Nvidia stock is not obviously cheap, but it may be cheaper than its own history and cheaper than the earnings power the market is assigning to it. That is the...
Is Nvidia Stock Too Cheap Right Now? Why Wall Street Thinks NVDA May Be Undervalued Before Earnings
Nvidia stock is not obviously cheap, but it may be cheaper than its own history and cheaper than the earnings power the market is assigning to it. That is the real argument before earnings on August 26: Wall Street is not asking whether Nvidia is expensive in the abstract; it is asking whether the company’s growth, margins, and AI chip demand justify a valuation that still may lag its fundamentals. A price target like Bank of America’s $350 says the market may be underpricing the next leg of earnings power.
Key Takeaways- Nvidia’s valuation looks rich on plain old earnings multiples, but that can miss the point if growth remains explosive.
- Bank of America’s case is simple: the stock may be trading at a discount to peers on a forward basis, despite Nvidia’s stronger moat.
- The real issue is not whether AI is real; it is whether AI spending keeps flowing into Nvidia’s chips, software stack, and networking gear.
- Earnings week can expose the gap between story and numbers, which is where many investors get sloppy.
- A sane investor should think in terms of valuation, growth, margins, cash flow, and portfolio position size, not headline hype.
What is Nvidia stock, really?
Nvidia stock represents ownership in a company that designs the chips, systems, and software that power much of modern AI computing. That sounds clean. It isn’t. The equity is really a claim on several moving parts at once: graphics processing units, data center networking, software tools, and the huge spending cycle from cloud providers, model developers, and enterprises trying to catch up.
Most people reduce Nvidia to “the AI stock.” That is lazy. Nvidia is better understood as an infrastructure supplier for high-end compute, with a moat built from hardware performance, software compatibility, developer familiarity, and a supply chain that has been hard to match. When I run the numbers, the market rarely prices these things correctly during a surge cycle. It tends to either overpay for stories or underprice sustained compounding until the facts are impossible to ignore.
So is Nvidia cheap? Not in the ordinary sense. On trailing measures, it still looks expensive to investors who screen for low P/E ratios and move on. But that is an amateur’s shortcut. For a company growing this fast, forward earnings, revenue growth, gross margin durability, and free cash flow conversion matter more than a backward-looking multiple. The truth is, a company compounding at this pace can look pricey for years while still creating real shareholder value.
That said, there is no magic here. If AI capex slows, if customers stretch deployments, or if competitors chip away at pricing, the valuation can compress fast. Markets are rude that way. A stock can be both high quality and vulnerable if investors pay too much for the future. This is why the question is not “Is Nvidia good?” It clearly is. The real question is “Is the market still underestimating how much good Nvidia can do?”
For a broader framework on how stocks are judged against the rest of the market, see our guide to stock valuation basics. If you want the mechanics behind capital allocation, the article on portfolio diversification helps. And if you are comparing AI stocks, our piece on AI investing strategy gives a cleaner lens than trading chatter.
Core benefits and why investors care
Nvidia matters because it has the sort of operating leverage most companies only dream about. Revenue can rise fast while margins stay strong, and that creates a compounding effect that shows up in cash flow, market share, and investor psychology. Frankly, that is why this stock keeps attracting both bulls and skeptics.
- Dominant AI hardware position: Nvidia’s GPUs remain central to training and inference workloads. That gives it pricing power, and pricing power is what separates a good business from a fragile one.
- Software and ecosystem moat: CUDA and related tools keep developers inside Nvidia’s orbit. This is not a trivial detail. Switching costs matter.
- Scale in data center demand: Hyperscalers, labs, and enterprises need massive compute, and Nvidia sits near the front of that spending stream.
- Strong free cash flow potential: Growth stocks that actually throw off cash are a different breed. They can fund R&D, buffer cycles, and support repurchases.
- Portfolio diversification role: For some investors, Nvidia offers exposure to AI infrastructure without buying a basket of weaker names that look “cheap” for a reason.
