海盗鸭

Recently, people have been comparing $MUB , $NVDAB , and $SPCXB again, asking which one to invest a thousand dollars in. The three companies are at completely different stages: Nvidia is the AI shovel-knight leader, but its valuation has already priced in a lot of expectations. Micron is playing the storage cycle reversal logic—high potential and leverage, but also big volatility. SpaceX is an unlisted scarce asset that only just went live on Binance bStock. Instead of agonizing over which one to pick, it’s better to figure out which line you’re betting on—compute power, storage, or space infrastructure. Different approaches lead to completely different answers.

7月29号微软要发Q4财报了，市场上已经有人在喊"三大理由提前买入"。$MSFTB 的逻辑其实一直很清晰：Azure云增速、Copilot商业化进展、还有跟OpenAI绑定带来的AI估值溢价。但"三大理由"这种文章每个财报季都有，真正决定涨跌的还是Azure增速能不能beat预期那两三个百分点。上季度已经证明过AI负载在拉动云收入了，这次如果趋势延续，短期应该还有支撑。

$METAB So far this year it’s down nearly 10%, and the Q2 earnings report is due on July 29. The reasons for the drop aren’t complicated: expectations for ad growth are slowing, and Meta’s AI investment has been too aggressive, leading the market to question the return on investment. But from another angle, even though Reality Labs is still burning cash, the cash flow from Meta’s core advertising business remains solid. At this point before the earnings report, funds are caught in a tug-of-war between those looking to bet on a rebound and those waiting for confirmation. The key is how management guides full-year (or second-half) ad revenue, and whether AI spending will be increased further. With only a few days left until the earnings release, volatility will definitely rise.

$TSLAB Q2 After the earnings miss, it dropped 15% in one go, yet Wall Street’s average target price still implies there’s another 29% upside. This “the more it falls, the more people like it” phenomenon isn’t the first time we’ve seen it with Tesla—analysts’ target prices often reflect the long-term narrative (autonomous driving, robotics, energy), while the market’s short-term selloff focuses on this-quarter deliveries and profit margins. The two sides are not talking about the same thing. A 15% pullback isn’t extreme for Tesla, but if the guidance next quarter remains conservative, no matter how high the target price is, sentiment won’t hold.

Bank of America Securities has adjusted its 2026 target price for $AMDB . We haven’t seen the full report with the exact figures yet, but in terms of direction, the large banks are willing to reset their expectations at this level, indicating confidence that they still expect AMD to make further inroads in the AI-chip market and compete for share. AMD’s issue isn’t that the technology isn’t good—it’s that the rate at which client volumes scale is far behind Nvidia. Adjusting the target price doesn’t mean it will be realized immediately; there are still several quarters of earnings reports in between that need to validate the outlook.

$MSFTB stock price is approaching its lowest level in nearly a year, but the July 29 earnings release could be a turning point. The market is weighing concerns that the return on AI investment may not meet expectations. Whether the Azure growth rate of a few percentage points is enough to dispel that worry is still unclear. Microsoft’s challenge is that it has spent a huge amount of money building out its AI infrastructure, and Wall Street now wants to see whether the revenue side can keep up. In the days leading up to the earnings report, this is actually a window for observation—it's more reliable to watch how institutions rebalance their portfolios than to guess the direction.

Rate-cut expectations and meme hype are both heating up—this combination is pretty interesting. The market is debating which is the privacy coin king, XMR or ZEC, but judging from the charts, XMR’s consecutive all-time highs are already telling the story. Back to the main plot: at $BTC 92000, price has been ranging sideways for several days. $ETH and $SOL are also edging up but not really going anywhere, as if waiting for a directional catalyst. If rate cuts really do come, it should be bullish for risk assets. But right now, the gap between what the Fed is saying and what the market is pricing is still quite large—don’t bet too early.

In pre-market trading, $TSLAB and $GOOGLB continue to dominate the charts. Tesla’s overnight gain was quite notable, and Musk’s "return" narrative clearly hit a nerve in the market. Google, on the other hand, had further breakdowns of earnings details—its cloud business and advertising both showed bright spots, but investors are still questioning the return on its AI investments. The price action of these two companies essentially reflects the market’s two core tensions right now: for Tesla, whether the "founder risk" can be eliminated; for Google, whether "AI investment" can be monetized. The overall direction hasn’t changed, but in terms of timing, there may be some back-and-forth in the short term.

LayerZero and Keeta’s collaboration enables native transfers of tokenized bank deposits between $ETH , $SOL , and Base. This direction holds a lot of imagination—once bank deposits are on-chain, they can flow freely like stablecoins, while still carrying the backing of bank credit. The RWA track has been very active this year; cross-chain interoperability is the final piece of the puzzle. Whoever lays it out first can capture the benefits of institutional capital flowing in.

$AMDB and $INTCB moved sharply after-hours at the same time—AMD’s earnings report is out, and Intel also reported. Seeing these two chip companies side by side is especially interesting: AMD is catching up to Nvidia on AI chips, while Intel is struggling with its foundry transformation. Their directions are completely different, yet both are at critical turning points. If AMD’s data center revenue continues to grow at a high rate, the valuation logic can hold; if Intel doesn’t achieve a real breakthrough in foundry orders, the market’s patience will be further eroded. The divergence in the chip sector will only become more and more obvious this year.

