluation of current bankroll status and risk level
• Double down decision analysis: Evaluates expected ROI and risk-reward ratio for doubling down

🧠 Decision Support System

• Basic strategy integration: Built-in professional blackjack basic strategy
• Real-time data analysis: Displays key indicators like bust probability, 21-point probability, etc.
• Entertainment cost estimation: Helps players understand expected entertainment costs

🚀 Quick Start

Requirements

• Go 1.22+
• UTF-8 compatible terminal

Install Dependencies

go mod tidy

Run the Game

# Method 1: Run directly
go run ./cmd

# Method 2: Build and run
go build -o blackjack ./cmd
./blackjack

🎮 Game Controls

Basic Actions

• h / hit - Hit (take a card)
• s / stand - Stand
• d / double / doubledown - Double down
• q / quit - Quit game
• y / yes - Continue game
• n / no - End game

Menu Options

• 1 - Start game
• 2 - View game rules
• 3 - Exit program

🃏 Game Rules

🎯 Game Objective

Get your hand as close to 21 as possible without going over, while beating the dealer's hand.

🎴 Card Values

• Number cards (2-10): Face value
• Face cards (J,Q,K): 10 points each
• Ace: Intelligently calculated as 1 or 11 (whichever is more favorable)

💰 Betting System

• Starting chips: 1000
• Betting options: 10, 25, 50, 100, 200 chips
• Smart betting: Automatic all-in option when chips are insufficient
• Payout rules:
  • 🏆 Regular win: 1:1
  • 🌟 Blackjack win: 3:2 (non-double situations)
  • 🤝 Push: Return original amount

⚡ Double Down Feature

• Trigger condition: Available on first two cards
• Chip requirement: Current chips ≥ current bet amount
• Game rule: Can only take one more card after doubling
• Payout adjustment: Blackjack pays 1:1 after doubling

🎲 Game Flow

1. Betting phase → Choose bet amount + bankroll management advice
2. Dealing phase → Player and dealer each get 2 cards (dealer has 1 hidden card)
3. Player turn → Choose hit/stand/double + probability analysis
4. Dealer turn → Dealer follows automatic rules
5. Settlement phase → Compare points and settle chips

📊 Intelligent Analysis System

🎯 Real-time Probability Analysis

The game displays detailed probability analysis each turn:

📊 Current Win Probability Analysis
────────────────────────────────────────
🟢 Player Win Probability: 45.6%
🔴 Dealer Win Probability: 16.7%
🟡 Push Probability:       37.6%

📈 Detailed Analysis:
   💥 Player Bust Probability: 85.1%
   💥 Dealer Bust Probability: 22.7%
   🎯 Player 21 Probability: 8.5%
   🎯 Dealer 21 Probability: 6.3%

🎯 Action Win Rate Comparison:
   ✋ Stand: 45.8% ⭐ (Recommended)
   👆 Hit: 41.9%
   ⚡ Double: 36.3%

💰 Kelly Criterion Bankroll Management

Betting Phase Advice

💰 Bankroll Management Advice:
📊 Recommended Bet: 15 chips (1.5% of bankroll)
💡 Your bankroll status is good, moderate betting recommended
🟢 Risk Status: Sufficient funds, controllable risk
🎮 Expected Entertainment Cost: 0.06% per hand

Double Down Decision Analysis

💰 Kelly Criterion Double Analysis:
⚡ Recommend Double (Expected ROI: 18.0%)
🔴 Double Risk Level: High (Kelly fraction: 0.180)

🧮 Tiered Bankroll Management Strategy

Bankroll Level	Recommended %	Risk Level	Strategy Description
≥ 1000 chips	1.5%	🟢 Low	Sufficient funds, moderate betting
≥ 500 chips	1.0%	🟢 Low	Conservative betting, risk control
≥ 200 chips	0.5%	🟡 Medium	More conservative, minimum betting
< 200 chips	Minimum	🔴 High	Recommend caution or leave game

🧠 Probability Calculation Principles

🎲 Monte Carlo Simulation Method

The game uses Monte Carlo simulation to calculate various probabilities:

1. Simulation count: 10,000 simulations per analysis
2. Strategy simulation: Player uses basic strategy, dealer follows fixed rules
3. Remaining deck: Based on current known cards and remaining deck
4. Statistical results: Statistics on frequency of various outcomes

📈 Advantages Over Direct Calculation

Aspect	Monte Carlo Simulation	Direct Calculation
Complexity Management	✅ Handles complex strategies simply	❌ Combinatorial explosion
Strategy Integration	✅ Naturally integrates basic strategy	❌ Difficult to handle strategy changes
Extensibility	✅ Easy to add new rules	❌ Need to rewrite calculation logic
Multi-variable Handling	✅ Naturally handles multiple factors	❌ Complexity explodes with dimensions
Dynamic Adaptation	✅ Automatically adapts to deck changes	❌ Need to re-derive formulas

💡 Kelly Criterion Application

Basic Formula

f* = (bp - q) / b

• f*: Optimal betting fraction
• b: Odds (1:1 or 3:2)
• p: Win probability
• q: Loss probability

Practical Application Strategy

1. Conservative factor: Use 25% of Kelly fraction to reduce risk
2. Tiered management: Different strategies based on bankroll status
3. Entertainment-oriented: Focus on bankroll management rather than strict expected returns