Here’s the kicker: a lot of investors confuse “not cheap” with “not attractive.” That is sloppy thinking. A stock can be expensive and still be the best house on the block. The trick is not to overpay for momentum. I’ve looked at enough portfolios to know that the biggest mistake is often the same one on repeat: people buy the trend after the easy money is gone, then act surprised when the multiple cools off.
Still, there is a reason Bank of America’s argument has teeth. If Nvidia is truly trading at a discount relative to peers on a forward basis, then the market may be using too conservative an earnings path. That matters because valuation is not a static label. It moves with earnings revisions, margin expansion, and sentiment. If the company keeps beating and raising, the “too expensive” crowd has to keep moving the goalposts.
From a stewardship angle, there is a moral layer here too, though nobody likes to say it plainly. Capital should be directed toward productive work, useful innovation, and businesses that create value for people, not just speculation for its own sake. Nvidia’s chips support real systems, real labor, and real productivity. That does not make the stock automatically cheap. It does mean the business sits on firmer ground than many market darlings.
Step-by-step: how to think about Nvidia before earnings
Do not guess.
That sounds obvious, yet most investors still do it, especially before earnings when excitement rises and discipline goes out the window. If you want to judge Nvidia properly, you need a process, not a vibe. I’ve analyzed hundreds of earnings setups, and the ones that work best use the same plain framework every time.
- Start with revenue growth expectations Look at the market’s current forecast for data center revenue, total revenue growth, and next-quarter guidance. If consensus is already sky-high, the stock needs a very strong beat to move much.
- Check gross margin trends Nvidia’s business is not just about units sold. It is about what those units earn after costs. If margins stay strong or expand, the company is still holding pricing power.
- Read the capex cycle carefully Hyperscalers are not charities. They spend when returns look good. If cloud providers keep increasing AI infrastructure budgets, Nvidia benefits. If they pause, the stock can wobble.
- Separate earnings quality from headline EPS One-time tax items, share count changes, and accounting noise can distort the picture. Focus on cash flow, not just earnings per share. Numbers can lie. Cash is harder to fake.
- Compare valuation to growth, not to mood A high multiple is less troubling when growth is still steep and the moat is intact. Compare Nvidia against other semiconductor names, but also against its own revenue and profit growth.
- Size the position properly This is where most people trip. Even a great business should not become a reckless chunk of your portfolio. Diversification is not cowardice. It is prudence.
- Decide in advance what would change your mind If guidance weakens, if margins compress, or if customer demand shifts, you need a rule. Otherwise you are just making emotional noise in a new costume.
When I test investment decisions, I care less about the headline and more about the probability distribution around the next four quarters. That is the real game. Not drama. Not message-board folklore. Probability.
If you want a cleaner framework for allocation decisions, our article on asset allocation is worth your time. For risk control, the guide to risk management in investing explains why good positions can still produce bad outcomes if sized poorly.
Comparison table: Nvidia versus AMD
AMD is the obvious comparison. It is the company investors reach for when they want a cheaper semiconductor name tied to AI and data center growth. But cheaper is not the same as better.
| Metric | Nvidia (NVDA) | AMD |
|---|
| AI leadership | Market leader in AI GPUs and systems | Strong challenger, but still behind in many AI workloads |
| Software ecosystem | Deep developer moat with CUDA | Improving, but less entrenched |
| Gross margin profile | Typically stronger | Lower on average |
| Market perception | Premium quality, high expectations | Often seen as the value alternative |
| AI infrastructure exposure | Broad, across chips, networking, systems | Meaningful, but narrower |
| Valuation case | Can look cheap on forward growth-adjusted measures | Can look cheaper on simple multiples |
| Earnings sensitivity | High, because expectations are huge | Also high, but with less pricing power |
The comparison looks simple on paper. It is not.
AMD may trade at a lower multiple, but the market often discounts it for a reason: weaker pricing power, less software lock-in, and a smaller share of the AI spend that matters most today. Nvidia’s premium exists because the business has earned it. That said, a premium can still become excessive if the growth curve bends down. That is why valuation debates should always be tied to earnings revision risk, not just a spreadsheet snapshot.
