Skip to content
All library documents

Shorting Crypto Pump Coins with AI Screening and Trailing Risk Controls

Article FMZ digest · Author: 发明者量化-小小梦

Summary

This strategy shorts high-gaining Binance USDT perpetual contracts, screening for coins that may be near a reversal. It gathers open interest, funding, market capitalization, and daily candle data, then applies hard exclusions and an AI model’s weighted six-factor score. Only candidates above the stated score threshold proceed to short entry, sized at a fixed amount. The article describes OI relative to market capitalization as its most heavily weighted measure and uses candle patterns and funding conditions as additional inputs.

A separate, faster risk-control workflow monitors positions. It can add to a short as price rises, subject to two additions and a stated total exposure cap, and uses a trailing profit drawdown after unrealized gains reach a trigger level; a manual stop-loss is also supported. The document supplies rules and code excerpts, but no historical performance results or validation of the AI signals. Scaling into a losing short can magnify losses, and the author cautions that persistent upward markets or fundamental catalysts can defeat the reversal thesis.

Key ideas

  • The entry screen selects top-gaining USDT perpetuals and excludes coins already held or failing hard filters.
  • The AI scoring model combines leverage, candle patterns, liquidity, volume, price gains, and funding.
  • Each initial short uses fixed sizing, while capped additions increase exposure if price continues rising.
  • A trailing take-profit follows peak unrealized profit, and a manual stop-loss is available.
  • The strategy may suffer sustained losses in a strong bull market and is presented without performance validation.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.