Ripple’s latest move is significant—RLUSD, the stablecoin, has been officially launched via Ripple Mint, and the company has also invested in compliant payment infrastructure Notabene. Both plays point in the same direction: stablecoin payments at the enterprise level. The stablecoin race is now seeing real money being poured in—Circle has just gone public, Tether is continuing to print, and Ripple’s decision to enter through compliance is essentially a differentiated strategy. However, the $XRP price may not immediately reflect these setups in the short term; what the market cares about more is overall risk appetite. Looking over the medium to long term, if RLUSD can win a few major customers, the narrative will be completely different.

$SPCXB Short sellers’ proportion further rose to 32%, and Musk directly warned short sellers on social media that they “won’t last.” The last time he said something like this was $TSLAB , when he was talking during a time when they were being shorted—later, an epic short squeeze indeed played out. But SpaceX and Tesla aren’t exactly the same: liquidity is worse and information transparency is lower, so the logic behind short sellers may be more solid. Still, a shorting ratio of 32% by itself is basically a powder keg—any positive news could ignite it. Both sides have their points; position management matters more than directional judgment.

$ETH The staking exit queue has almost been emptied—nobody wants to unstake. This data is quite interesting: it suggests that at the current price, holders would rather keep their positions locked to keep earning yield than sell out. In a way, it indicates that the lock-up of shares is strengthening. But on the other hand, if the market suddenly deteriorates, once this group of passive holders collectively changes its mind, the selling pressure could become quite concentrated. Short-term bias is bullish, but in the medium term, watch the fragility of this structure.

A developer found that Claude chat logs were indexed by Google, and they even contained encrypted wallet data. The crawler with the number $GOOGLB collected all the shared links. This serves as a reminder of an old problem: by default, if you don’t manually turn off link sharing in AI tools, it’s no different from posting your private key on a public bulletin board. If you’ve used it, promptly revoke the old links—regardless of whether the content includes sensitive information. No matter how convenient the tool is, you can’t lose your security awareness.

$GOLD is consolidating below 4,100 ahead of next week’s interest-rate decision. Gold and crypto have recently started moving in different directions—gold is trading sideways near highs, while the crypto market has been tracking the weakening of U.S. Treasury yields. If next week’s rate-cut expectations fail, both sides may face pressure in the short term. But the medium-term logic is different: gold is supported by central bank purchases, while crypto relies more on liquidity and risk appetite.

RLUSD’s minting platform and compliant network access are good news, but at the same time transfer volume fell 25% to $10.95 billion—this seems contradictory. The $XRP ecosystem has always had this problem: the infrastructure is being built, but actual usage doesn’t keep up. The increase in the number of holders suggests retail investors are entering the market, but has there really been growth in institutional trading demand? When the data conflicts, I tend to trust transaction volume first.

The rising wedge pattern from the technical side has been quite clear on the $ETH daily line already. The target level points toward around 1600. Of course, the chart pattern is only a matter of probability, not destiny. But combined with recent on-chain data, the net inflows from large addresses have been slowing down, and the derivatives market’s long positions are also rather crowded. Signals of short-term pressure are indeed stacking up. If it breaks below the current support, 1600 isn’t impossible—but don’t forget that every time ETH falls hard, people say, “This time it’s different,” and then it rallies back again.

Strive’s SATA products have recovered most of their losses from June. They’re only 3% away from par value right now. These structured products related to $BTC can quickly close the discount, indicating that institutions’ demand for BTC exposure is still there. Compared with last year’s situation where GBTC traded at a long-term, large discount, market efficiency is indeed much better now. Watch whether it can return to above par—that would be an indicator of market sentiment.

Cramer recently highlighted $ORCLB again, saying it deserves attention. To be honest, whether the stock he hypes is accurate or not aside, Oracle is indeed at a point of divergence right now: the bulls are watching for AI order growth and customer lock-in, while the bears are calculating capital expenditure return. The stock price has already run quite a bit this year, and the growth expectations implied by the current valuation are not low. If you only started paying attention to this stock because of something Cramer said, it’s a good idea to look through the last two quarters’ earnings call transcripts first—management’s guidance on cloud ARR is more reliable than any analyst’s mouth.

ETF issuers are now treating Anthropic like the next $SPCXB to chase, trying to replicate the playbook of SpaceX back then—using ETFs to package an unlisted AI unicorn and sell it to retail investors. The logic isn’t hard to follow: Anthropic’s valuation has surged quickly, retail investors can’t buy primary shares, and there’s definitely room for intermediaries to profit from the spread. But keep in mind that the underlying liquidity and the valuation benchmarks for products like this are quite vague—very different from directly buying stocks. With SpaceX, at least there were Starship launch milestones to anchor expectations; with Anthropic, outsiders can’t really see the timeline of commercialization. If you want to jump on board, first understand the fund’s structure before acting—don’t rush in just because it says “AI.”