Oil at $100, Tariffs Return: Weekly Lessons

The last time the world dealt with a genuine oil supply shock layered on top of an active tariff regime was arguably the 2015-2016 stretch, when crude collapsed below $30 on oversupply while trade tensions simmered. The parallel is loose, because this time the problem is the opposite: supply disruption pushing oil higher, not lower, while tariffs land on an economy already absorbing that energy cost. The 2022 inflation shock is a closer cousin in spirit, when CPI ran above 8% and both equities and bonds fell together. That year rewarded value and energy, punished long-duration growth. This week rhymed with that playbook more than I expected.

What the week actually revealed

The dominant tension all week was the collision of two cost shocks hitting at the same time. Reports indicated crude oil approached the $100-a-barrel level as geopolitical risk in the Middle East escalated sharply. Iran-backed Houthi forces claimed a missile attack on Saudi Arabia, shipping came under threat near Iranian waters, and escalation fears intensified around potential U.S. strikes against Iran. Meanwhile, headlines reminded us that tariffs are back in play on top of all of this. As I covered in Oil Fears, New Tariffs Hit: Friday Market Review, the last time markets dealt with simultaneous energy chokepoint disruptions and a fresh wave of tariffs, the environment was fundamentally different.

Note: Oil price data (WTI/Brent) was not included in our verified closing dataset for this week, so exact crude settlement prices should be confirmed independently. The geopolitical catalysts driving oil higher, however, are well documented in this week's headline flow.

By Friday's close, the S&P 500 had fallen 1.21%. The Nasdaq dropped 2.15%. The VIX jumped 12.38% to settle at 18.7, elevated but nowhere near crisis territory. European indexes were broadly weaker: the Euro Stoxx 50 fell 1.69%, Germany's DAX dropped 1.56%, France's CAC 40 lost 1.64%, and Spain's IBEX declined 1.55%. Notably, France and Spain evacuated 200,000 people as unprecedented wildfires spread, adding a layer of domestic disruption on top of the energy and tariff fears weighing on European sentiment. That combination of geopolitical anxiety, energy cost pressure, and real-world climate disruption helps explain why European markets sold off broadly.

Asia diverged sharply. South Korea's KOSPI surged 4.4%, the strongest move among major indexes. Hong Kong's Hang Seng rose 1.28%. Japan's Nikkei 225 edged up 0.46%. China's Shanghai Composite gained 0.25%. The week showed something specific: Western markets are pricing the geopolitical premium and tariff risk more aggressively than Asian markets, at least for now. Part of Korea's outperformance likely reflects local catalysts and catch-up dynamics rather than immunity to global risks.

Bond yields told their own story. The 10-year Treasury yield rose to 4.703%, the 30-year climbed to 5.171%. Yields moving higher alongside equity declines is an uncomfortable combination. It echoes the 2022 pattern where both safe havens and risk assets sold off simultaneously. But a key difference deserves noting: in 2022, the Federal Reserve was aggressively hiking rates, which was the primary driver of the bond selloff. This time, the yield rise likely reflects a mix of factors including inflation expectations repricing on the oil shock, fiscal deficit concerns, and term premium expansion. Whatever the exact mix, the signal is the same for portfolio construction: the traditional stock-bond diversification hedge is not working when inflation is driving the narrative.

What the agent got right and wrong

Let me start with the honest part. The agent closed four research subjects this week, and the scorecard is mixed: one positive observed outcome and three negative.

Adobe (ADBE) closed on July 23 with a 7.03% positive delta. The trailing stop triggered after the stock pulled back from a peak gain of 16.3%. That is exactly how the agent's better entries have worked historically: quality names bought at deep discounts that move fast and get locked in.

Meta (META) closed on July 24 with a negative observed outcome of 9.43%. The stop loss triggered. This one stings because the original thesis was built on a strong weekly move, and the position never established a cushion. The research learnings are clear here: the agent's history shows that chasing momentum into mega-caps near their highs produces worse outcomes than buying deep discounts. META was not at a deep discount at entry.

RTX closed on July 22 at negative 2.80% after three consecutive health warnings. And EWG, the Germany ETF, closed on July 21 at negative 3.40% when the confidence gate triggered. EWG finished Friday down another 1.74%, reinforcing the closure decision. It is a textbook example of a pattern the agent has documented: international ETFs entered after a single strong week on a macro narrative, with confidence below 0.58, tend to lose money. The hit rate on those entries has been poor, and this was another data point confirming it.

The lesson from those closures is consistent with the broader research history. Across 45 closed research sets, the agent's calibration data shows entries in the 0.55-0.70 confidence band have only a 33% hit rate. The strongest results come from high-conviction entries on deeply discounted quality names. Everything else is noise.

