Let's start with what happened today. The volatility is actually a combination of several mechanical factors, not a single event.

The biggest one is the semiannual Russell Index reconstitution, involving over $12 trillion in tracked assets. Nvidia has replaced Apple as the largest weight in the Russell 1000, and newly large-cap stocks like SpaceX have been included. Passive funds must mechanically rebalance according to the new weightings. That scary-looking volatility at the close was essentially these funds executing mechanical buys and sells according to the rules.

At the same time, there was also the routine quarter-end pension fund selling. This quarter, AI and semiconductors have risen so sharply that many institutions' actual positions have deviated significantly from their target allocations. At the end of the quarter, they need to sell the overweight portions and buy back assets like bonds to bring the ratios back in line. This is purely a routine portfolio management action, unrelated to company performance. Such large capital shifts can cause significant market disruption around month-end. Next Tuesday, June 30, is the last trading day of the month, so the rebalancing is not yet fully complete.

There were also automatic sell orders from CTAs.

These three things combined caused the violent volatility seen at the close today. Such volatility can sometimes present opportunities.

On Friday, the S&P was almost flat compared to yesterday, the Nasdaq closed slightly lower, and the Philadelphia Semiconductor Index dropped 5%.

On Friday, 324 stocks in the S&P 500 rose, only 178 fell, and 8 out of 11 sectors posted weekly gains. This suggests that large-cap funds are gradually returning to other traditional, lagging heavyweights, showing signs of a healthy mean reversion. Money hasn't actually left the market; it's rotating between sectors, with no panic-driven one-way exodus. However, I believe it will rotate back before the semiconductor earnings season.

Today's sector heatmap also reveals a lot. The sector divergence is very clear.

Among the declining sectors, semiconductors fell 3.07%, one of the biggest drops today, related to the previously mentioned dispute between storage and Apple. Internet content and information fell 1.42%, diversified consumer electronics and banks fell 0.96%, capital markets fell 1.63%, asset management fell 0.38%, communications equipment fell 3.39%, and specialized industrial machinery fell 2.47%. Semiconductor equipment and materials fared even worse, dropping 4.63%.

But there were also quite a few rising sectors, with solid gains. Software infrastructure rose 4.13%, general pharmaceutical manufacturers rose 3.82%, application software rose 5.23%, internet retail rose 2.35%, diversified insurance rose 1.89%, consumer electronics rose 3.10%, biotechnology rose 2.11%, computer hardware surged the most at 7.82%, automotive manufacturing rose 1.45%, debt and loans rose 1.42%, entertainment rose 2.43%, non-alcoholic beverages rose 2.00%, healthcare plans rose 2.02%, and medical equipment rose 1.77%.

This divergence reveals the logic of capital rotation: part of the funds are withdrawing from the most crowded semiconductor hardware and moving into software, biotech, consumer electronics, and other sectors that had relatively underperformed recently. Money hasn't left the market; it's rotating internally.

My personal judgment is that this semiconductor correction is temporary. I expect funds to flow back into this sector next week and the week after, especially with the sentiment buildup before SK Hynix's listing, plus the catalyst of the July earnings season. Semiconductors are likely to become the focal point of capital again. This current position feels more like a halftime break.

There were also two intra-industry events today.

Apple publicly complained about high memory costs, triggering a broad decline in Asian tech stocks, and the Korean stock market triggered circuit breakers twice. The market fears that if memory chips take too much profit share, cloud giants may slow capital expenditure, dragging down the entire supply chain. However, Micron's stance is also tough, directly rebuffing Apple, saying that during the 2023 industry downturn, Apple's ruthless price suppression caused today's shortage. This indicates that upstream core hardware manufacturers are regaining pricing power.

