Next week, the most important six weeks of the market's second half begins. PepsiCo and Delta Air Lines open the Q2 earnings season around July 9, the big banks, JPMorgan and Citigroup among them, report in mid July, and the technology giants that hold up this entire market follow in the back half of the month. Microsoft, Alphabet, Meta, Apple, and Amazon report through late July, with Nvidia closing the show in late August.
Every earnings season matters. This one carries unusual weight, because the market is asking it to justify a very specific and very expensive story. The S&P 500 sits near record highs, up about 7.7 percent for the year. The Shiller price to earnings ratio stands above 41, the second highest level in more than 140 years. The top ten companies represent roughly 40 percent of the index. AI infrastructure beneficiaries account for about half of all S&P 500 earnings growth. And the Federal Reserve has just told markets that the next rate move is more likely a hike than a cut, which means valuation support from monetary policy is not coming.
When a market this expensive meets a Fed this unfriendly, earnings are the only pillar left. Here is the playbook for the next six weeks, built the way institutions build theirs: know the expectations, know the calendar, know what actually moves prices, and decide your actions in advance.
The bar these companies must clear
Earnings season is never about whether results are good. It is about whether results beat what was already priced in. That distinction is the source of most retail confusion, the quarters where a company grows profits 30 percent and the stock falls 8 percent.
So here is what is priced in. Analysts expect another strong quarter, and the recent record supports them: in the first quarter, S&P 500 results beat expectations by 6 percent, the strongest beat rate in four years, with earnings per share growing nearly 28 percent on revenue growth close to 12 percent. Since the quarter began, estimates have risen in only 5 of 16 sectors, with the increases concentrated exactly where you would guess. Energy earnings are projected to more than double, up roughly 114 percent, on the oil shock that followed the Iran conflict and the disruption of the Strait of Hormuz. Basic materials and technology follow with growth expectations in the 40 percent range.
Notice the structure of that setup. The market's earnings growth is concentrated in two trades: AI infrastructure and energy. One is powered by 725 billion dollars of planned capital spending this year from just four companies, Google, Amazon, Microsoft, and Meta, up 77 percent from last year. The other is powered by a war premium in oil that could fade if the conflict continues winding down. Both engines are real. Both are also single points of failure, and this earnings season is a stress test of each.
The three questions that decide the second half
Institutions do not listen to earnings calls for the results. The results are known within seconds and priced within minutes. They listen for guidance on the handful of questions that determine the next two quarters. This season, three questions tower over everything.
Question one: does the AI capital spending hold? Watch the hyperscaler capex guidance like it is the only number that exists, because for this market, it nearly is. The AI trade has a circular quality that strategists have started to flag openly: the companies whose earnings justify the market's valuation depend on continued data center spending by a handful of buyers, whose own valuations depend on that spending eventually producing returns. As long as capex guidance rises, the loop is virtuous. The first quarter a major hyperscaler trims its spending plans, everything priced off that spending, semiconductors, memory, servers, power, data center real estate, reprices in an afternoon. AI infrastructure earnings estimates for 2026 have been revised up more than 50 percent since late 2024 while estimates for the rest of the index drifted slightly lower. That divergence is the market's entire skeleton.
Question two: what does the consumer say? The index level story is euphoric, but the household level data is not. Real wages are shrinking against 4.1 percent inflation, the saving rate has fallen to 3 percent, and consumer sentiment sits near historic lows for anyone without a stock portfolio. The banks report first and will tell you about credit card delinquencies and loan demand. PepsiCo and Delta will tell you whether households are still absorbing price increases. If the consumer facing reports show cracks while the AI complex reports blowouts, the two track economy becomes a two track market, and the 490 stocks that are not the AI trade will tell you about it.
Question three: can anyone pass along the costs? With input costs elevated by energy and tariffs, margins are the tell. First quarter operating margins hit a record near 16 percent, proof of extraordinary pricing power. Listen for how often management teams mention cost pressure they cannot pass through. Margins crack before earnings do, and guidance cracks before margins.
What institutions do differently in these six weeks
Here is the uncomfortable truth about earnings season: the average investor should mostly do nothing during it, and the discipline to do nothing is precisely what separates the professionals.
Institutions prepare rather than react. Before the season starts, they know every reporting date for every meaningful position, they know the consensus numbers, and they have already decided what they would do in each scenario. When the report hits, they are executing a decision made calmly two weeks earlier, not forming one during a 12 percent after hours move.
