Dragon-Tiger List Intelligent Analysis System

Problem

Common Dragon-Tiger List workflows rank stocks by seats, net buying, and institutional or speculator resonance, but short-horizon A-share limit-up trades depend heavily on market regime. A strong resonance signal can fail in a freezing phase, while weaker setups can work during recovery. This skill addresses evaluating Dragon-Tiger signals within an emotion cycle, reducing reliance on single-stock strength alone. It targets short-term A-share trading around next-day sentiment, sector rotation, and seat resonance, producing conservative, tiered recommendations with position guidance.

How it works

• Cycle detection: uses metrics such as limit-up count, consecutive-board height, board ladder depth, and limit-down count to estimate sentiment temperature and classify the market into ice, recovery, climax, and ebb phases.
• Signal tiering: reuses seat recognition, resonance analysis, and signal modules to classify stocks into A+, A, B+, B, C, and D buckets based on institutional or speculator behavior, sector momentum, turnover, market-cap elasticity, and other filters.
• Regime adaptation: adds score boosts in recovery and downgrades recommendations in the ebb phase, with position sizing adjusted by cycle stage.
• Report output: can generate structured output for a chosen --date, --top, and --output-dir, typically producing an HTML report for preview.

Boundaries

• Classifications are based on historical backtests and do not guarantee returns; strong institutional or resonance patterns may still be small-sample effects.
• Emotion-cycle detection relies on closing data, so intraday sentiment can change quickly; it should not be treated as an automatic intraday risk signal.
• Use it as short-term research reference only, paired with stop-loss discipline, position control, and capital preservation, especially during ebb phases.