Trading With Claude: Day 28 — The Semiconductor That Ignored the War
Dushyant Remivasan14 min read·Just now--
Posted by Dushyant Remivasan | April 23, 2026
🟢 Day 28 Result: POSITIVE — SOXX intraday +$5.43 realized | Running closed P&L: +$31.43 | One session remaining
Day 28 was the session where the experiment found its clearest trade of the final week, not in oil, not in gold, and not in anything connected to the Strait of Hormuz. The best play of the day was in semiconductors — a sector that has spent the entire 56-day Iran war building its own independent narrative entirely decoupled from geopolitical noise. SOXX was entered at $440.89, a partial exit at $445.76 locked in the first tranche, and the remaining share was stopped out at $441.45 on a midday pullback. Total gain: $5.43. The experiment’s running closed P&L now stands at $31.43 — a 125% return on the original $25 seed capital with one session remaining.
Today’s Headlines — What the Market Was Watching
The Finnhub feed this morning told a familiar story with one new dimension.
“Iran takes seized ships to port, countries check on seafarers’ safety.” The two container ships seized Wednesday are confirmed in Iranian territorial waters. This is no longer a threat or a warning — it is a completed action. The IRGC demonstrated that it can physically stop and hold commercial vessels crossing the Strait during an active ceasefire, and the international community’s response was to ask about crew welfare rather than take any countermeasures. Iran’s leverage in the Strait has been formally demonstrated.
“Wall St muted as Middle East impasse, mixed earnings hurt sentiment.” The broader market is not rallying or panicking — it is waiting. The frozen standoff creates a suspended-animation tape where large directional bets are difficult to justify, which is exactly why the cleanest trades of the final week have been in names with zero Iran exposure.
“US business activity recovers in April, war with Iran is boosting prices.” The S&P Global PMI survey confirmed US manufacturing and services activity expanded in April despite the war, but noted that input prices are rising. This is the stagflation signal confirming itself in real-time business survey data rather than just commodity prices. Companies are absorbing higher costs and passing them on. The inflation component of the Fed trap is getting worse every week the Strait stays closed.
“Iran war impact seeps ever deeper into global economy.” Reuters documented the expanding footprint of the supply disruption — from energy to food to manufacturing inputs to consumer goods. The war’s economic impact is no longer contained to oil markets. It is in paint, in airlines, in Diet Coke cans in India, in the cost of condoms in Malaysia. The disruption has become structural.
“Son of former Shah of Iran appeals to Western countries for support.” A detail easy to overlook but analytically significant — the former Shah’s son is making public appeals for regime change support, which is exactly the dynamic Trump referenced when he called Iran’s government “seriously fractured.” The internal political instability that makes a unified Iranian proposal impossible is not just diplomatic language — it has a political geography. The IRGC and the parliamentary leadership are not operating from the same position, and until that fracture heals or one faction wins, no deal can be signed.
The SOXX Trade — Full Record
The entry thesis was straightforward. SOXX had spent the entire Iran war as the experiment’s cleanest non-correlated asset — an ETF whose constituent companies are building the physical infrastructure of artificial intelligence and whose revenue growth is driven by data center capex commitments that were contracted before the war began and will be honored regardless of what happens in Islamabad. No Iranian petrochemical halt affects Nvidia’s GPU shipments to a Texas data center. No IRGC gunboat on the Strait changes Meta’s $115 billion capex plan for 2026.
The five-minute chart confirmed the thesis before entry was taken. The overnight session showed SOXX grinding from $430.50 to $439 in a clean ascending staircase through the pre-market hours. The regular session open at 8:30 AM CT produced the cleanest single candle of the day — a gap-up from $435 to $441 on the highest volume bar of the entire chart, institutional accumulation on the open. The subsequent 90-minute consolidation at $441–442 showed healthy digestion rather than distribution. The staircase from $442 to $445 between 9:30 and 10:35 AM CT was exactly the low-volatility, higher-lows structure that signals a position working in your favor.