BofA’s view that Nvidia may be undervalued relative to peers suggests the market could be misclassifying Nvidia as merely a “big winner” rather than the dominant platform. If that is true, then the peer comparison matters less than the size of the AI capex pool. The real question is who captures the largest share of that pool over time.
For context on semiconductor competition and chip cycles, our breakdown of semiconductor stocks gives a fuller picture than ticker-watching ever will.
Common mistakes investors make
People make this harder than it is.
The biggest mistake is treating Nvidia like a coin flip between “cheap” and “expensive,” when the better question is whether the future cash flows justify the current price. Markets do not reward laziness, and they especially do not reward investors who confuse a hot stock with a dead-simple trade.
- Using only trailing P/E: This is the classic trap. Trailing earnings can be stale when growth is moving fast.
- Ignoring guidance: Earnings beats are nice, but guidance drives the next move. If the outlook disappoints, the stock can fall even after strong results.
- Buying after the headline frenzy: By the time everyone on TV says the stock is amazing, the easy upside may already be gone.
- Overconcentrating: A great stock is still one stock. If it goes wrong, your portfolio does too.
- Treating AI demand as infinite: It is not. Spending cycles can pause, digest, or shift to other vendors.
- Assuming moats never erode: They do. Sometimes slowly, sometimes all at once.
Let’s be real: the internet is full of people who confuse conviction with certainty. That is not analysis. It is noise. A better approach is to ask what must be true for the bull case to work, then compare that with what must be true for the bear case to matter.
The bear case is not insane. If Nvidia’s growth normalizes, if margins cool, or if customer concentration creates pressure, the valuation can compress hard. That is the price of owning a leader with a large market cap and heavy expectations. The more people expect perfection, the less room there is for sloppy execution.
The same prudence applies to investor behavior. A good business should be treated with respect, not worship. That is especially true for anyone managing household savings, retirement accounts, or charitable capital. Stewardship means weighing return against risk, and profit against purpose. Greed is a lousy allocator.
Frequently asked questions
Is Nvidia stock too expensive before earnings?
It may look expensive on trailing earnings, but that is not the whole story. If growth, margins, and free cash flow remain strong, the stock can still be reasonably valued on a forward basis. The market cares more about what Nvidia earns next year than what it earned last quarter.
Why do analysts keep raising Nvidia price targets?
Because earnings estimates keep moving higher. When a company beats, raises, and holds its margin profile, analysts often have to revise models upward. That does not mean every target is right. It means the business continues to outperform prior assumptions.
What could cause Nvidia stock to fall after earnings?
Weak guidance, slowing data center demand, margin pressure, or signs that AI spending is pausing. Stocks priced for perfection can drop even after a strong quarter if expectations were too high. The market is unforgiving. Always has been.
Is Nvidia better than AMD as an AI investment?
For now, Nvidia has the stronger moat, deeper software stack, and greater share of AI infrastructure spending. AMD may offer a lower valuation, but lower valuation alone is not enough. Investors should compare earnings power, execution risk, and portfolio fit, not just ticker price.
Nvidia is not a bargain-bin stock. It is something subtler and, frankly, more interesting: a dominant business whose valuation may still lag its actual earning power. That can happen when a company grows so fast that investors keep assuming the next quarter will be the last great one. They keep waiting for gravity to show up, and sometimes it does not.
When I look at Nvidia ahead of earnings, I do not ask whether it is “cheap” in the generic sense. That question is too crude. I ask whether the market is underestimating the durability of demand, the strength of the moat, and the size of the cash flows still to come. If Bank of America is right, the gap between price and value may be wider than the crowd thinks.
But do not get cute. A good stock can still be a bad purchase if you pay a silly price, and no amount of AI excitement changes that. Keep the position sized with discipline, compare it against the rest of your portfolio, and remember that capital is a tool. Use it carefully. In markets, as in life, stewardship beats swagger.