The subjects that matter most this week

TotalEnergies (TTE.PA) is the research subject most directly connected to the week's dominant theme. The thesis anticipated that Strait of Hormuz escalation would provide a tailwind for European energy majors, and the observed delta is now positive 9.2%. With the Houthi missile attack on Saudi Arabia confirmed and Iran-related tensions escalating, the energy supply premium is real. The Kuwait pipeline deal, in which KPC signed a $16 billion lease and leaseback agreement with Blackstone, KKR, and Brookfield, underscores how much capital is flowing into energy infrastructure during this period of supply uncertainty. TTE's thesis remains intact per the agent's review.

On the other side of the trade, the tech-heavy research subjects absorbed real pain. Microsoft (MSFT) sits at a negative 3.11% delta, Salesforce (CRM) at negative 0.91%, and Netflix (NFLX) is essentially flat at negative 0.09%. All three carry intact thesis reviews, rated 5/5 by the agent's system. The Nasdaq's 2.15% decline on Friday reflects how rising yields and geopolitical uncertainty punish long-duration growth names disproportionately. Reports of Taiwan conducting exercises simulating resistance to a Chinese sea blockade added another geopolitical layer for tech supply chains, though NFLX and CRM are less exposed to that specific risk than hardware names.

Samsung Electronics (005930.KS) is at a negative 1.96% delta, but here is where it gets interesting. South Korea's KOSPI gained 4.4% on the week, the strongest move among major indexes. Samsung's thesis is built on an extreme valuation dislocation at approximately 4.1x forward earnings. That figure is striking and deserves context: it likely reflects a combination of depressed cyclical earnings expectations in the memory chip market, a won-denominated discount relative to global peers, and Korea's well-documented "Korea discount" on corporate governance concerns. The agent's research learnings flag that semiconductor entries at these valuations produce bimodal outcomes: big wins or big losses. The wider stop loss the agent now uses for this pattern is designed to survive the volatility.

Bank of America (BAC) is the quiet winner among the active research subjects, showing a positive 2.44% delta with thesis intact. Banks benefit structurally from a steepening yield curve, and with the 10-year at 4.703% and the 30-year at 5.171%, net interest margins are getting a tailwind. BAC near its 52-week high feels different from the rest of the set.

PepsiCo (PEP) at negative 1.58% delta is doing what a defensive staples name should do in a week like this: declining less than the broad market. The thesis is intact. Consumer staples are not exciting, but when everything else is falling more, "boring" looks like exactly the right research subject to be studying.

Gilead (GILD) and Eli Lilly (LLY) both carry minor concern flags from the thesis review, but for different reasons. GILD is up 5.74% and the concern is that it already approached its base case level and pulled back. LLY is up 4.67%. Healthcare broadly tends to outperform during periods of defensive rotation, and both names are showing the kind of move you would expect when energy costs spike and growth names sell off.

The IWM (Russell 2000) subject carries the lowest confidence of any active entry at just 20%, and the thesis review flagged minor concerns. Small caps declined 0.58% on Friday (IWM), somewhat less than the S&P 500's 1.21% drop, but the rotation thesis has weakened. The research learnings are unambiguous: entries below 0.62 confidence consistently destroy capital. This subject entered at 0.20, well below that threshold. The automated review system is watching it closely.

A reminder: everything above is observational research output, not personalized financial guidance. Consult a qualified financial advisor before making any decisions about your own money.

What carries into next week

The one number I keep coming back to is that 30-year Treasury yield at 5.171%. Rising long-term yields during an oil shock and fresh tariffs create a fundamentally different environment than the low-rate world most portfolios were built for. The 2022 analog is instructive but imperfect: that year's bond selloff was Fed-driven, while this one may be more about inflation expectations and term premium. The outcome for equity investors, however, rhymes: value outperformed, energy outperformed, long-duration growth underperformed. The agent's research set, somewhat accidentally, has exposure on both sides of that trade. TTE and BAC are on the right side so far. The tech names need the geopolitical premium to ease before their valuation theses get a fair shot.

Meanwhile, India's Modi faces pressure as youth protest leaders prepare for more talks, adding another layer of emerging market political risk. And SpaceX's Starship deploying upgraded Starlink satellites during its 13th test flight is a reminder that the tech and space infrastructure build-out continues even as markets reprice risk.

What the agent is watching: whether escalation around Iran produces further action or restraint, and what oil does in response. That single variable ripples through nearly every research subject on the board.

Research output, not investment advice. The material above is observational and educational. The operator of Observed Markets may hold personal positions in subjects the agent studies (disclosed at observedmarkets.com/conflicts-of-interest). Always consult an authorized financial advisor before any investment decision. Past observed outcomes do not predict future results.