07/20/2026 Trading Recap and Outlook

Good morning, traders. Let’s get started.

Last week was mentally and emotionally exhausting. On the one hand, the market was bombarded with macroeconomic data such as CPI and PPI, forcing investors to repeatedly reprice the path of interest rates. On the other hand, semiconductor positions remained under persistent deleveraging pressure, producing extremely violent intraday swings and repeated back-and-forth price action.

On the trading side, we bought RAM near the lows. Given the event risk surrounding CPI, part of the position was stopped out in advance to avoid excessive volatility, while another portion was sold almost precisely at the local high.

We also proactively reduced our NVDL position to lower the portfolio’s concentrated exposure to semiconductors and high-beta factors.

Among our SOXL short-put positions, the puts with strike prices of 155, 150, and 140 were assigned. The resulting equity positions have now largely recovered to breakeven and are showing small unrealized gains.

In addition, our other major positions in FNGU and MAGS generated substantial profits, and we have been gradually taking profits.

We also used small call-option positions in PLTR and AMZN for tactical trades and took profits promptly after they moved in our favor.

EWZ, meanwhile, fell sharply following the announcement of a new round of tariffs involving Brazil. We followed our discipline and exited the position near breakeven.

Overall, the focus of last week’s trading was not to extract the maximum possible profit from every position. Instead, it was to continuously control exposure, realize profits, and prevent the risk from any single trade from developing into an irreversible portfolio drawdown in an environment of extreme volatility and uncertainty.

In our Discord, I update my latest thoughts every day. There are also many experienced traders in the group who continuously share their own insights. If you have not yet joined the subscriber chat, please click the link below:

https://sidestack.io/franktrading

Without further ado, let’s move on to the next section.

The SPX declined approximately 1.6% this week and closed at 7,457.69. The NDX fell 4.1% to 28,592.66. The DJI declined approximately 0.9% to 52,146, while the RUT fell approximately 0.5% to 2,962.

The VIX, meanwhile, rose 24.9% and closed at 18.77.

The NDX significantly underperformed the SPX. Semiconductors, memory, optical connectivity, NeoCloud companies, and AI-inference beneficiaries broadly suffered double-digit declines.

At the same time, non-AI areas such as financials, energy, real estate, and consumer staples remained relatively stable. This indicates that the market was not experiencing a systemic risk-off event in which all stocks fell simultaneously. Instead, this was a highly concentrated deleveraging event within AI and Momentum exposures.

Energy was the strongest-performing sector, rising 4.85% for the week and ranking first among all sectors. Real estate rose 2.13%, ranking second, while consumer staples gained 1.25% and ranked third.

Financials advanced 0.99%, while healthcare posted a modest gain of 0.18%.

Among the declining sectors, utilities fell 0.43%, consumer discretionary declined 1.18%, industrials lost 1.39%, and materials fell 1.43%.

Communication services declined 2.71%, while technology was the weakest-performing sector, falling 3.61% and ranking last among the 11 major sectors.

Energy was the only sector showing a clear trend advantage, supported simultaneously by rising crude oil prices and improving refining margins.

The strength in real estate and consumer staples suggests that some capital has rotated out of high-volatility technology and into interest-rate-sensitive assets, stable-cash-flow businesses, and lower-correlation exposures.

Financials benefited from bank earnings, stronger trading revenue, and improving M&A income.

Technology and communication services finished at the bottom, indicating that the correction has spread beyond semiconductors into the broader growth complex.

A very important divergence emerged between fund flows and price performance.

During the week ending July 15, U.S. equity funds recorded approximately $4.8 billion in net outflows, their first weekly outflow in three weeks.

However, technology funds still attracted approximately $1.57 billion in inflows. Globally, technology funds received approximately $3.37 billion, although that represented the lowest level of inflows in three weeks.

At the same time, global bond funds recorded their 15th consecutive week of inflows, absorbing approximately $16.16 billion during the week.

This indicates that the decline in technology stocks was not caused by all medium- and long-term investors selling simultaneously.

A more accurate description is that long-term capital continued to allocate to technology, while hedge funds and highly leveraged Momentum strategies rapidly reduced their positions.

According to Goldman Sachs data, U.S. equities experienced net selling equivalent to approximately negative 1.1 standard deviations relative to the past year.

Single stocks saw their largest net selling in three weeks, at approximately negative 1.4 standard deviations. Short selling was approximately 1.9 times as large as long-position reductions.

