Short AMD via optionsWhat I did: Long Oct 4 vertical 145/135 AMD put spread @ 2.06, as a swing trade.
The rationale:
NASDAQ:AMD seems to struggle with the 3-month VWMA (white line), while in the context of a declining 3-month and 1-month trend.
Fundamentally, AMD competes with NASDAQ:INTC on CPUs and with NASDAQ:NVDA on graphics cards. Of the three, NVDA is the most profitable by far, with a 10-yr median FCF margin of 27.5% (INTC 19.3%, AMD 2.1%).
In terms of 10-yr revenue CAGR, NVDA is also the most compelling with 30.9% (INTC 0.3% ,AMD 15.6%).
Despite this, AMD does not offer a significant valuation discount relative to NVDA. AMD's next-twelve-month estimated PE multiple is 35, just as NVDA's. And in terms of forward EV/EBITDA, AMD is currently currently MORE expensive, with a 33.1 multiple vs. 29.4 for NVDA. INTC is at this point unprofitable, and at best a turnaround candidate, so I am excluding it from further consideration.
I do understand that NVDA's large exposure to the AI infrastructure build-out presents unique downside potential once that wave will have run its course. But a temporary benefit is still preferable to no benefit at all, so I think as long as NVDA grows faster and operates more profitably than AMD, its valuation multiples should present an upper bound to AMD's. In other words, unless NVDA can progress to new highs, I don't see AMD going up. And, as we have seen today, NVDA is struggling to recapture its past highs.
I believe that AMD is susceptible here to significant volatility from its association to NVDA. A full reversal of trailing 10-day swing back towards 133+/- seems quite possible, especially since I see the market anticipate a lot of good news from inflation/Fed direction. While I do believe that the Fed will come through with a 50bp cut on Wednesday, I think that cut will be justified with a long discussion of poor recent economic data, casting doubt on the feasibility of a "soft landing".
My trade is simply a swing with the intent of hedging some of my long exposures. But it also exists in a larger context: For the last couple of years the argument has been made that semiconductors are no longer cyclical stocks, but rather idiosyncratic growth machines, like software. If that narrative gets challenged by reality, then I think AMD has a long way down. Its current multiples are one third above trough valuations, and if analysts were to revise down their estimates, then the downside grows in proportion.