Earnings Season Is Two-Thirds Over. Here's How It's Going and What It Means for the Market.

About 84% of S&P 500 companies that have reported have beaten EPS estimates, above the 10-year average of 76%, with aggregate earnings growth running around 27% year over year. The main drivers are the megacap tech names in the Magnificent Seven, especially Alphabet, Amazon, Meta, and Nvidia, which delivered outsized contributions to earnings growth. Nvidia reported $68.1 billion in fiscal Q4 sales and EPS of $1.62, about 6% above consensus.

Analysis

The key market implication is not simply that megacap tech is beating expectations, but that index-level earnings are becoming increasingly dependent on a narrow set of names with unusually volatile non-operating items. That makes the headline EPS growth rate look stronger than the underlying revenue/operating-profit trajectory, which should reduce confidence in forward multiple expansion unless breadth improves. In practice, this is bullish for passive/index holders in the very short run, but it also concentrates disappointment risk into the next print from the same cohort.

The second-order winner is not just the reported companies but the AI supply chain beneath them: compute, networking, power, and advanced packaging remain the cleaner way to express this theme because they are less exposed to accounting noise and more tied to capex budgets. If management teams keep leaning into AI infrastructure spend, the earnings impulse should extend into the next 2-3 quarters even if consumer internet growth normalizes. The vulnerability is that any deceleration in capex or a stronger-than-expected regulatory/tax drag would hit sentiment disproportionately because investors are now anchoring on sustained outperformance from a handful of mega-caps.

The contrarian read is that the market may be overconfident in the durability of the earnings beat rate. A large share of this quarter’s upside came from one-time gains and tax effects, which do not scale into a repeatable EPS trend; if the next cycle strips those out, the perceived earnings quality will deteriorate quickly. That creates a window for tactical fades in the names with the biggest earnings-driven moves, while keeping a structural long on the broader AI infrastructure trade.

Key Decisions for Investors

• Stay tactically long QQQ / SPY into the next few weeks, but size it as a momentum trade rather than a durable fundamental call; upside persists as long as the market keeps extrapolating index earnings breadth. Tighten risk if breadth narrows further or if the next megacap report disappoints on forward guidance.
• Initiate a pair trade: long NVDA (or a basket of AI infrastructure suppliers) vs short a basket of mega-cap internet names with heavy non-operating EPS contribution, over 1-3 months. The goal is to own cleaner operating leverage while fading accounting-driven earnings quality compression.
• Buy AMZN or META pullback hedges via 1-2 month put spreads if the post-earnings move extends beyond fundamentals; these are the most vulnerable to mean reversion once one-time benefits are stripped out.
• Prefer semis / networking / power infrastructure over the Magnificent Seven as a 6-12 month expression of the AI capex cycle; the risk-reward is better because revenue is tied to procurement budgets, not sentiment-driven multiple expansion.