On OpenAI, reports say that due to the massive financing scale and competition for funds with capital-intensive firms like SpaceX, bank advisors have suggested postponing the $1 trillion valuation IPO to next year or even 2027. I personally think this is good news, not bad. The IPO delay means the pressure from capital absorption is postponed, extending the life of the current bull market. The affected parties include SoftBank, which holds its shares, investment banks like Goldman Sachs and Morgan Stanley, and Oracle, which will see short-term pressure on their stock prices.

Within the semiconductor sector, there is also divergence—not all chip stocks fell equally.

Although Micron and SanDisk have seen large fluctuations recently, their daily candles have never effectively broken below the EMA20 on pullbacks. If Micron can fill the daily gap around 1086, that would be a good buying opportunity, consistent with the logic that upward gap fills are almost inevitable. Plus, with SK Hynix entering the US stock market on July 10, the memory line remains worth focusing on in the coming period.

AMD and Intel have performed quite strongly, with their stock prices steadily consolidating above the 21-day moving average and EMA20, not following the broader market's sell-off. These two are relatively resilient within the semiconductor sector, indicating that capital within the sector is picking and choosing, not selling everything indiscriminately.

MRVL is also relatively strong, as investment banks raised its target price to 375, and a recent JP Morgan report suggested that customized chip shipments could surpass GPUs by 2027.

GLW and COHR are also relatively strong in the optical sector.

NOK is slightly weaker, currently consolidating around EMA 50. If the broader market stabilizes, NOK's adjustment may be near completion. We can look forward to the earnings report on July 22, which I believe will be an important catalyst, leading to decent stock price gains. Accumulating cheaper shares before the earnings report is advisable.

SK Hynix's US listing deserves separate attention—it's a key variable for the next two weeks.

As Nvidia's largest HBM memory supplier, SK Hynix plans to officially list in the US two weeks later (July 10), with a financing scale of nearly $30 billion.

I believe the semiconductor sector is likely to strengthen ahead of the July 10 listing due to expectations, but on the listing day itself, it could be a "sell the news" inflection point. Be cautious and prepare in advance.

For example: If DRAM rises to 90-95 by July 10, and Micron rises to 1300-1400, consider trimming positions appropriately. Not because fundamentals have deteriorated, but for short-term swing trades with leveraged call options, be careful to avoid "sell the news." Long-term holders can stay put, as SK Hynix's earnings report is on July 29.

Think about it from another angle: If you were company management or an underwriter, holding a stock with such strong performance, needing to raise $29 billion in two weeks, and ensuring a smooth financing with a valuation discount (compared to Micron) likely being compressed to guarantee success, what would you do? You'd probably try everything to maximize hype before the listing to make the pricing look better. So, during this 14-day sprint to the finish, the price will likely rise. Then, with the July 29 earnings report following, these two events together are the most noteworthy time points in the first half of July.

Next week also has several points worth watching.

On the macro data front, from Tuesday to Friday, JOLTS job openings, non-farm payrolls, and unemployment rate data will be released intensively. These employment figures directly affect market expectations for the Fed's path. More importantly, on Wednesday, July 1, newly appointed Chair Warsh will deliver a significant speech, one of his more intensive public statements since taking office. The market will closely watch his wording direction, which connects to the previously tracked logic of constructive ambiguity and downplaying the dot plot.

On individual stocks, Oracle experienced a low-probability mechanical sell-off in recent days. Technically, it's like a fully stretched rubber band, with short-term oversold rebound potential at any time. The worst-case scenario is the weekly 200-day moving average around $145, where it will likely find support.

Both gold and Bitcoin show similar bullish divergence signals. Gold hit a new low, but the RSI printed a higher low. This divergence, where price makes a new low but the indicator does not, often signals a short-term rebound. Bitcoin also experienced extreme selling, consolidating near the lower band of weekly implied volatility for several days. While the coin price made new lows, the daily RSI also rose, showing the same bullish divergence.

Looking at these signals together, the direction is consistent: whether in tech stocks, safe-haven assets, or cryptocurrencies, this round of selling has pushed short-term sentiment to an extreme level. These directions are likely to improve in early July.