Two shares were entered at $440.89 with a stop at $434.00 — below the session low of $434.86. Target 1 was set at $446.00. The first share was sold manually at $445.76 when SOXX touched near the target, capturing $4.87 per share. The stop on the remaining share was raised to $441.50 — above entry, locking in a minimum gain on the second tranche. That stop triggered at $441.45 when SOXX pulled back midday, capturing $0.56 on the second share.
SOXX Trade -- Complete Record
┌──────────────────┬────────┬──────────┬──────────┬─────────┐
│ Tranche │ Shares │ Entry │ Exit │ Gain │
├──────────────────┼────────┼──────────┼──────────┼─────────┤
│ T1 manual sell │ 1 │ $440.89 │ $445.76 │ +$4.87 │
│ T2 stop triggered│ 1 │ $440.89 │ $441.45 │ +$0.56 │
├──────────────────┼────────┼──────────┼──────────┼─────────┤
│ Total │ 2 │ $440.89 │ $443.61 │ +$5.43 │
│ │ │ │ (avg) │ │
└──────────────────┴────────┴──────────┴──────────┴─────────┘The stop-raise-after-T1 mechanic worked exactly as designed. After the first share was sold at $445.76, the worst case on the remaining share became a small gain rather than a loss. When SOXX pulled back, the raised stop at $441.50 executed cleanly and the position was closed with profit locked on both tranches. This is the partial exit discipline practiced on UCO Day 15 and MP Day 23–25 — applied correctly on the experiment’s penultimate session.
The Strait of Hormuz — Day 54, Still Closed
Shipping traffic through the Strait of Hormuz has been largely blocked since February 28, 2026, when the United States and Israel launched air operations against Iran. The IRGC has now launched 21 confirmed attacks on merchant ships and has reportedly laid sea mines in the strait. Before the war, approximately 25% of the world’s seaborne oil trade and 20% of global LNG passed through the chokepoint daily.
The seizure of two cargo ships Wednesday is the 21st and 22nd confirmed IRGC actions against commercial shipping since the war began. Iran has now moved those ships to an Iranian port. The crews of those ships are nationals of multiple countries waiting for their governments to negotiate their release. Every day this continues, the political cost of reopening the Strait rises for Iran because it would require releasing the ships and crews, removing the sea mines, and allowing vessels past IRGC patrol zones — all of which is incompatible with the “we do not negotiate under threat” posture Tehran has publicly committed to.
The frozen standoff is not an accident. It is Iran’s strategy. Deny the US a clean military victory, demonstrate control of the world’s most important oil chokepoint, fracture the Iranian government sufficiently to delay any proposal, and wait for domestic US political pressure — falling approval ratings, rising gas prices, food inflation — to soften the American negotiating position. The S&P 500 broke below its 200-day moving average for the first time since May 2025 as the Iran war began, and the subsequent recovery has been uneven, with semiconductors leading while energy and precious metals oscillate on every diplomatic headline.
The economic cost is now being measured in aggregate rather than daily price moves. Supply disruptions are running at approximately 13 million barrels of crude, condensates, and natural gas liquids per day. Cumulative disruption has already exceeded half a billion barrels, and experts warn that even a deal tomorrow would not immediately unwind the damage — it could take months to claw back lost supply.
The Semiconductor Sector — Why SOXX Hit a 25-Year Record in April
SOXX has surged 27.7% in April alone — a move so large it eclipses every monthly return the fund has posted since its launch in July 2001. The ETF absorbed $2.05 billion in April inflows, more than double its previous monthly record. Combined with the VanEck Semiconductor ETF’s $3.4 billion in inflows, the two funds took in $5.45 billion in a single month — a category record. SOXX reclaimed all-time highs last seen in late February and posted 12 positive sessions out of 14 in April.
