I Built a Trading Bot That Refuses to Trade
说明与用途
一个只在两层 AI 给予 60% 以上置信度时才下单的交易机器人,避免情绪化交易。
问题
Most trading bots trade too often, driven by emotion, leading to losses. The goal was to create a bot that trades only when mathematically justified, reducing overtrading and emotional decisions.
结果
In a live test on 21 Oct 2026, AI #1 was 85.4% confident but Laya returned 12.0% confidence, causing the bot to stay silent. The two‑gate logic prevented a trade, demonstrating the bot’s ability to refuse trading when confidence is low.
工作流
- Scan market for volatility and suitable stocks
- Analyze price action of selected stock
- Generate three risk‑reward setups and compute confidence
- If confidence <60% stop
- Send setups to second AI
- Second AI compares to past trades and selects best setup
- If second AI confidence ≥60% produce trade plan
- Otherwise output silence
证据
Most trading bots are trigger‑happy — and most traders lose money for one reason: emotion. FOMO entry, revenge trade, overtrading.
Step 1 — Market scan. The first AI analyzes the whole market: which stocks have volatility, and which stock is actually worth trading.
Step 3 — Three setups, one confidence score. At that point it builds 3 trade setups based on risk‑reward, and gives a confidence score. Below 60%? Stop.
AI #2 (laya) picked setup_1 (RR 1:1.5) with 12.0% confidence.
Hard rule: mode: "paper" in the config. This project cannot place real orders — it's educational by design.
- 记录日期: 2026-10-25T06:11:17.232Z