Why I Keep Datadog, Vertiv, and Arista Off My Radar

The core of value investing is simple, yet often overlooked in the heat of a market cycle. It demands two things: a durable competitive advantage, or what we call a "moat," and a margin of safety between the market price and your estimate of intrinsic value. A wide moat protects a company's economic profits from rivals, while a margin of safety cushions the investor against errors in judgment or unforeseen downturns. Without both, you're not investing; you're speculating.

This discipline is a bulwark against a common trap: mistaking a low valuation for a bargain. As the concept of a "value trap" illustrates, a stock can trade at a low multiple for years because the business itself is deteriorating or facing structural headwinds. The market's price is often the wisest verdict, reflecting a lack of confidence in the company's ability to generate future cash flows. True value requires a business that is not just cheap, but fundamentally sound and capable of compounding over the long term.

A powerful reality check for any investor, especially when considering high-multiple growth stocks, is Charlie Munger's 50% drop test. The late Berkshire Hathaway vice chair argued that if you cannot stomach a 50% decline in your portfolio, you are not fit to be a common shareholder. This isn't about predicting crashes, but about confronting your own risk tolerance. The test separates those who can endure the volatility inherent in owning a business from those who will panic and sell at the worst possible time. History shows even the greatest companies, from Berkshire to Amazon, have weathered such drops. The point is to ensure your portfolio's composition aligns with your emotional capacity to hold through the inevitable turbulence.

This brings us to the current market environment. The tech sector is undergoing a profound repricing as investors grapple with the long-term implications of artificial intelligence. The "picking of winners and losers" has begun, and the stakes are high. In this fog, the risk of misjudging a company's economic durability and its ability to generate cash flows for decades increases significantly. A stock's current price may be driven more by AI hype than by a clear, sustainable path to profits. For the value investor, this volatility is noise. The focus must remain on the underlying business: does it possess a wide, defensible moat, and is its current price offering a sufficient margin of safety against the uncertainty of the AI-driven future?

Assessing the Favorites: Business Quality and Valuation

The three stocks favored by Wall Street analysts present a classic tension for the value investor. Each operates in the AI-driven tech cycle, but their underlying business quality and current valuations tell different stories. Let's apply the filter of durable moat and margin of safety.

Datadog exemplifies a company with a clear path to profitability. The firm targets an adjusted operating margin of over 25%, a milestone that signals a transition from pure growth to cash-generating machine. Its strength lies in a wide moat: the company's platform offers unique, contextual data that helps enterprises manage the rising complexity of AI-integrated IT stacks. This creates a powerful "land and expand" dynamic, with significant room to grow, especially in security, where it currently captures only a small fraction of large customer revenue. Yet, the sustainability of its AI-driven demand is a question. The business is also heavily concentrated, with 70% of its customers generating over $1 million in annual recurring revenue. While this indicates high-value relationships, it also means the company's fortunes are tied to a relatively small pool of clients. For a value investor, this concentration is a moat on one side and a vulnerability on the other.

Vertiv, meanwhile, provides essential services for the AI infrastructure build-out, particularly power and cooling solutions for data centers. Its recent results were strong, with organic orders surging 252% in the fourth quarter. However, its business model is fundamentally narrow and highly cyclical. The company's fortunes rise and fall with the capital expenditure cycles of its large customers, which are themselves subject to the volatile nature of tech investment. This creates a narrow moat; while Vertiv is a necessary vendor, it does not possess the pricing power or technological differentiation that would insulate it from downturns. The value proposition here is tied to a cyclical peak, not a durable competitive advantage.

Arista Networks presents a different challenge. The company has built a powerful moat in AI networking, becoming a critical supplier for major cloud providers. This positions it at the heart of a powerful secular trend. Yet, its valuation already prices in near-perfect execution for years. The market is paying a premium for a company that is executing flawlessly in a high-growth niche. This leaves little room for error. For a value investor, this is the essence of a "buy the rumor, sell the news" setup. The company's quality is undeniable, but the margin of safety has been compressed to almost nothing. The price today demands flawless growth, which is a difficult standard to meet over the long term.

In each case, the analyst consensus sees a winner. The value investor must ask if the price reflects a margin of safety. For Datadog, the moat is wide but concentration is a risk. For Vertiv, the service is necessary but the cycle is a trap. For Arista, the moat is strong but the price is full. These are not value traps, but they are not value buys either-at least not at current levels.