Understanding why semiconductors decoupled from the war narrative requires understanding where the demand is coming from. Nvidia CEO Jensen Huang estimated that between $3 trillion and $4 trillion will be spent on AI infrastructure by the end of the decade. Amazon projected $200 billion in 2026 capex, Google between $175 and $185 billion, and Meta between $115 and $135 billion. All told, hyperscalers are planning to spend nearly $700 billion on data center projects in 2026 alone. None of those contracts depend on Hormuz.
Oracle’s April catalyst: Oracle unveiled new generative AI enhancements within its Utilities Industry Suite, integrating AI into the Aconex project management platform to help utility providers optimize grid reliability and reduce soaring operating costs — capabilities directly relevant to the energy crisis created by the Iran war. Oracle also announced a new public cloud region in Casablanca, Morocco, accelerating sovereign cloud demand. The company reported 84% IaaS revenue growth and $553 billion in remaining performance obligations — an all-time high backlog. ORCL returned above its 20-day and 50-day moving averages following the announcements, with Wall Street consensus at a strong buy and a mean price target implying more than 65% upside from current levels.
Meta’s AI infrastructure commitment: Meta stock rallied after reporting quarterly results that showed AI is directly boosting the company’s core advertising revenue, validating the capex investment thesis that skeptics had questioned. The company’s 2026 infrastructure spending plan of $115 to $135 billion represents an 80% increase from 2025 levels.
What is coming next in the semiconductor space: Texas Instruments reported earnings today — highlighted by Jim Cramer as the star of the earnings show. Intel, AMD, and Broadcom all have earnings in the next two weeks. The FOMC meets April 28–29. Any signal from Powell that rate cuts are back on the table — supported by evidence that oil-driven inflation is transitory rather than structural — would be rocket fuel for semiconductor valuations, which are rate-sensitive given their long-duration growth profiles.
A fresh analysis of AI infrastructure stocks noted that AMD, MRVL, MPWR, SOXX, and AMZN are among the most overbought names on RSI and Stochastic readings, meaning traders should focus on entry zones and pullbacks rather than chasing strength. The strongest message from the sector is that AI, semiconductors, optical networking, and digital infrastructure remain the market’s leadership groups — but the next layer is that many of the best names are in late-stage momentum territory, making trade selection and entry discipline critical.
According to World Semiconductor Trade Statistics, global semiconductor sales are expected to surge 26.3% year over year to $975.4 billion in 2026, with AI accelerators, high-bandwidth memory, advanced logic chips, and supporting infrastructure as the primary growth engines. The expansion is broad-based across regions and product categories.
The Agriculture Sector — The Most Underreported Crisis of the War
While oil and gold dominate the trading headlines, the most structurally consequential impact of the Hormuz closure may be in agriculture. The experiment tracked CF Industries as its primary agriculture play, and the thesis behind that position deserves a thorough analysis as the final session approaches.
In 2024, up to 30% of global fertilizer trade passed through the Strait of Hormuz from the Persian Gulf to export markets, as well as an estimated 20% of LNG — a key fertilizer feedstock — and 27% of globally traded oil. The Iran conflict represents the third major shock to fertilizer markets in six years since COVID-19.
FOB granular urea in Egypt — the benchmark for global nitrogen fertilizer pricing — jumped to approximately $700 per metric ton, up from $400 to $490 before the war began. Oxford Economics confirmed that urea and ammonia prices surged by around 50% and 20% respectively since the war started. A total of 54 agricultural groups wrote to President Trump calling for “much-needed market relief for America’s farmers” as planting season began amid surging fuel and fertilizer prices.
The farm-level economics have become brutal. The USDA projects that corn costs $5 per bushel to grow in 2026 but will sell for $4.20. Soybeans cost $12.27 to produce but will sell for $10.30. Seventy percent of US farmers report they cannot afford enough fertilizer. Farm bankruptcies hit 315 filings in 2025 and total farm debt is expected to reach a record $624.7 billion in 2026.