They also respect the base rates. Single stock moves around earnings have gotten violent, particularly in the AI complex where positioning is crowded, and the first move is frequently reversed within days as the algorithmic flows wash out. Buying the first red candle of an earnings disappointment, or chasing the first green one, is statistically the worst timed trade most retail investors make all year.
The retail translation is a short list. Know when your holdings report, so nothing ambushes you. Never hold a position so large that one earnings call changes your financial life, a rule this market's concentration makes urgent, given that the average S&P 500 member has already seen a 21 percent drawdown this year beneath the index's calm surface. Use scheduled, automated buying through the season, so volatility becomes your dollar cost averaging discount instead of your emotional trigger. And write your reactions in advance: if a core holding falls 15 percent on guidance, is your plan to add, hold, or reassess? Deciding now costs nothing. Deciding at 4:05 pm on report day costs whatever your adrenaline says it costs.
The tooling for this is not complicated. A consolidated dashboard like Empower shows your true exposure to each reporting company across every account and fund you own, which is the number you need before deciding whether any single report matters to you. And an automated platform like M1 Finance executes your scheduled buys into your target allocation on autopilot, which is the mechanical discipline that turns a volatile six weeks into an accumulation opportunity.
How to read a beat, and how to spot a real one
One more layer of institutional literacy before the reports start, because the single most misleading word in financial media over the next six weeks will be the word beat.
Companies beat estimates roughly three quarters of the time, in good markets and bad. This is not because corporate America is perpetually surprising. It is because the expectations game is managed. Analysts publish estimates, management teams guide those estimates down through the quarter with carefully calibrated caution, and then results clear the lowered bar by a comfortable margin. Everyone plays, everyone knows, and the headline beat rate stays high through booms and recessions alike.
So the beat itself carries almost no information. What carries information is the shape around it. A real positive surprise pairs the earnings beat with a revenue beat, because you cannot financially engineer customers. It comes with raised forward guidance, because management just told you the next quarter looks better, not merely the last one. And it survives the conference call, where analysts probe margins, demand trends, and spending plans. The reports that beat on earnings, miss on revenue, and guide down are the ones that fall 10 percent, and the headline will still call them a beat.
For this specific season, apply that lens ruthlessly to the AI complex, where positioning is most crowded and the reflex to celebrate headline numbers is strongest. First quarter results beat expectations by 6 percent in aggregate. If the second quarter beat margin narrows toward the historical norm while guidance language gets cautious, that is deceleration hiding inside a season of beats, and deceleration at a 41 Shiller multiple is how corrections start. Read the guidance, not the headline.
The calendar and the trap
Mark the shape of the season: banks and early consumer names in mid July, the megacap technology reports clustered in the final week of July, Nvidia standing alone in late August as the season's true finale. The gap between big tech's July reports and Nvidia's August report is historically a nervous stretch, where the market has digested the capex guidance but not yet seen the results of its most important company.
One trap to name explicitly, because every earnings season claims victims with it: options markets already price the expected move. Buying calls before a report because you expect good numbers is not a trade on the company, it is a bet that results beat the implied volatility already embedded in the option premium, a far higher bar. The institutional norm is the opposite posture: position size such that no single print matters, then let six weeks of reports tell you about the economy while everyone else trades their amygdala.
Your pre season checklist
I have compiled the complete system into the Earnings Season Field Guide: the full Q2 reporting calendar with the dates that matter, the three question framework with the specific metrics to pull from each report, the position sizing rules, and the pre written reaction plan template. Reply to this email with the word EARNINGS and I will send it to you at no cost.
Know someone who trades earnings season on vibes? Forward them this issue. Refer 3 readers to Money Systems Lab and you unlock the full playbook library. Refer 10 and you receive lifetime premium access to every system we publish. Your referral link is at the bottom of this email.
Money Systems Lab publishes on Beehiiv and runs its production systems on Make, for the same reason this entire issue preaches preparation: reliable output comes from architecture, not adrenaline.
The reports are coming. Decide everything before they do.
Taylor Voss
Money Systems Lab Institutional-grade financial intelligence for everyone else
Disclosure: Some links in this issue are affiliate links, which means Money Systems Lab may earn a commission at no additional cost to you. We only recommend tools we would use ourselves. Nothing in this newsletter is personalized financial advice. Do your own research and consult a licensed professional before making investment decisions.