研究日誌
週報、交易案例分析、放棄的機會、系統中斷紀錄，以及每次規則調整後的檢討。
案例分析
INTC: the naked window can go your way too
A worked example of the same mechanism that produced this project's largest loss — this time landing favorably.
2026-09-12
案例分析
SNDK: a zero-dollar naked window, and a $109 loss anyway
The naked-window price move was exactly zero on this trade. It still lost money — because the two legs drifted apart while both were supposedly hedged.
2026-09-12
案例分析
ARM: a 15-day outage, not a strategy failure
The largest single loss in this project's history came from a scanner outage, not from the funding-rate thesis being wrong.
2026-09-11
放棄的機會
HYUNDAI: three losses, then blacklisted — and why the automatic system should have caught it
A symbol that kept qualifying on paper but kept losing in practice, and the gap between an automatic rule and its intent.
2026-09-11
週報
Weekly review: September 1–8, 2026
Five closed trades, all net-positive on strategy accounting — but the naked-leg reconciliation tells a more mixed story.
2026-09-11
規則變更
A stale Friday-close price got locked in as a Monday-open fill
Three trades showed exactly $0.00 naked-leg P&L after a weekend gap — an implausible result that led straight to a real pricing bug.
2026-09-01
規則變更
Why 'pause on weekends' wasn't enough — replacing a day-of-week check with the real thing
A weekend-only naked-leg pause missed most of the actual problem, because ordinary weeknight gaps are just as common as weekends.
2026-08-25
規則變更
A config value nobody was reading — 23 trades and $219.72 lost to a wiring bug
A minimum-hold safeguard existed in the settings file for two months. It never once ran.
2026-08-06
放棄的機會
We tested blocking 'already-declining' funding entries — and our first result was wrong
A promising-looking filter improvement turned out to be two backtest bugs, not a real edge.
2026-07-24