The six-month outlook for agriculture is the most bearish sector story in the entire experiment. The USDA Prospective Planting report estimates that plantings of corn and wheat acreage will fall 3 percent each in 2026 relative to 2025. The UN World Food Programme estimates that if oil prices remain above $100 per barrel through June 2026, the number of people facing acute hunger could increase by 45 million. The WFP estimates that the impact of energy prices on food prices could peak approximately four months following the onset of the Iran war — meaning the worst food price inflation is likely still ahead.
Wood Mackenzie estimates that even with a ceasefire in place, it will take months to get Middle East energy production back to normal. The US Energy Information Administration has warned that fuel prices could continue a months-long rise even after Hormuz reopens. A possible super El Nino weather pattern has been forecast by the European Centre for Medium-Range Weather Forecasts, which could compound food security concerns if it materializes alongside the ongoing price shock.
For investors, this creates a specific asymmetry. CF Industries, Mosaic, Nutrien, and the broader nitrogen fertilizer complex benefit from elevated urea prices as long as Hormuz stays closed. These companies sell at $700 what they produce with inputs priced at $400 — the margin expansion is enormous. The risk is that a sudden peace deal would crater urea prices as Middle Eastern production returned to market. But the CSIS and IFPRI research suggests that even a deal tomorrow would take months to normalize supply chains, meaning elevated fertilizer prices are structurally sticky even in a resolution scenario.
The CF Industries trade in this experiment — entered at $116.10 on Day 16 and exited on Day 25 at $125+ — captured the front end of this repricing. The full agricultural supply chain crisis is just beginning to work through to consumer food prices.
The 200-Day Moving Average — What It Is, Where It Works, and Where It Fails
Several trades in this experiment were either taken or avoided based on the chart structure around the 200-day moving average. You have been researching this strategy, so this is the honest analysis.
What it is: The 200-day moving average is the simple or exponential average of a security’s closing price over the prior 200 trading days — roughly 40 weeks. It functions as a smoothing device for long-term trend assessment. The most commonly used signal combines the 50-day and 200-day averages: a buy signal is generated when the 50-day crosses above the 200-day (the “golden cross”) and a sell signal when the 50-day crosses below (the “death cross”).
Where it works: The 200-day MA is most powerful as a trend filter for long-only equity positions. Historically, the market produces the highest returns when trading above the 200-day SMA. Strategies that hold equities when price is above the 200-day and move to T-bills when below have historically reduced drawdown significantly versus a pure buy-and-hold approach — particularly during the 2000–2002 and 2008–2009 bear markets. Paul Tudor Jones has publicly stated that his primary rule is to never own a stock that is trading below its 200-day moving average.
The strategy works best in four specific contexts. First, in trending equity markets where momentum persists for months at a time. Second, as a filter to avoid buying into a confirmed downtrend — if a stock is below its 200-day, you are buying something the market has been distributing for months, which is structurally disadvantaged. Third, for position sizing — a stock above its 200-day warrants more conviction than one below it. Fourth, as a stop signal for long-term holdings — when a position crosses below the 200-day after an extended uptrend, it is a warning that the trend is changing.
The S&P 500’s break below its 200-day moving average in early 2026 — for the first time since May 2025 — was accompanied by a rotation from a sideways phase to a distribution phase, signaling a bearish trend. The subsequent Iran war added energy and inflation pressure. The recovery above the 200-day has been uneven, with tech and semiconductors leading while cyclicals lag.
Where it fails: The 200-day MA has three well-documented failure modes that are directly relevant to this experiment.
The first failure mode is in binary-event driven markets. Every Iran war trade in this experiment was driven by a binary event — a ceasefire announcement, a ship seizure, a Trump Truth Social post. The 200-day moving average is a lagging indicator built for sustained trends. It cannot tell you what happens to GDX in the 30 minutes after Trump says “I’m extending the ceasefire.” The entire ITA position was stopped out because a binary diplomatic event overrode the technical trend. In event-driven trading, the 200-day is nearly useless as a timing tool.
