MU & SNDK Update: Where We Are Tuesday Night, and What's Actually Next

Quick update on MU and SNDK now that the dust has settled on Tuesday's session.

Micron closed near the lows of the gap, SanDisk gave back roughly 200 points and stayed there, and the rest of the U.S. memory complex (WDC, STX) traded with them. The Korean session that started this — KOSPI down 10%, Samsung and SK Hynix liquidated, halt tripped — did not reverse into the close. That is the relevant overnight signal heading into Micron's print.

Below: where the two names actually closed, what the options tape is pricing, the three scenarios for Wednesday after the bell, and how I'm sized.

MU finished the day in the lower third of the intraday range after gapping down roughly 100 points at the open. There was a midday bounce attempt that failed, and the close on the low end tells you the marginal seller was not done. Volume was a multiple of the 30-day average. That is a real distribution day, not noise.

SNDK was worse on a percentage basis. After the ~200-point gap, it traded sideways in a wide range and closed near the middle of that range, which is technically constructive relative to MU but only because the damage had already been done at the open. Short interest in SNDK had been climbing into the print already; today's move likely shook out some weak longs and gave shorts a level to defend.

Across the complex: WDC down meaningfully, STX down less, the HBM-adjacent equipment names (LRCX, AMAT, KLAC) down in sympathy but on lighter volume. The dispersion tells you the market is differentiating — this is concentrated in pure-play memory, not the entire semi cap-equipment chain.

The MU weekly straddle going into Wednesday's print is pricing one of the larger implied moves of any single-name catalyst this earnings season. That is partially elevated baseline vol because of the macro tape, and partially the market correctly identifying that the print is now a binary catalyst rather than an incremental one.

What is more interesting is the skew. Downside puts on MU got bid harder than upside calls on the gap-down, which is normal — but the put skew has stayed bid into the close. That tells you the marginal options buyer is still hedging downside, not chasing upside. In a healthy setup heading into a print, you would expect at least some call buying as people try to play a relief rally. We are not seeing that yet.

SNDK options are less liquid but the same story shows up in spread pricing: downside protection is bid, upside is offered. The market is positioned defensively into the print, which is itself a small contrarian tell — the most painful outcome for the consensus positioning is now an upside surprise.

The single most important macro datapoint between now and Wednesday's print is not a U.S. data release. It is whether Samsung Electronics and SK Hynix stabilize in the Tuesday-into-Wednesday Asian session. If those two names bounce hard overnight, the Tuesday move in U.S. memory was a one-day cascade and MU walks into the print with the wind at its back. If Korea opens weak again and follows through, the U.S. memory tape opens offered, and Micron is reporting into a tape that is actively demanding reassurance.

Watch the KOSPI open, watch SK Hynix specifically, and watch the won. A weaker won historically correlates with foreign selling of Korean tech, which is exactly what happened today.

Scenario 1 — Clean beat and raise. MU prints in-line to slightly better on Q3, guides Q4 gross margin above the 81% level, and management language on HBM pricing is firm ('continuing to expand,' 'demand exceeds supply through 2027,' 'capex tied to committed customer agreements'). In this case the stock gaps up 8–15% after hours, SNDK follows in sympathy, and the broader AI complex gets a short-cover relief rally into Friday. The Tuesday gap becomes the reference low and gets bought on any pullback.

Scenario 2 — In-line and cautious. MU prints in-line, guides Q4 gross margin roughly flat at the 81% level, and management uses softer language ('we expect pricing to remain firm at current levels,' 'mix optimization between HBM and conventional DRAM,' 'visibility into early 2027'). The stock chops in a wide after-hours range, probably finishes within a few percent of the print level, and the trade becomes a tape-watch through Thursday. This is the most likely outcome and the one the options market is least prepared for, because both wings of the straddle lose.

Scenario 3 — Disappointment. MU prints in-line on the headline but guides Q4 gross margin below 81%, or mentions 'pricing moderation,' 'customer mix headwinds,' or pulls forward FY27 capex without clear demand attribution. In this case the stock gaps down another 10–20% after hours, SNDK follows down hard, the broader AI-momentum unwind extends into Thursday, and the rotation trades (gold, energy, utilities, healthcare) that worked Tuesday extend with it. This is the lowest-probability outcome but the one with the largest tape consequence.

One sentence on HBM4 yield ramp. Anything that confirms HBM4 is moving from sampling to volume on schedule keeps the bull case intact. Any hedging language about 'qualification timelines' is a yellow flag.

The bit-growth versus pricing trade-off. Memory companies always have a choice between selling more bits at lower prices or fewer bits at higher prices. The healthiest answer is 'pricing strength supporting both bit growth and ASP expansion.' If management starts talking about volume offsetting price, that is the cycle turning.

Customer concentration commentary. If management mentions a strategic agreement with a named hyperscaler that backstops the capex step-up, that is the bull-case anchor. If they decline to break out commitments by customer, the market will assume the capex is more speculative than they are letting on.

Inventory days. A sequential decline in finished-goods inventory days is bullish. A sequential build, even small, is a tell that demand at the current price level is starting to soften.

SNDK is the cleaner derivative trade. It does not report Wednesday, but it trades on the read-through. If Micron confirms HBM and NAND pricing are both holding, SNDK should reclaim a meaningful piece of the Tuesday gap on Thursday morning. If Micron disappoints, SNDK has further to fall because its multiple has expanded faster than its earnings revisions over the past three months.

The specific level on SNDK is the Tuesday low. A lower low after the Micron print confirms the unwind is broadening from MU into the rest of the NAND complex. Holding the Tuesday low on a sympathy reaction is the cleanest signal that the worst is already priced in.

Smaller than I would be on a typical setup. The gap into the print has already collapsed the asymmetric risk-reward that existed at Monday's close. I am holding a modest long in MU into the print, paired with downside protection I bought when implied vol was lower last week. I am flat SNDK into the print and will react Thursday morning rather than guess Wednesday night.

If Scenario 1 plays out, I add on the first 30-minute pullback Thursday rather than chase the after-hours move. If Scenario 2 plays out, I do nothing — the trade has not been given to me yet. If Scenario 3 plays out, I close the MU long, let my puts run, and rotate the proceeds into the rotation winners (gold miners, energy, defensive utilities) that the Tuesday tape already identified as the destination for the unwind flow.

Position sizing matters more than direction here. Anyone who is full size in either direction into a binary catalyst with this much overnight macro risk is taking on risk that the expected value does not justify. That is the single most important sentence in this update.

MU and SNDK closed Tuesday near the lows on heavy volume after a forced Korean cascade liquidated Samsung and SK Hynix. The options market is positioned defensively into Wednesday's Micron print, which makes an upside surprise the most painful tape outcome. Three scenarios — clean beat, in-line cautious, disappointment — each have a specific trade plan, and the in-line cautious case is both the most likely and the least well-priced.

I am holding a small MU long with downside protection into the print and waiting to react to SNDK Thursday morning. The most important thing right now is not picking a direction. It is being sized to be wrong and still in business Thursday at the open.

This article is a personal opinion piece by Guy Gentile. It is not investment advice, a research report, an offer to buy or sell securities, or a recommendation. The author may at any time be long, short, or flat positions in MU, SNDK, WDC, STX, related semiconductor names, options, or hedges, and may change those positions without notice. Do your own work.