Nvidia Is Down 30%. Do You Own the Stock, or Does the Stock Own You?
Nvidia reports earnings on Tuesday afternoon, with the shares roughly 30% below their October high. I have no useful edge in guessing whether the stock goes up or down after the number, and I am not going to pretend otherwise. I’ve openly been bearish on the stock, and it’s going my way. Nvidia could beat expectations again and rally hard. It could produce another excellent quarter and fall because investors wanted even more. Both have happened to great companies before.
What interests me more is what happens to investors when a stock like (NVDA) starts going the wrong way.
More News from Barchart
We are most emotionally attached to stocks that have generally treated us well. They are usually the ones that make us a lot of money. (NVDA) has done that for an extraordinary number of people. It has rewarded patience, punished skeptics, and recovered from setbacks that looked serious at the time. If you owned it early enough, (NVDA) may be the best investment you have ever made. That history is valuable. It can also become a problem.
A stock that has proved you right repeatedly starts getting a different standard of evidence. Bad news needs more proof. Competition gets dismissed more quickly. Valuation becomes something other investors worry about. A disappointing quarter becomes temporary because the last one was also temporary. You are still doing analysis, but increasingly the analysis starts with the answer you want.
I have seen this many times over 35 years in markets. Once a stock has made someone serious money, selling it can begin to feel less like changing an investment decision and more like admitting something about themselves. That is where I become careful.
Great Companies Can Train Their Shareholders Badly
(NVDA) is an extraordinary business. There is no point pretending otherwise simply because the stock has fallen. The company reported $96.2 billion of revenue last quarter, more than twice the level of a year earlier, while data center revenue grew 117%. It has become one of the central pieces of the global AI infrastructure buildout. But the quality of the business has never been the only question.
I wrote earlier this year about why great businesses can still become terrible investments because price eventually matters no matter how good the company is. (MSFT) is probably the cleanest historical example. The business remained extremely good after the technology bubble. Revenue grew, free cash flow grew, and the company became more valuable economically. Yet somebody who bought near the 1999 peak spent years waiting for the business to grow into the price they had paid. The investor could be right about Microsoft and still be wrong about Microsoft stock. That sounds obvious today. It was considerably less obvious when everyone already knew that (MSFT) was one of the best companies in the world.
Cisco was similar. The internet was not a fad. (CSCO) sold equipment that helped build it. The bulls were right about the importance of networking and about the explosion of internet traffic. What they got wrong was the amount of future success already sitting inside the share price. I think AI investors should remember that. Nvidia can remain one of the most important companies in the world and still produce periods when the stock is a poor investment. The two ideas do not contradict each other.
In August I wrote about (NVDA) helping mobilize more than $500 billion of outside capital for AI infrastructure. My concern was not that AI demand was fake. It was almost the opposite. The infrastructure needed to meet that demand had grown so large that I wanted to understand who was funding it and what returns would eventually justify it.
Then (NVDA) disclosed $279 billion of commitments to future supply and manufacturing capacity. Again, that was not evidence of a broken business. It showed just how confidently (NVDA) was positioning itself for demand several years into the future. The company is no longer simply filling today’s orders. It is making enormous decisions about what AI demand will look like later this decade. Those are questions shareholders should want to discuss. The moment you find yourself getting irritated because somebody raises them, I would pay attention to that reaction.
The Problem With a Stock That Has Always Come Back
There is a peculiar psychological trap with exceptional winners. Every recovery makes the next decline easier to dismiss. Nvidia fell before. It came back. Nvidia looked expensive before. Earnings caught up. Competitors appeared before. (NVDA) stayed ahead. Eventually the historical pattern becomes the investment thesis.
That is dangerous because businesses do not usually change in one dramatic moment. The evidence accumulates slowly. A customer starts developing its own chip. Capital spending moves higher. Financing becomes more complicated. Margins stop expanding. A competitor improves a little a little better. None of those things necessarily destroys the thesis, but if you love the stock, you can explain each one away separately and never stop to ask whether the collection of evidence has changed anything.
Nvidia’s largest customers illustrate the issue nicely. (AMZN), (GOOG), (MSFT), and (META) have enormous incentives to develop their own silicon because Nvidia earns extraordinary economics from them. I have written about this topic before, including Nvidia’s move deeper into the developer ecosystem through its roughly $13 billion Hugging Face transaction. Nvidia clearly understands that its moat must keep moving beyond the chip itself into software, networking, systems, and developer relationships. I don’t read that as Nvidia losing. I read it as capitalism working.
When a supplier earns exceptional margins, customers eventually look for ways to keep more of those economics for themselves. Nvidia knows it. Its customers know it. Shareholders should know it too.
(NKE) offers another version of this problem. For years investors could rely on the brand, distribution, athlete relationships, and a history of premium economics. Then the evidence started changing. China weakened. Hoka and On became serious competitors. Product innovation became less convincing, and management had to repair distribution decisions it had made only a few years earlier. I have written repeatedly about Nike because the investment question kept changing even though the logo did not.
That is a useful lesson. Familiarity with a company is not the same as understanding what the company is worth today.
I Would Ask Nvidia Shareholders One Question
Imagine you wake up tomorrow and your (NVDA) position has disappeared. There is no tax consequence, no trading cost, and no mistake at the broker. The exact value of the position is sitting in cash in your account. Would you buy Nvidia back? More importantly, would you buy the same amount? I like this exercise because it removes history. You no longer own the stock because you bought it at $40, $100, or $180. You cannot tell yourself you are “playing with house money” because there is no such thing. Every dollar in the account is yours today. Now you need to make the investment decision again.
What do you think Nvidia earns in three years? What happens to margins? How much AI infrastructure eventually produces acceptable returns for the companies buying it? What multiple are you willing to pay? What would make you admit that your assumptions were wrong? Those questions are much harder than saying that Nvidia is down 30%.
I would do the same exercise with any stock that has become important to you. (AAPL). (TSLA). (AMZN). (NKE). (MSFT). Whatever it is. The more money it has made you, the more useful the exercise becomes because successful investments create emotional capital alongside financial capital.
There is nothing wrong with conviction. You need it. Some of the best money I have made came from being willing to sit with an investment when other people disagreed with me.
But conviction should come from continuing to test the facts, not from remembering how right you were last time.
Nvidia’s earnings tomorrow will give investors another batch of information. Revenue, margins, demand, supply commitments, and management’s outlook will all matter. The share price reaction will matter too, particularly if good news no longer produces the response investors have become accustomed to.
What I would not do is decide in advance that a fall means the market is wrong or that a rally proves the thesis is intact.
(NVDA) does not know that you own it. It does not know your cost basis, and it has no obligation to get you back to the old high. The only useful question is whether the future economics justify today’s price.
The hardest stock to sell is often the one that made you feel smartest. That is precisely why it deserves the hardest questions.
On the date of publication, Jim Osman did not directly have a position in any of the securities mentioned in this article. All information and data in this article is solely for informational purposes. This article was originally published on Barchart.com