Let me share my recent trading actions over the past few days. I've been using the same strategy for four consecutive days.

Each day, I've been buying in batches during the first 30 minutes of the pre-market session. I'll continue to do so next Monday and Tuesday. These are the last two days of pension fund selling. If you haven't used all your bullets, you can follow this rhythm to buy.

I still believe Q will come to 696 to 700. Next week is the 250th anniversary of US Independence Day. Wall Street, or Trump, will likely want to support the market. Sentiment during this period should be relatively positive. Currently, the Fear & Greed Index is around 25. If Q truly dips to 696-700 next Monday and Tuesday, with the Fear & Greed Index simultaneously around 20, that would be the perfect bottom-fishing window in this correction, with both technical and emotional signals confirming.

The S&P has now fallen for four consecutive weeks. Historically, a rebound is likely in the following weeks. Plus, the earnings season is approaching, and I believe this quarter's earnings won't be bad.

I've been buying only common stocks in this bottom-fishing wave, not using options. So I don't have much sensitivity to short-term fluctuations. I have a good mindset and can let it fluctuate as it wants. If you want to use options for leverage, try to avoid short-term Call options. Such contracts suffer rapid time decay and are hard to hold, and your emotions can easily be dragged by volatility. If you must use options, deep in-the-money (ITM) LEAP Calls with 1-2 year maturities are relatively more stable. They are essentially closer to holding the underlying stock and are less likely to be shaken out by short-term fluctuations.

Here are some historical statistics worth including to help judge the significance of the current position.

A set of statistics after the S&P's four-week losing streak: Over the next 1 week, average return 0.8% to 1.2%, with about 62% probability of an up week—belonging to emotional rebounds, technical short covering after overselling. Over the next 1 month, average return 2.1%, with 68% win rate—a period of volatile bottoming, requiring identification of true market support. Over the next 3 months, average return 4.5%, with 73% win rate—entering a steady recovery phase, with fundamentals dominating and funds refocusing on core heavyweights. Over the next 6 months, average return 8.4%, with 78% win rate—returning to a bull track, with an obvious upward trend, often reaching new highs.

A four-week losing streak in the S&P is not extremely rare in US stock history, but it usually means short-term speculative funds have been largely washed out. The longer the time window forward, the higher the probability of positive returns.

Why this four-week drop combined with seasonal factors is worth referencing.

In the last week of June, due to pension funds mechanically selling stocks to buy bonds, historical performance is typically weak, sometimes even accelerating a false breakdown. However, once July arrives, global capital from the new quarter will begin to rebuild positions and flow in. Historical statistics show that the first two weeks of July are among the trading day combinations with the highest bullish win rates of the year.

Looking at all this together, the current position looks more like a low point created by the combination of seasonal patterns and technical overselling, not a signal of a true trend reversal to the downside.

Let me summarize my judgment.

The decline and volatility of recent days are essentially the result of overlapping mechanical fund flows from index reconstitution, pension rebalancing, and quarter-end portfolio window dressing. The market's fundamental picture has not deteriorated. Within the AI supply chain, there is indeed a knockout competition for profit redistribution, but that's a structural change. The improvement in market breadth for two consecutive days, along with the clear divergence and resilience within the semiconductor sector, indicates a healthy mean reversion. Money hasn't left; it's just being mechanically redistributed.

On the operational front, continue buying in batches according to plan. If Q reaches 696-700, that would be the best point to add positions in this correction. After four consecutive weeks of S&P declines, historical patterns and the upcoming earnings season point to a higher probability of recovery. The window before SK Hynix's July 10 listing remains worth monitoring; trim positions if the stock rises too much. Next week's employment data and Warsh's speech are the two biggest variables for the new week, but all the bullish divergence signals seen so far point to improved sentiment in early July.

I'll share S&P 500, Q, and BTC charts later for analysis. Since bottom-fishing depends on the broader market, I'll use the broader market to show everyone how to bottom-fish individual stocks.