The second failure mode is in whipsaw conditions. When a market oscillates above and below the 200-day repeatedly in a short time, the crossover strategy generates false signals that each produce small losses. A volatile, news-driven tape like the Iran war produces this pattern relentlessly. The cost of repeatedly buying the golden cross and selling the death cross in a choppy market can exceed the benefit of the signal during genuine trend changes.
The third failure mode is that it is a backward-looking average. The 200-day moving average tells you where the price was on average over the past 200 days. It tells you nothing about where the price is going. SOXX was below its 200-day in early April during the worst of the war panic. The 200-day said “sell.” The correct trade was to buy aggressively. The Iran war decoupling thesis had already changed the fundamental picture — the 200-day was still processing a fear that had already peaked.
The practical takeaway for this experiment: The 200-day MA would have been a useful filter for multi-week position sizing — do not build large positions in names that are breaking below their 200-day into a war-driven downtrend. It would have been useless as a timing tool for any intraday or event-driven trade. The UCO trades, the GDX trade, the MP trade — none of them were 200-day crossover setups. They were thesis-driven entries based on a specific macro event with a defined catalyst and a defined stop.
The best application of the 200-day for a trader following a macro thesis like this experiment is as a reality check: if your thesis says buy oil and oil is in a confirmed downtrend below its 200-day with no catalyst to reverse it, the 200-day is telling you the market disagrees with you. Respect that disagreement until the catalyst arrives.
Account Scorecard — Full Experiment
┌──────────────────────────────────────┬──────────┐
│ Trade │ P&L │
├──────────────────────────────────────┼──────────┤
│ Days 1-14 cumulative │ +$2.38 │
│ Day 15: CF and UCO exits │ +$11.06 │
│ Day 16: CF re-entry + LNG exits │ +$13.21 │
│ Day 17: SLV stop │ -$7.02 │
│ Day 18: LNG stop │ -$15.48 │
│ Day 23: MP T1 (2 shares at $62.00) │ +$8.80 │
│ Day 24: GDX T1 + SLV T1+T2 │ +$16.97 │
│ Day 25: MP T2 │ +$6.73 │
│ Day 26: ITA + GDX + COPX stops │ -$18.45 │
│ Day 27: UCO intraday │ +$7.80 │
│ Day 28: SOXX T1 + T2 stop │ +$5.43 │
├──────────────────────────────────────┼──────────┤
│ Running total │ +$31.43 │
└──────────────────────────────────────┴──────────┘Cash: $1,353.20 | Zero open positions | One session remaining | 1 day trade available
Day 29 — The Final Session
One session remains. One day trade available. The FOMC meets April 28–29 — five days after the experiment closes, so the rate decision falls outside the documentation window. The final session on Friday April 24 will be the last documented trade decision of the experiment.
The macro setup for Friday: the frozen standoff continues, Iran’s seized ships are in port, no talks are scheduled, and the market is processing mixed signals from earnings season. SOXX’s AI infrastructure thesis has 12 positive sessions in April. UCO’s oil thesis is intact but the pre-market gap-and-trap pattern has made clean entries harder to find each session as the overnight move exhausts the day’s range before the regular session opens.
The honest position for the final session: the experiment has demonstrated its thesis, documented its lessons, and built a result of +$31.43 on a $25 starting stake. The final day trade is an option, not an obligation. The most important thing it can document is not another gain — it is the decision-making process that has driven every entry and exit across 28 sessions of real-money trading.
The experiment ends Friday. The war does not.
This blog documents a real-money trading experiment for educational purposes only. Nothing written here constitutes financial advice. All positions involve real risk of loss. Never trade with money you cannot afford to lose.
Next post: Day 29 — The final session. One day trade. The experiment concludes.
