Core Trading Philosophy

The primary objective of this framework is to document daily market conditions for statistical analysis and backtesting purposes. The short-term core strategy adheres to a "Leader-Rest-Leader" model, avoiding adjustments across all levels. The operational guideline suggests that while a stock may be tradable during 10% of market conditions, remaining empty (cash position) is appropriate for the remaining 90% of the time.

Statistical observation indicates a promotion rate of approximately 57% for stocks advancing from the second to the third limit-up board.

Data Screening Methodologies

Two primary methods are utilized for identifying potential targets using database queries.

Method 1: Indicator-Based Selection

This approach filters stocks based on specific technical indicators stored in the signal database.

SELECT * 
FROM strategy_signals 
WHERE signal_type LIKE '%indicator_based%' 
  AND trade_date = '2024-02-23';

Method 2: Trend Analysis for Leaders

This method identifies leading stocks by confirming sustained upward momentum across multiple price points over a three-day period. The criteria include:

1. Lowest price increasing for 3 consecutive days.
2. Average price increasing for 3 consecutive days.
3. Closing price increasing for 3 consecutive days.
4. Highest price increasing for 3 consecutive days.
5. Opening price increasing for 3 consecutive days.
6. Moving averages displaying a divergent (bullish) formation.

SELECT * 
FROM strategy_signals 
WHERE signal_type LIKE '%trend_analysis%' 
  AND trade_date = '2024-02-23';

Method 3: Five-Step Filtering Process

A comprehensive query combines volume, flow, and price action data to isolate high-probability setups. The logic filters for specific opening ranges, volume ratios, and alignment with limit-up candidates.

SELECT 
    ticker,
    volume_rank,
    net_inflow_rank,
    stock_code,
    stock_name,
    large_order_amount,
    price_change_pct,
    institutional_net_flow,
    large_order_net_ratio,
    CAST(last_limit_time AS DATETIME) AS limit_time,
    limit_analysis,
    three_day_gain_pct,
    consecutive_boards,
    order_block_amount,
    order_block_rank,
    retail_flow_metric,
    total_amount,
    profit_pool_pct,
    open_amount,
    open_change_pct,
    volume_ratio
FROM daily_stock_snapshot
WHERE open_change_pct BETWEEN 1 AND 5
  AND open_amount BETWEEN 100 AND 5000
  AND volume_ratio BETWEEN 1 AND 5
  AND stock_name IN (
      SELECT short_name 
      FROM auction_limit_ups 
      WHERE trade_date = '2024-02-23'
  );

Market Sentiment Indicators

Selection hierarchy follows the structure: Market Phase → Sector → Individual Stock.

• ADR (Advance-Decline Ratio): Typically ranges between 0.5 and 1.5.
• ADR > 1.5: Indicates a high probability of market pullback.
• ADR < 0.65: Suggests a high probability of market rebound.
• ADR < 0.3 or 0.5: Often signals the formation of a market bottom.

Leader Classification

Leaders are categorized into Market General Leaders and Sector-Specific Leaders. Recent focus has been on the Sora concept sector.

Profit Effect

The core truth layer relies on identifying the fastest-rising个股 (individual stocks) and selecting the strongest performers for the subsequent session.

Strategy Models

The following trading patterns are defined within the system:

1. Leader Model: Focus on market dominators.
2. First to Second Board: Buying leaders at the second limit-up.
3. One Yang Penetrates Four Lines: Capturing explosive breakout points.
4. Hot Spot Selection: Turnover board + Limit up before 10 AM + Hot Sector + Order block > 100M + Price above 5-day MA.
5. Explosion Point Dragon Return: Re-entry after initial surge.
6. Limit-Up Accumulation Model.
7. Smooth Trend Model.
8. Triple Protection Model.
9. Old Duck Head: Leader's first negative day after surge.

Execution and Risk Management

Position sizing is maintained at approximately 60%. Psychological discipline is critical; maintain stability and trust the buy signals.

Daily indicators for validation include:

• Advance/Decline Ratio: Win rate increases if greater than 50%.
• Market Popularity.
• Dragon-Tiger List (Top Buyers/Sellers).
• Theme Database.

Aggregated Signal Logic

The final selection process aggregates multiple strategies. The logic requires a stock to satisfy atleast two model conditions, exhibit bullish moving averages, face resistance levels, show positive institutional net flow, and demonstrate a decrease in retail participation.

SELECT * 
FROM daily_stock_snapshot A
WHERE A.stock_name IN (
    SELECT T.name FROM (
        SELECT 
            X.stock_name, 
            Y.signal_count 
        FROM strategy_signals X 
        INNER JOIN (
            SELECT stock_name, COUNT(1) AS signal_count 
            FROM strategy_signals 
            WHERE trade_date = '2024-02-23' 
            GROUP BY stock_name 
            ORDER BY signal_count DESC 
            OFFSET 0 ROWS FETCH NEXT 1000 ROWS ONLY
        ) Y ON X.stock_name = Y.stock_name
        WHERE X.trade_date = '2024-02-23' 
          AND X.price_change_pct > 9 
          AND Y.signal_count >= 2 
          AND X.institutional_net_flow > 0 
          AND X.retail_flow_metric < 0 
        ORDER BY Y.signal_count DESC, X.consecutive_boards DESC
    ) T
)
AND A.current_price > A.ma_120 
AND A.current_price > A.ma_250 
AND A.ma_120 > A.ma_250
AND A.institutional_net_flow > 0
ORDER BY A.consecutive_boards DESC;

Directional Focus

The most fermentable theme for the session is Artificial Intelligence. The stock pool consists primarily of historical leader stocks. Capital inflow exceeding 200 million, consecutive large bullish candles, and rapid limit-up actions are key identifiers for core targets.

Performance Review

Previous day strategy success rate was approximately 50%. Current screening suggests a rolling operation with 50% position allocation, aiming for a perfect execution score. No specific leader was identified in the immediate screen, with observation focused on mechanical performance.