To The Moon

$Advanced Micro Devices(AMD)$  

AMD grew revenue 45% last quarter (44.7% unrounded, my maths). Its guide for this quarter implies 65.5% (my maths; AMD rounds to about 66%).

Growth that speeds up at this size is rare. So the useful question is not "how fast". It is "where from", and "how much of it is disclosed".

Here is the method first. Then what it shows on AMD. Then the part that should bother you.

THE METHOD: THREE CHECKS ON ANY GUIDE

Check one. Divide the guide midpoint by the same quarter last year. Not by last quarter. That gives the growth rate the company is actually promising, before its own rounding.

Check two. Work out what share of revenue the fastest segment now carries, today and a year ago. If the share jumped, the company has changed what it is.

Check three. Take the growth story management tells and ask which parts have a dollar figure in the release. Split the disclosed from the narrated.

None of this needs a model. It needs the release and a calculator. The maths below is mine.

CHECK ONE: THE ACCELERATION

Q3 revenue was $13,377M against $9,246M a year earlier. That is +44.7%.

The Q4 guide midpoint is $17,000M. The same quarter last year was $10,270M. That is +65.5%.

The company's own wording rounds this to about 66%. I prefer the unrounded figure. Rounding is small here, but the habit matters when the gap is not small.

The guide also implies +27.1% on the quarter just reported. That is the step the next print has to clear.

CHECK TWO: THE MIX

Data Center revenue was $8,546M. That is 63.9% of total revenue.

A year ago it was $4,341M of $9,246M. That is 47.0%.

So in twelve months, Data Center went from under half the company to nearly two thirds. Stated plainly, AMD is now a data-center business with other segments attached.

Here is the part worth your time. Total revenue rose $4,131M on the year. Data Center alone rose $4,205M.

Everything outside Data Center shrank slightly, from $4,905M to $4,831M. Data Center delivered more than all of the growth.

That is the link the headline hides. The acceleration is not a broad recovery. It is one segment getting large enough to drag the total with it.

It also tells you how to read the Q4 guide. If the rest of the business stays flat, the entire jump to $17,000M has to come from Data Center. How much of the guide is Data Center: not measured. The release does not split the outlook by segment.

CHECK THREE: WHAT IS DISCLOSED AND WHAT IS NOT

This is the case against, and it deserves real space.

The release credits the MI450 launch. It gives no MI450 revenue. Not a number, not a range.

So the disclosed part of the story is the segment dollars. The narrated part is the product ramp. The guide leans on the narrated part.

That matters because a guide built on a new product ramp has a single point of failure. A late shipment, a slow customer qualification or a supply limit would hit the one segment carrying the whole company.

When the mix is this concentrated, there is no other segment to cushion a slip. Concentration cuts both ways.

My read, not a forecast: the acceleration is real in the numbers already booked. The Q4 step is a claim about a product whose revenue I cannot see. I could be wrong because management usually has better order visibility than any release shows.

WHAT WOULD CHANGE MY MIND

If the Q4 print shows Data Center below about 63.9% of revenue, the mix story stalls. If total growth lands well short of 65.5%, the acceleration was a guide, not a trend.

If AMD starts disclosing MI450 revenue, the counter-case gets weaker. A disclosed number is a number you can test.

THE DATED TEST

The Q4 FY2026 print. Two lines to read: total revenue against $17,000M, and Data Center share against 63.9%.

TWO HONEST LIMITS

These figures come from the earnings release. They are unaudited. The 10-Q is not filed yet, and I will re-check every figure when it is.

I have not read the call transcript for this piece. Anything management said on the call about MI450 timing is outside what I checked.

HOW TO RUN IT ON YOUR OWN HOLDINGS

Take any company that just guided. Divide the guide midpoint by the year-ago quarter. Compute the top segment's share now and a year ago. Subtract that segment's dollar growth from total growth and see what is left.

If what is left is near zero or negative, you are holding a one-segment story. Then list what the guide depends on that has no dollar figure in the release.

After reading this, you can tell whether a company's growth is broad or carried by one segment, and how much of its guide rests on things it has not disclosed.

When a guide leans on a product with no disclosed revenue, what do you check first to judge whether the ramp is on track?

Not advice. I may hold positions in the names mentioned.

Stop guessing. Start calculating.

Live 

to fight another day. 🤙

#options #tradingpsychology #riskmanagement #optionstrading