I sold my 15 GM shares on 30 October 2026, nine days after I bought them. This closes the trade I wrote about in Testing AI trading: GM. It hit the target, and I took it.
Trade summary
| Action | Sell (close) |
| Ticker | GM (NYSE) |
| Quantity | 15 shares |
| Exit price | USD 85.86 (limit) |
| Executed | 30 Oct 2026, 22:19 SGT (10:19 ET) |
| Entry | USD 76.87, 21 Oct 2026 |
| Realised P&L | +USD 132.73 (about +11.5%) after fees |
| Days held | 9 |
| Followed the plan | Yes |
Why I sold
GM closed at USD 85.72 on 29 October, above my USD 84.80 target. It hit the target and I didn’t want to be greedy. I could have moved the stop up and held, but a stop only triggers in US market hours. If the price fell overnight, the stop would wait, and the sale wouldn’t be confirmed.
Basic TA at exit
GM, as of the 29 October 2026 close, from IBKR daily bars.
| Indicator | Value | Read |
|---|---|---|
| Price | USD 85.72 | Close before I sold |
| MA20 | 80.88 | Price above: bullish |
| MA50 | 83.08 | Price above: bullish |
| RSI 14 | 59.7 | Neutral (overbought is 70 and above) |
| Trend | Four up days, 76.31 to 85.72 | Strong short-term up |
| Support | 82.40, then about 81 (MA20) | First floor about 4% below |
| Resistance | 87 to 88, then 91.85 | Next ceiling 1.5% to 2.7% above |
| Volume | 5.57M vs 4.79M average | 16% above average, buyers behind the move |
Overall read: bullish, with room to about 87 or 88. I sold on the plan, not on the chart.
What I did first
Early on 30 October (Singapore time), Claude suggested raising the stop from USD 74.50 to USD 82.30, just under the previous day’s low. I did it. That locked in a gain of about USD 81 if things went wrong at the open.
What executed
At the US open, my first sell at USD 85.80 was rejected. IBKR’s order list showed why: the 82.30 stop was still live, so the 15 shares were already promised to it. I cancelled the stop, then placed a sell at USD 85.86. Claude suggested dropping it to 85.65 to fill straight away. I held at 85.86, and it filled five minutes later, in two pieces (1 share, then 14). Commission was USD 1.12.
Did I follow the plan
Yes. The stop stayed in until the sale, and I sold above the target, as planned.
What Claude got right and wrong
- Right: the ticker, size, price, stop and target, which made about 11.5%.
- Right: raising the stop before the open.
- Wrong: Claude told me to sell first, then cancel the stop. It’s the other way round.
- My call over Claude’s: I kept my limit at 85.86 instead of 85.65. It filled, and was worth about USD 3 more.
What I’d change
Cancel the stop first, as soon as I decide to sell, before placing the sell order.
What happened next
30 October 2026. GM kept going after I sold. It closed the day at USD 87.40, with a high of 88.02. Holding would have made about USD 23 more on 15 shares. That’s fine. The target was the plan, and I took it.
5 Nov 2026, scoring note. The target agreed before I bought was 84.80. I sold at 85.86, my own price, after the target was reached. Under the rule I set on 5 November (target means sold at the agreed price; anything else is manual), this exit is now scored as manual. The trade, the price and the profit are unchanged.
Sources
- IBKR Trades and order list, 30 October 2026
- IBKR GM daily chart, as of 31 October 2026
Updates to this post
| Date | What changed |
|---|---|
| Title updated to match the manual score | |
| Outcome rescored target to manual under the 5 Nov rule; scoring note added | |
| Turned the chart table into a Basic TA block, and added % distances to support and resistance |
This is a personal record of a small, capped trading experiment, not financial advice. The trade ideas came from Claude; the decisions were mine.