← sources

Nvidia is so weird

source: reddit (reddit:ValueInvesting)published 2026-11-15 21:17ingested 2026-11-16 02:39

retained text

So, according to just about all the fundamentals there are, Nvidia is massively undervalued. They have a current P/E of 21 and a forward P/E of around 10, while being projected to grow by 90% this year and 70% next year.

Since the start of 2025, they have absolutely crushed earnings every single time, and yet the stock has barely moved. The market does not seem to care about Nvidia crushing earnings because it's focused on whether AI is a bubble or not. Yet AMD, Intel, and other AI plays with smaller market caps have exploded on good earnings. 

Whenever Nvidia's stock starts to grow, it seems to be held back by its insane market cap. It almost seems as though the market has simply decided that Nvidia, no matter the fundamentals, can't have a substantially larger market cap than whichever company is the second most valuable at the time. Yet Nvidia seems to have so many more risks associated with it compared to Apple, Google, Microsoft, etc. 

Like, they haven't moved despite smashing earnings time and time again. At the start of this year, estimates for EPS and revenue growth were at 50% for the current year and 35% for next year. Those numbers have since increased to 90% for the current year and 70% for next year, yet the stock has barely even moved during this period. Even right before all the Jensen stuff became public, it was trading lower than at the start of 2026, despite SOXX having moved 45% YTD.

Now, the stock has barely moved for two years despite crushing earnings. Just imagine what would happen if they have just one slightly bad earnings report. Similarly, what if LLMs struggle to grow? This would obviously have a massive impact on NVIDIA. Unless robotics and self driving vehciles really start to become a thing in the mean time. Still currently the only way current spending on data centers makes any sense today is if those LLMs explode in earnings over the next couple of years.

In addition to all this, you have increased competition, with Anthropic and OpenAI starting to make their own chips, as well as all the other Mag 7 companies. There is also a potential AI may start to erode the CUDA moat. There are just so many risks currently, and short term the downside seems to be so much bigger than the upside. 

I don't get why anyone would invest in NVIDA given all this. Like there seems to be huge risk-reward inbalance going forward. It seems earnings for Nvidia over the next couple of quarters will only matter if they are bad. 

And yet, if you can't justify investing in nvidia at current valuations, I have no idea how you can justify investing on any other SOXX stock.

Nvidia has become one of the major players when it comes to self driving vehicles, and pretty much  the undisputable leader when it comes to robotics. Both of which are markets that may grow to be even bigger then the LLM market will ever be, and still they are the company trading at the lowest mulitples across the entire industry. Not to mention the CUDA-moat and how essential it still is to training. I dont get how any of this makes any sense? Like at one hand it seems desperately undervalued compared to just about any other stock in the industry and on the other hand its already worth 4 trillion dollars.

document fetched 2026-11-16 00:38 · canonical https://reddit.com/r/ValueInvesting/comments/1xx70th/nvidia_is_so_weird