The biggest fear during mega-cap tech earnings season isn’t weak results—it’s expectations that are too high. Stocks like PLTR, AMD, and NVDA have posted mind-boggling gains over the past few years: PLTR rose about 167% in 2023 and another 340% in 2024, NVDA gained about 239% in 2023 and 171% in 2024, and AMD rose about 127% in 2023. With gains like these, current prices have long since priced in “smooth sailing for the next five years,” embedding very high growth assumptions in their valuations.

If guidance is even slightly softer than expected, or gross margins or next quarter’s revenue outlook don’t continue to beat expectations, valuation multiple compression can hit much harder than the earnings miss. Results might fall short by only a few percentage points, but the multiple could drop from 40x to 25x. A 5%–10% after-hours sell-off wouldn’t be unusual. Over the past two years, both NVDA and AMD have seen sharp pullbacks despite decent revenue, because their guidance or gross margin commentary wasn’t strong enough.

I came up through traditional finance and am used to starting with risk budgets, volatility, and drawdown tolerance when managing positions. At these levels, I only dare to take small positions in these stocks, keeping each one to 3%–5% of my portfolio. I also trim before earnings. I just can’t hold a large position through the report.

The key this week is the capital expenditure guidance from several giants. Cloud providers like Microsoft, Alphabet, Amazon, and Meta are expected to spend a combined total of more than $300 billion in capital expenditures in 2025. Now that the AI story has been told for a while, we need to see who can actually turn it into profits instead of relying on the narrative to drive valuations.

How are you handling these high-flying stocks during earnings season—trimming your positions or riding it out?