MarketRhapsody

Understand Why Price Moves

We show not just what happened in the market, but why: real client trade flow with millisecond precision.

How It Works

Why This Is Valuable

  • ✓ Labeled trades: entry/exit, new/existing clients
  • ✓ Millisecond timestamps for sync with tape/orderbook
  • ✓ Thousands of events → see crowd behavior

Simple Example

ETH surges from $4,800 to $4,950. Price charts say "went up". Our flow shows: retail piled in at $4,930, while larger wallets were quietly selling. This way you can catch FOMO exhaustion earlier.

Complete Exchange Data Model

This is essentially a full exchange model powered by our super-advanced AI, available to any client for a small fee. You get AI-modeled trader profiles with complete picture: order types, P&L, execution timing, and behavioral patterns.

📊 Data Fields

Core Trade Data

  • account_id: AI-modeled trader profiles
  • instrument: Trading pair (ETH/USDT, BTC/USDT, etc.)
  • direction: Buy/Sell
  • order_type: Limit, Stop, Market

Timing & Execution

  • created_at: Order placement timestamp
  • finished_at: Execution timestamp
  • price: Execution price
  • quantity: Trade size

P&L Analysis

  • settlement_usd: Realized P&L in USD
  • usd_amount: Trade value in USD
  • final_amount: Net position after fees

📈 Sample Data

enc_10daec09,eth/usdt,buy,limit
2025-01-13 14:39:23.448
2025-02-02 22:40:05.527
$2830.0, 0.07 ETH
+$198.16 USD

What This Shows:

  • ✓ Order Types: Limit orders, stop losses, market orders
  • ✓ Execution Delay: 20 days from placement to execution
  • ✓ Profit Tracking: Realized gains/losses per trade
  • ✓ AI Trader Modeling: Consistent behavioral patterns across instruments

💡 Analysis Possibilities

• Track order execution patterns by type
• Measure stop-loss vs limit order performance
• Analyze trader behavior across timeframes
• Build P&L prediction models
• Reverse-engineer AI trader behavioral patterns

🎯 Available Instruments

BTC/USDT
ETH/USDT
DOT/USDT
MKR/USDT
ALGO/USDT
INJ/USDT
+ More

AI Model: Entry Zone Prediction

What It Learns

  • Labels: entry/exit, new/existing accounts
  • Features: returns, volatility, flow imbalance
  • Microstructure: size bursts, inter-arrival times

Target

P(entry cluster in price band k in next N minutes/hours).

Model highlights price bands with highest probability of new entries.

Mini Example

Price hovers at $4,950, small buys increase with fresh accounts → probability rises for $4,945–$4,955 zone over 10 minutes.

📥 Download Demo Dataset