Technology was both the worst-performing sector and the sector with the largest net selling, with selling intensity reaching negative 1.7 standard deviations.

Financials, healthcare, energy, and industrials, by contrast, recorded net buying.

Therefore, this week was not simply a case of “capital leaving technology.” Three forces were operating simultaneously:

First, AI and semiconductor longs were being forced to reduce exposure.

Second, short sellers began actively targeting companies whose fundamentals and valuations were most vulnerable to debate.

Third, medium- and long-term investors continued to absorb shares of selected high-quality technology companies.

This also explains why some individual stocks repeatedly experienced intraday swings of 5% to 10% without developing into a one-directional liquidity crisis across the entire technology sector.

The divergence across global markets was even more extreme than in the United States.

The European STOXX 600 was roughly unchanged for the week. Because Europe has a lower technology weighting and larger energy and financial-sector weightings, it was significantly less affected by the semiconductor selloff than the United States and North Asia.

The primary pressures facing Europe came from energy prices and the market’s repricing of the ECB’s future rate-hike path.

Across Asia-Pacific markets:

• The Nikkei 225 fell approximately 6.4%.

• The Korean market declined nearly 10%.

• Taiwan fell approximately 6%.

• The CSI 300 declined approximately 5.3%.

• The Hang Seng Index, by contrast, rose approximately 1.6%.

Korea and Taiwan contain some of the world’s most concentrated AI-hardware and semiconductor positioning, making them the regions most severely affected by Momentum deleveraging.

The MSCI Emerging Markets Index has fallen approximately 10% from its June 22 high, while the EM Momentum factor has declined approximately 18%.

Hynix, Samsung, the Korean market, and Taiwanese technology stocks were the largest contributors to the decline.

However, it remains necessary to distinguish between price action and fundamentals.

The tracking estimate for second-quarter emerging-market EPS growth was revised upward from approximately 64% before earnings season to approximately 66%.

Even excluding technology hardware and commodities, emerging-market earnings are still expected to grow approximately 14% year over year, while the median company is expected to grow approximately 15%.

The breadth of earnings revisions has also turned positive.

In other words, the declines in North Asia have been significantly larger than the deterioration in earnings expectations. This is more consistent with position liquidation than with an industry-wide earnings collapse.

The relative strength of the Hang Seng market reflects two factors.

First, positioning and valuations in Hong Kong and China had previously been significantly lower than in Korea and Taiwan.

Second, the Long Korea/Short China and Hong Kong pair trade has begun to unwind.

Investors are selling the most crowded AI-hardware positions while covering the Chinese internet and Hong Kong equity shorts that had previously been used as hedges.

India and Southeast Asia also outperformed on a relative basis.

If the next phase of the emerging-market cycle develops into broader market participation, the most likely beneficiaries will be banks, capital-goods companies, metals and mining businesses, and domestic cyclical sectors—not only TSMC, Samsung, and Hynix.

Brazil experienced some retracement this week, but the underlying rate-cutting and domestic-cycle thesis has not been broken.

Unlike North Asia, Brazilian Momentum has remained relatively resilient since June 22.

EWZ is therefore currently being influenced more by global risk appetite and oil-price volatility than by deterioration in domestic earnings expectations.

WTI rose approximately 15% this week and closed near $82.49 per barrel, while Brent climbed to $88.10.

The immediate driver was the renewed escalation of the U.S.-Iran conflict and the risk to navigation through the Strait of Hormuz. However, positioning also amplified the move.

Before the rally, CTAs had gradually shifted to net-short positioning, while Managed Money net exposure had moved close to neutral.

After oil broke higher, short covering, systematic Momentum strategies flipping long, and inflows into energy funds created a positive feedback loop.

Since July 10, crude oil open interest has declined by approximately $1 billion, also indicating that the rally included substantial short covering.

However, the options market has not become as consistently bullish as the spot market.

The one-month 25-delta skew has shifted back in favor of puts, while both the put/call ratio and put open interest have increased.

This means institutions are acknowledging near-term supply risks while simultaneously hedging against a retracement following the sharp rise in oil prices.

Weekend risk increased further.

The United States carried out strikes against Iran for an eighth consecutive night. Earlier, an Iranian attack in Jordan killed two U.S. service members and left one missing.

Markets in Qatar, Bahrain, and Kuwait all declined on Sunday.