The Compounding Engine and Capital Allocation

The true test of a business is not its headline growth rate, but the quality of its compounding engine. A durable, high-return model generates cash flows that can be reinvested at attractive rates, compounding intrinsic value over decades. This is the heart of what Warren Buffett calls "owner earnings." For the value investor, the focus must shift from short-term top-line expansion to assessing a company's ability to consistently earn high returns on capital and allocate that capital wisely.

This is where the current AI-driven hype cycle creates a significant risk. Companies are being valued on their participation in a powerful trend, but the sustainability of that participation is not guaranteed. A wide moat is essential to protect those returns from erosion. Consider Arista Networks, which has built a critical position in AI networking. Its strength is a narrow, technology-specific moat. While this provides a temporary advantage, it is vulnerable to disruption or commoditization. A true economic moat, like that of a software platform with high switching costs, is broader and more durable. It allows a company to earn high returns even when the initial trend fades. The danger is mistaking a temporary advantage for a permanent one.

Capital allocation discipline is the other half of the equation. A company can have a great business but destroy shareholder value if it squanders cash on poor acquisitions or excessive share buybacks at inflated prices. The evidence suggests that the market is now rewarding quality more explicitly. As Robert Novy-Marx's research indicates, incorporating quality metrics into value investing can help distinguish between a "bargain stock" and a "value trap." The latter are cheap for a good reason-often because the business model cannot generate high, sustainable returns. The value trap is not just a cheap stock; it is a business that cannot compound.

This quality focus is reflected in the current market's rotation. Value stocks have outperformed growth in 2026, a trend that could reverse with a Federal Reserve rate cut. Yet, this outperformance underscores a broader investor reassessment. The easy part of picking value stocks based solely on cheap valuation ratios may be over. The smarter money is now looking for companies that combine reasonable prices with demonstrable quality: wide moats, high returns on capital, and disciplined capital allocation. This is the setup for avoiding value traps and capturing the true upside of a value strategy.

For the three stocks in question, the investment thesis hinges on their ability to navigate this quality filter. Datadog's path to a 25% operating margin is a step toward a cash-generating engine, but its concentration risk challenges the durability of its returns. Vertiv's cyclical model offers no guarantee of high returns on capital, regardless of the AI build-out. Arista's critical role in networking is undeniable, but its premium valuation leaves no margin for error in capital allocation or execution. In a market that is beginning to price in quality, these are not the companies that will compound value most reliably. The patient investor waits for a business with a wide moat and a margin of safety, not just a seat at the AI table.

Catalysts and Risks: What Would Change the Thesis

For the value investor, the path forward hinges on specific catalysts that would confirm or contradict the current skepticism. The key is to identify the precise events or data points that would justify a premium valuation or, conversely, expose the underlying risks.

The primary catalyst for any of these growth stocks would be sustained, high-quality profitability that justifies their current valuations. For Datadog, this means not just hitting its adjusted operating margin target of over 25%, but doing so consistently while maintaining its growth trajectory. A clear signal would be a multi-quarter run of margins expanding toward that goal, coupled with a visible ramp in its security and AI product segments. For Arista, the catalyst is demonstrating that its critical networking role translates into durable, high-return capital allocation. Evidence of a widening moat-perhaps through new, defensible technology or deeper integration with cloud providers-would be a positive sign. The bottom line is cash flow generation that exceeds the cost of capital, proving the business can compound value.

The most significant risk, however, is a shift in the macro environment that disproportionately impacts these high-multiple, cyclical, or AI-dependent businesses. The evidence points to a potential reversal in the value/growth trade. As noted, value stocks have outperformed growth in 2026, but this trend could reverse if the Federal Reserve begins cutting interest rates. For companies priced for perfection, rising rates are a headwind; a pivot to lower rates could make their expensive valuations even less palatable. More broadly, any sign of a slowdown in AI-driven capital expenditure-whether from economic weakness, regulatory friction, or technological stagnation-would pressure the revenue growth of all three. Vertiv, with its cyclical model tied to large customer capex, would be especially vulnerable.

A value investor should also watch for any sign of a widening moat or a significant improvement in capital efficiency that could create a new, attractive entry point. This is the essence of finding a margin of safety. For instance, if a company like Datadog successfully diversifies its customer base beyond its high-value concentration, it would strengthen its moat and reduce a key vulnerability. If Arista's technology becomes so entrenched that competitors cannot catch up, that would be a classic widening moat. The patient investor looks for these quality improvements not to chase a rally, but to see if the price eventually aligns with a more durable business.

Ultimately, the thesis is about patience and discipline. The catalysts are about confirming quality; the risks are about protecting capital. As Charlie Munger's 50% drop test reminds us, the market will deliver volatility. The value investor's job is to ensure their holdings are businesses that can withstand it, not just stocks that can.