Alphabet delivered impeccable growth on its most important business segment: Google Cloud reported year-over-year growth of 82%: $24.8B v $13.6B.

However, it came at a cost.

Investors decided to prioritize the cost over the growth, shaving 7% off the stock.

Depends on whether you are an investor or trader.

Don’t be a hero. Your money doesn’t pay you to do that.

GOOG now sits below its major moving averages. Statistically speaking, volatility is higher below the 200-day SMA (200d). Assuming you are interested in putting money to work here — initiating a new position or increasing a current one — you want to wait for the stock to sustain price above that 200d for a few sessions.

If downside continues, look for GOOG to find support at the 100-week SMA. It is the next major level of support for the stock. However, I would still wait for the stock to reclaim the 200d. At least for now, I won’t trust a recovery in the stock without a recovery of that 200d level.

Oh, and one last thing, don’t take its trailing ~16x P/E at face value.

Roughly $6.26 — about 69% — of Alphabet's $9.11 reported Q2 2026 diluted EPS came from unrealized (paper) gains on equity securities: ~$99.0 billion in the quarter, $21.9 billion via Anthropic, $77.1 billion via Space X.

In the interest of fairness, it is worth noting that the market isn’t oblivious to this. You can find in page 9 of the report. These stakes should be worth something. These interests benefit Alphabet’s shareholders. We give Berkshire credit for its equity portfolio after all, despite the main business being insurance.

The point I am making is that the “cheap P/E” argument isn’t as clean as bulls would have you believe. At the same time, they shouldn’t be worth nothing or overshadow the tremendous underlying performance (80%+ on cloud) as bears would have you believe.

If you were long going into the quarter and you are disciplined, you have already been stopped out. The best advice I can give any trader — regardless of their experience — is to avoid taking the knockout punch. Capital is finite. Take what’s left of it, and find a better opportunity. We’ll talk about those in a moment, but first, a final question.

I wish I could be more constructive.

If GOOG is any indication, this market is more keen to punish for CAPEX than reward for growth. Consensus is that none of the hyperscalers — remember, AAPL is not a hyperscaler — is slowing CAPEX this quarter. If Google Cloud’s 82% growth, far surpassing the whisper of ~70%, wasn’t enough to justify the spending, I have no idea what level of growth will.

As such, as both a trader and an investor, I am waiting until after their quarterly reports before taking a look. My cautious posture is also informed by the technical damage done this week.

Prior to the report, the cohort was clearly building momentum: a yellow light about to turn green. Afterward, it’s the opposite: a yellow light about to turn red. You can make an argument a topping pattern is emerging.

That said, rotation and dispersion are alive in the MAGS. Apple has been the biggest winner, closing in on new all-time high despite the broader weakness. Nvidia is perking up too. As we know, this market has been extraordinarily rotational. In the two days since Alphabet’s report, there has been a clear standout: industrials.

Better to show than tell. Let’s take a look at the heat maps and the sector performances since GOOG’s quarter.

Evidently, risk-on/AI rotation has found a new favorite in the industrials.

N=2. I know. Early days.

But the fact that semiconductors, memory, and semiconductor equipment names (within the XLK) didn't catch a major bid after Alphabet and Tesla confirmed the check-writers weren't putting away their check-books anytime soon — while industrials regained momentum — is something. Money wanted to chase AI risk. And it didn't choose the semis; instead, it chose the industrials.

And, the Intel (INTC) disaster only adds to this feeling that rotation has a new favorite. The stock immediately went higher after the quarter in the post-market, lost the gains over night, opened the following session lower, and kept falling. Rumor has it INTC would still be falling now, if not for the fact that the stock market closes at 4PM… and some people want this thing to run 24/7… careful what you wish for.

Further complicating the buy-thesis for semiconductors is the major speculative unwind occurring in Korea:

“Last week as over 1.2 million South Korean retail trading accounts triggered margin calls— equaling roughly 1 in every 30 working-age adults — following a brutal market crash driven by an unwinding of heavily leveraged bets on chipmakers like Samsung Electronics and SK Hynix.”

Source: Bull Theory

When I put it all together, I conclude investors are concentrating in a different picks-and-shovels AI play: the industrials, the companies actually building the data center. The top 5 names are Caterpillar (CAT), GE Aerospace (GE), GE Vernova (GEV), RTX (RTX), and Boeing (BA).

As for the technical situation, it looks healthy. On Thursday, as Alphabet and Tesla sparked a tech sell-off, the XLI rallied out of a downtrend. I want in. I am waiting for the ETF to make a new all-time high or retest the 50-day SMA (50d). The 50d aligns with resistance in March and support in June. If this level shows itself as support again, that is another indication of a strong bull trend in place and would be enough for me to ante up.