The risk of a gap higher in oil prices next Monday is therefore materially greater than it was last week.

My assessment is that crude oil is currently being driven by a combination of actual military conflict, supply logistics, and a short squeeze.

After a 15% rally, it is not appropriate to chase the market indiscriminately following a gap higher.

The most important variables to monitor are:

• Actual vessel traffic through the Strait of Hormuz

• Middle Eastern crude-loading volumes

• Saudi exports through alternative routes such as Yanbu

• War-risk insurance premiums and VLCC freight rates

• Brent front-month spreads

• Refining crack spreads

If oil prices continue rising while front-month spreads and actual shipping volumes do not deteriorate in parallel, it would indicate that an increasing share of the additional rally is being driven by risk premium rather than by a physical supply shortage.

Gold declined approximately 2.8% this week and closed near $4,010, while silver fell approximately 6.5%.

Despite a clear escalation in geopolitical risk, precious metals still declined.

This suggests that the marginal pricing power remains in the hands of real interest rates and monetary-policy expectations rather than war-related safe-haven demand.

The U.S. 10-year real yield remains around 2.34%, which is extremely high in absolute terms.

As long as real yields fail to decline on a sustained basis, gold will struggle to establish a durable trend based solely on geopolitical risk.

Copper remained at elevated levels, with little overall change in price. However, exchange inventories continued to decline, indicating that near-term physical-market conditions have not deteriorated materially.

Aluminum was roughly unchanged.

Industrial metals are currently being influenced by two opposing forces:

• AI, power grids, data centers, electrification, and global capital expenditure are supporting demand.

• Slowing Chinese growth, a strong U.S. dollar, and declining global risk appetite are suppressing valuations.

Copper is therefore not purely a China macro trade.

As long as orders related to data centers, power grids, and capital goods remain strong, copper may continue to show resilience even when Chinese economic data weakens.

Corn, soybeans, and wheat generally strengthened on Friday, with wheat outperforming.

The market is reassessing U.S. planted acreage, weather conditions, and export demand.

Two medium-term risks require continued monitoring.

First, the transmission of the Gulf conflict into natural-gas, urea, and nitrogen-fertilizer costs.

Second, the potential impact of a strong El Niño on the Indian monsoon and on the production of sugar, coffee, palm oil, and grains.

Agricultural commodities have not yet established a trend as clear as crude oil, but their volatility may already have bottomed.

U.S. Treasuries experienced an initial bear steepening followed by a bull steepening during the week.

On Monday, oil prices and expectations of rate hikes pushed the 2-year yield as high as approximately 4.28%, while the 10-year yield rose to approximately 4.61%.

However, after CPI and PPI came in cooler than expected, expectations for near-term rate hikes declined significantly at the front end.

By Friday, the 2-year yield had fallen to 4.18%, the 10-year yield was around 4.55%, and the 30-year yield remained as high as 5.06%.

The 2-year/10-year spread widened to approximately 37 basis points.

The final weekly pattern was a modest bull steepening:

• The 2-year yield fell approximately 3 basis points from the previous week.

• The 10-year yield declined only approximately 1 basis point.

• The long end clearly refused to follow the front end significantly lower.

June CPI declined 0.4% month over month and rose 3.5% year over year.

Core CPI was unchanged month over month and slowed to 2.6% year over year.

PPI declined 0.3% month over month, although the core measure excluding food, energy, and trade services still increased 5.1% year over year.

Retail sales rose 0.2% month over month, indicating that consumer demand has not yet materially stalled.

Lower energy prices temporarily suppressed headline inflation, while core inflation improved. However, the economy and underlying demand remain resilient.

This was sufficient to reduce the probability of an immediate July rate hike, but it was not enough to produce a sustained decline in long-term yields.

The U.S. Dollar Index fell approximately 0.2% for the week, while the 10-year and 30-year yields remained above 4.5% and 5%, respectively.

This indicates that fiscal supply, the term premium, and long-term demand for capital continue to pressure bond prices.

For equities, the most important development was not the decline in the 2-year yield from 4.28% to 4.18%.

The key issue is that the 10-year real yield remains above 2.3%.

As long as the real cost of capital remains elevated, the market will continue to demand that AI companies prove their capital expenditure can be converted into cash flow.

Compared with the continued pressure from elevated long-term real yields, credit markets remained relatively calm this week.