🏗️ Architecture Design

📁 Project Structure

go-blackjack/
├── cmd/                         # 🚀 Program entry
│   └── main.go
├── internal/                    # 🔒 Internal modules
│   ├── domain/                  # 🎯 Domain layer - Core business logic
│   │   └── entities/
│   │       ├── game.go          # Game aggregate root
│   │       ├── player.go        # Player entity
│   │       ├── dealer.go        # Dealer entity
│   │       ├── card.go          # Card entity
│   │       ├── deck.go          # Deck entity
│   │       ├── hand.go          # Hand entity
│   │       └── types.go         # Type definitions
│   ├── application/             # 🔄 Application layer - Use case orchestration
│   │   ├── services/
│   │   │   ├── game.go          # Game application service
│   │   │   └── probability.go   # Probability calculation service
│   │   └── dtos/
│   │       └── game.go          # Data transfer objects
│   └── interfaces/              # 🖥️ Interface layer - User interaction
│       └── cli/
│           ├── game.go          # CLI handler
│           └── display.go       # Display service
├── go.mod                       # 📦 Dependency management
├── go.sum
└── README.md                    # 📖 Project documentation

🎯 Architecture Layers

Domain Layer

• Responsibility: Core business logic and rules
• Characteristics: No external dependencies, pure business code
• Contains: Game entities, business rules, state management

// Game aggregate root - Unified game state management
type Game struct {
    ID          string
    Player      *Player
    Dealer      *Dealer
    Deck        *Deck
    State       GameState // State machine management
    RoundNumber int
}

Application Layer

• Responsibility: Use case orchestration, domain object coordination
• Characteristics: Handles business processes, data transformation
• Contains: Application services, probability calculation, DTO objects

// Game application service - Orchestrates game use cases
type GameApplicationService struct {
    game            *entities.Game
    probabilityCalc *ProbabilityCalculator
}

// Probability calculator - Monte Carlo simulation
type ProbabilityCalculator struct {
    trials int // Number of simulations
    rng    *rand.Rand
}

Interface Layer

• Responsibility: User interaction, input/output handling
• Characteristics: Extensible for multiple interface implementations
• Contains: CLI handler, display service

// CLI game handler
type GameHandler struct {
    gameService *services.GameApplicationService
    display     *DisplayService
}

🔄 Data Flow

User Input → CLI Handler → Application Service → Probability Calculator
                    ↓                              ↓
Domain Entities ← Application Service ← Monte Carlo Simulation
                    ↓
CLI Display ← Kelly Formula Analysis

🧪 Core Features

🎮 State Machine Management

type GameState int
const (
    StateWaitingToBet GameState = iota
    StatePlayerTurn
    StateDealerTurn
    StateGameOver
)

🃏 Intelligent Card Value Calculation

Automatically calculates optimal Ace value (1 or 11) to ensure best possible hand value.

📊 Probability Analysis Engine

type ProbabilityResult struct {
    PlayerWinProbability  float64
    DealerWinProbability  float64
    PlayerBustProbability float64
    ActionAnalysis        *ActionAnalysis
}

💰 Kelly Criterion Decision Making

type KellyRecommendation struct {
    RecommendedBetAmount   int
    RecommendedBetFraction float64
    ShouldDouble           bool
    RiskLevel              string
}

💡 Business Rule Validation

Strict business rule validation in the domain layer ensures game logic correctness.

🔧 Extensible Design

• Multi-interface support: Easy to add web, mobile interfaces
• Feature extension: Easy to add split, surrender features
• Algorithm optimization: Supports different probability calculation methods

🎯 Game Screenshots Examples

Betting Phase - Bankroll Management Advice

💰 Current Chips: 1000
Please select bet amount:
1. 10 chips
2. 25 chips
3. 50 chips
4. 100 chips
5. 200 chips
6. Exit game

💰 Bankroll Management Advice:
📊 Recommended Bet: 15 chips (1.5% of bankroll)
💡 Your bankroll status is good, moderate betting recommended
🟢 Risk Status: Sufficient funds, controllable risk
🎮 Expected Entertainment Cost: 0.06% per hand

Game Round - Intelligent Analysis

👨 Dealer Hand (first card hidden):
🂠 🃏A

👨 Player Hand (Value: 9):
🃏5 🃏4

────────────────────────────────────────
📊 Current Win Probability Analysis
────────────────────────────────────────
🟢 Player Win Probability: 59.1%
🔴 Dealer Win Probability: 34.0%
🟡 Push Probability:       6.8%

📈 Detailed Analysis:
   💥 Player Bust Probability: 0.0%
   💥 Dealer Bust Probability: 44.4%
   🎯 Player 21 Probability: 0.0%
   🎯 Dealer 21 Probability: 8.7%

🎯 Action Win Rate Comparison:
   ✋ Stand: 44.2%
   👆 Hit: 56.4%
   ⚡ Double: 59.0% ⭐ (Recommended)

🏆 Optimal Strategy Expected Win Rate: 59.0%

💰 Kelly Criterion Double Analysis:
   ⚡ Recommend Double (Expected ROI: 18.0%)
   🔴 Double Risk Level: High (Kelly fraction: 0.180)
────────────────────────────────────────

🎯 Future Plans

Short-term Goals

• 🧪 Complete unit test coverage
• 📈 Add historical statistics feature
• 🎮 Implement split functionality
• 🔄 Add surrender option

Medium-term Goals

• 🌐 Develop web interface version
• 💾 Implement game data persistence
• 📱 Mobile adaptation
• 🤖 AI opponent mode

Long-term Goals

• 🏆 Multiplayer functionality
• 📊 Detailed statistical reports
• 🎯 Custom rule settings
• 🌍 Internationalization support

🤝 Contributing

Issues and Pull Requests are welcome!

Development Environment Setup

git clone https://github.com/yourusername/go-blackjack.git
cd go-blackjack
go mod tidy
go test ./...

Code Style

• Follow Go official coding standards
• Use meaningful variable and function names
• Add appropriate comments and documentation

📄 License

This project is licensed under the MIT License - see the LICENSE file for details

---

⭐ If this project helps you, please give us a star!