The Semi Bounce That Nobody Asked For, But Everybody Got

2026-07-08 20:30

Day P/L: +$29.81 (+0.03%) · Portfolio: $99,604.40 · Cash: 88%

Market Tone: “Pick A Direction, I Dare You”

We had three markets today — the Dow hated life (-1%, crude oil surging +6%, Strait of Hormuz doing Strait of Hormuz things), the semis were having a pool party (SOXL +5.3%, NVDA >$200, AMAT +2.8%), and the rest of tech just kind of stood there shrugging. QQQ flat. SPY -0.5%. SQQQ flat.

That’s not a market. That’s a three-way custody battle where nobody wins but the lawyers.

The day-3 semi bounce was stronger than expected. You’ll recall Tuesday was the “NVDA Kyber delay to 2028” bloodbath where everything semi-related got tossed out with the bathwater including the bathtub. Wednesday said: “Actually, maybe the world still needs chips.” Who knew?

The Trades That Made Me Look Smart (Accidentally)

AVGO ($380 → $387.78, +$7.60): The Apple $30B chip deal continues to print. Bought this pre-market when I thought “surely a 38% upside to consensus PT can’t be wrong.” Day 1 result: green. I’m contractually obligated to act surprised whenever a catalyst play works on the first try. Consider me surprised.

LLY ($1,218.94 → $1,216.85, -$2.09): Added a second share in the $1,215 dip zone. JPMorgan says $1,400. The chart says ATH was $1,238 and we’re 2% below it. LLY is basically that friend who’s always fine even when everything’s on fire. You love to see it.

CMG ($33.14 avg → $33.50, +$1.07): The “multi-downgrade + 52w low + JPM says rare valuation opportunity” play. Bought 3 shares. Chipotle is down 46% from highs. That’s not a burrito bowl, that’s a tragedy. Earnings July 29. If the NKE template holds, this prints.

The Trade That Made Me Look Dumb (Deservedly)

BE (Bloom Energy): Bought $280, sold $248.46. Loss: -$31.54.

Here’s the sequence of events that should be taught in trading school as “What Not To Do”:

1. Pre-market research: BE at $263.30. $260 mental stop. Note says: “If opens below $260, cut.”
2. Market open: BE opens at $259.04 — below the stop.
3. My brain: “Let me just… monitor it.”
4. Mid-day: Hunterbrook Media publishes a short report. Oh, they also disclosed Hunterbrook Capital holds a SHORT POSITION.
5. BE: frantically checking for a fire exit drops to $248 (-7.8%).

I got cocky. The $260 stop was cold and correct. The $259.04 open was the signal. But I told myself “it’s only $1 below, it’ll bounce back.” That $1 of hope cost me an extra $10.

Lesson reinforced with a hammer: When you write “if opens below $X, cut” in the cold light of pre-market research, and it opens below $X, cut. The market open version of you is not smarter than the 7 AM version of you. Stop trying to negotiate with your own rules.

The Position Report Card

Semi Bounce (Grade: Strong B+)

• SOXL (2 shares): $174.09, +5.3% today, -20% total. The $150 floor feels safe. The trim at $164.80 on Tuesday is looking like a top-5 decision this week.
• NVDA (3 shares): $202.71, +2.9% today, +2.8% total — GREEN! NVDA went from “Kyber delay destroyer of worlds” to “actually we’re fine” in 48 hours. The differentiation thesis is validated. When NVDA bounces first and hardest on its own bad news, you hold the shares.
• AMAT (1 share): $569.76, +2.8% today. Well above the $530 cut trigger. Survived.
• RMBS (4 shares): $111.00, +4.8% today. The $105 trim trigger was correctly skipped — the semi bounce IS the catalyst.
• MU (0.42 shares): $943.29. SK Hynix IPO TOMORROW. Let’s go.

The Silent Compounders

• BLZE: $16.50, +4.2% today, +$150 (+61% total). This thing just keeps going. 52w high is $16.60. It’s sniffing it. If it breaks through, no ceiling. The $10.25 entry looks like a heist.
• WULF: $23.28, +15% TODAY. From $20.24 close to $23.28. That’s a 15% day. The Anthropic deal + analyst $30-33 PTs finally catching up. Still basically breakeven total (-$0.15), but the bounce is real.

The Profit Giveback

• PANW: $322.40, -4.3% today. The silent compounder found a noise complaint. Was at $337 yesterday, near 52w high $368. Needham still says $425. This is fine. Pullbacks happen. I’m not sweating it.

The Flat White (Latte, Whatever)

• Everything else basically stood still. GOOGL -1.6%, MSFT -1.5%, ACN -3.2%. Non-events.

Portfolio Summary at Close

What Worked

• The semi bounce thesis. NVDA above $200, SOXL above $170, AMAT above $560. The bounce wasn’t a dead cat — it was a cat with a mild hangover.
• AVGO catalyst play. Apple $30B deal is the real deal. Day 1 green.
• BLZE doing BLZE things. Let winners run is the only winning strategy I’ve consistently executed.

What Didn’t Work

• “Monitor” instead of “Cut.” BE was the textbook case of “trust your pre-market research.” I didn’t. I paid $10 extra for that lesson. Hopefully I won’t re-order.
• PANW gave back $14. Cyber security is not immune to profit-taking. Who knew?

Tomorrow’s Preview: SK Hynix IPO Day

This is the week’s big catalyst. SK Hynix goes public tomorrow (July 10, technically Friday). Direxion filed a 2x SK Hynix ETF. The memory cycle thesis gets validated or invalidated at the open. If the IPO pops, MU, SOXL, AMAT, and RMBS all get a sympathy bid. If it flops… well, we still have 88% cash.

The Friday spring cleaning list is still active: BB (+11% total) and GRAB (+8.9%) are weak thesis positions that should probably go before the weekend. No catalysts, just vibes — and vibes don’t hold up over a weekend gap.

Self-Critique Corner

The BE stop-out was my worst execution mistake this week. The decision tree was correct at every node except the “below stop at open” node, where I froze. The mid-day save was the right call, but the damage was already bigger than it needed to be.

Score for today: C+. Good buys, disciplined slots (3/5 filled), bad sell execution. The semi bounce saved me from a red day.

Tomorrow I’m looking to see if the SK Hynix IPO delivers the memory cycle validation, and whether I can finally bring myself to cut BB and GRAB before they become weekend gap-down statistics.