The 9/28 orders never got placed. That's on me. Missed trade number three, if you're scoring at home, and at this point I'm the weakest link in a two-person operation where the other person is a computer.
If they'd gone in, the sell would have filled at the open and the $130 buy-back would have filled on 10/12. The difference by 10/13: about $56. Not a disaster. Still annoying.
So we fixed the actual problem, which is me. Any rule that needs someone watching the market at 4pm doesn't work when that someone is at work. On 10/14 the rules became standing orders that live at the broker: a stop to sell all 8 shares if SOXL hits $95, replacing "sell if it closes under $100." Every call from now on goes in the same evening, and I confirm it's in. No more "I'll do it tomorrow."
The broker rejected the $200 sell order because the shares were already tied up in the stop. So that one gets checked by hand. Of course it does.
10/15: SOXL dropped to $95.23 during the day. Twenty-three cents above the stop. Then it closed at $107.35. If you want to know what 23 cents feels like, it feels like a lot.
10/16: the stop filled at $95.00. The account is now $768.82, all cash, down 23.1%. Loss on the shares: $231.18.
Same day, before any of the data came in, it wrote the rules for getting back in: no buying until Meta, Google, Microsoft and Amazon all report earnings. If at least three of the four keep or raise their AI spending plans, and none of a list of "someone's running out of money" warning signs goes off, buy back in with everything on 10/30. Written down. In advance. Look at us.
← Five Months of HOLD · Re-entry, Another Stop, and Then the Floor Fell Out →
Not investment advice. This is a log of one experiment: $1,000 and an AI that can put it anywhere.