I was pleasantly surprised at the relatively strong performance of NVDA in the pre-market yesterday when I was writing about the favorable market reaction to NVDA’s quarterly earnings announcement after the market close on Wednesday. That glow didn’t last very long after the opening bell, however, with the stock trading down by about 8.5% by the close–dragging other AI-related stocks down with it as well.

the issues

—earnings momentum. If a company posts quarterly earnings results that are getting sequentially stronger, meaning, say, +10% yoy, followed by +12%, +14%, +16%, two factors tend to propel the stock upward. They are: the earnings gains themselves and, crucially, the pattern of accelerating eps growth. This second is typically what causes the stock’s PE to expand.

The opposite of this last pattern, where eps growth is decelerating, typically causes the PE to contract. That’s the issue with NVDA.

—what everybody knows. NVDA has excellent financial disclosure, in my view. To anyone who has taken the briefest glance at the financials, the issue of decelerating earnings growth momentum has been clear for six months or more.

So I was especially interested in the latest NVDA results because I thought they would show how much of the multiple contraction issue had already been factored into the stock price. After all, NVDA had peaked at $150+ in early January and fallen by 18% since. So maybe this is what that decline was all about.

The pre-market yesterday seemed to me to be saying that, yes, investors had read the financials, knew that most of the company’s operating leverage had already been exhausted–and that therefore operating earnings would no longer be rising much more quickly than sales.

Regular trading yesterday said the opposite–that the decelerating eps growth momentum still surprised a significant number of holders.

Oh, well.

The issue now, as far as I’m concerned, is when to rebuy the shares I sold a few months ago. I don’t think there’s any rush, though.