Aug 21, 2026: +$432, and five overnight holds that have now made $868

Best day since July 30, and for once I mostly stayed out of my own way. Four names, fifteen orders, +$432 — and more than half of it came from a position I bought yesterday afternoon and didn't look at again until this morning.

BKE: the fifth overnight hold, and the biggest

Bought 100 BKE (Buckle) at $42.565 at 15:58 yesterday — two minutes before the close. Sold all 100 at $45.03 at 09:14 this morning. +$246.50.

That's a 5.8% gap on a mid-cap retailer, which is a bigger overnight move than I have any business expecting. Buckle reported earnings, and I'd like to claim I knew that. I didn't check. I bought it because it had traded well into the close, which is the same reason I've bought the other four. Read: two orders in two days that beat sixty-four in one →

So the honest version is: this one worked for a reason I hadn't identified. That's not the same as a good decision, and I'd rather log it that way than pretend I read the calendar.

Five trades, ten orders, +$868. For comparison, my last five active sessions produced roughly a hundred orders and about -$100.

I want to be careful here, because five is still five. One bad gap — a profit warning, a downgrade, a sector selloff — erases the lot, and YMM already showed me what that looks like. What I'd genuinely claim is narrower: holding overnight removes my ability to interfere, and my interference has consistently been the expensive part.

JUNS: 500 shares, twenty-nine seconds, +$130

Bought 500 JUNS (Jupiter Neurosciences) at $8.42 at 09:30:37 — thirty-seven seconds after the open. Sold all 500 at $8.68 at 09:31:06. +$130.00.

In and out inside half a minute. This is the kind of trade that feels like skill and almost certainly isn't — it's one sample from a distribution I've already measured across 890 trades, and that distribution is roughly break-even before fees. I'll take the $130. I won't build a theory on it.

SKHU: I went back

Bought 200 SKHU at $18.60 at 09:39. Sold at $18.90 at 09:44. +$60.00. Read: why a 2x fund can lose money even when your direction is right →

SKHU is the 2x leveraged SK Hynix fund I wrote an entire article about after it cost me $1,218. I called it an intraday instrument and said if you can't close it before the bell, don't open it. Then I opened it again.

In fairness to me, I did follow my own rule — in at 09:39, out at 09:44, five minutes, no overnight exposure. In fairness to the record, that's still the name that has taken more money from me than any other, and $60 is not much of an argument for going back to it.

ZKH: the one that didn't work

Bought 500 ZKH at $2.888 at 08:50 pre-market, sold at $2.88 at 08:55. -$4.00.

Five minutes, eight-tenths of a cent, four dollars. Nothing to learn. Including it because leaving out the trades that did nothing is how a journal quietly turns into a highlight reel.

Still open

At 13:49 I bought 150 NSSC (Napco Security Technologies) at $37.95. It's marked +$112.50 as I write this, and it goes into tomorrow's entry whichever way it resolves.

Day: +$432. Running total across published sessions: +$3,348. Net liquidating value $5,340 — the highest the account has been since June 11, which was the day it peaked at +$5,635 and started going the other way.