To The Moon

$Western Digital(WDC)$  

Mathematical Money | October 13, 2026

The KOSPI broke below 6,000 today, down more than 7% intraday and triggering a sell-side trading curb for the first time in two months. SK Hynix fell over 9%. Samsung Electronics over 8%.

$WDC$ is down about 6% as I write, at roughly $362 against Monday's close of $387. I own the shares, I own long-dated calls, and I wrote about this sector eight days ago. So I went and read what actually happened rather than assuming the move told me something about my position.

It mostly doesn't, but the part that does is uncomfortable, and I would rather set out both than pick the half that suits the position.

What actually happened in Korea

The trigger was CXMT, a Chinese memory maker, which announced progress on a new memory-cell design that packs more storage into less physical space. Samsung, SK Hynix and Micron together hold roughly 90% of the global DRAM market, so a credible Chinese entrant with a density advantage is a direct threat to all three.

On top of that, Broadcom's chief executive described the current AI investment environment as "casino-like", which knocked sentiment across everything AI-adjacent on the same day. So the selloff has two components: a specific competitive threat to DRAM, and a general wobble in AI enthusiasm. Neither of them is about hard drives.

Western Digital is in neither of those businesses

This is the part most of the commentary is skipping. Following the SanDisk spin-off, Western Digital is a pure-play hard disk drive company. It does not make DRAM. It does not make NAND flash any more either.

Its competitors are Seagate and Toshiba, not Samsung and SK Hynix. Its product sits in the bulk nearline tier, where hyperscalers store the data that is too large to justify putting on expensive flash. A Chinese memory maker improving DRAM cell density is, on the face of it, somebody else's problem entirely. And yet the stock fell 6%, because that is what happens when a sector gets sold and an index does not read filings.

The check that proves it

I ran the same test I ran eight days ago and it gave the same answer. Over the last five sessions: Western Digital −12.4%. Seagate −15.1%. Micron −7.0%. Three companies being sold hard.

NetApp +6.9%, and it closed at a 52-week high. NetApp buys drives and sells storage systems built around them. If this were a demand problem for stored data, NetApp would be falling hardest of the four. It is making new highs instead. The market is repricing who gets to keep the margin in the supply chain, not whether the data exists.

That was the finding last week and it has survived a second, completely unrelated shock. I find that more persuasive than I did the first time.

The part that is genuinely uncomfortable

Here is the honest transmission channel, and it is why I am not dismissing today entirely. Cheaper flash substitutes for hard drives at the margin. If Chinese competition pushes memory and NAND prices down structurally, some workloads that would have sat on high-capacity HDDs move to flash instead. It does not threaten the bulk archive tier, where the cost-per-terabyte gap is still enormous. It does compress the boundary between the two.

So the correct reading is not that today has nothing to do with Western Digital. It is that today has very little to do with its current quarter and something real to do with its 2029 one. That is a slower and smaller problem than a 6% day implies, and it is not nothing.

Where that leaves my position, stated plainly

The shares are down about 10% against my cost. The long-dated calls are down about 29%. That second number is the one that matters and I will not dress it up. These are January 2028 calls, and 70% of what they are currently worth is time premium rather than intrinsic value. For the position to break even the stock needs to rise roughly 43% from here before expiry. Volatility on this name is running near 80% annualised, so that is not an absurd ask, but it is a long way and the clock is running.

The short calls I wrote at $405 expire on Friday and will do so worthless. That is the one part of this that worked exactly as intended, and it is a reminder that the income leg keeps paying while the directional leg is losing.

What I'm actually doing

Nothing today. Selling into the second day of a sector-wide derating, on news about a different product category, is how a drawdown becomes a realised loss.

I will keep writing calls, and the rule stays the same as everywhere else: the strike plus the premium has to clear my cost. On the shares that means anything above about $402 sells at a profit. On the long-dated calls it is harder, because the strikes that pay enough premium to offset the time decay sit well below the level I need the stock to reach.

I am not adding. The position is already one of the larger ones in the book and conviction is not a reason to concentrate further into a falling knife.

The thing I will actually watch is not the share price, it is whether hyperscaler HDD orders change. Western Digital has described capacity visibility with customer agreements reaching into 2029. If those hold, today was sentiment. If they start getting renegotiated, today was early.

My view of the business has not changed today. The arithmetic on my long-dated calls has got materially worse regardless, and I notice those two things get conflated constantly, by me as much as anyone. A damaged structure and a broken thesis look identical on a red screen and they call for completely different responses.

What I would genuinely like to know is how far down a supply chain other people check before deciding a selloff applies to them. Twice in eight days the customer has told me something the headline didn't, and I only started looking because I got it wrong the first time.

#WDC #memory #storage #semiconductors #options

Stop guessing. Start calculating.

Live to fight another day. 🤙