MetaEra: U.S. Markets Rise, Silver Mining Stock Surges 8.69%

On September 2, the three major U.S. stock indices closed higher simultaneously: the Dow rose +0.56%, the S&P 500 +0.47%, and the Nasdaq +0.45%. August ADP private job growth came in at only 38,000, below the expected 47,000, cooling expectations of rate hikes and prompting gains in precious metals and utilities. The stock of the day was Hecla Mining (HL), which surged 8.69% to close at $20.77—while silver prices moved less than 1% on the same day. Today’s U.S. Stock Classroom poses a question: Over a year since mining companies began pivoting to AI, have they truly decoupled from cryptocurrency prices? Tonight at 12:30 UTC, pre-market results will be released by optical communications equipment provider Ciena (CIEN). All data in this article is based on U.S. stock market closing figures as of September 2.

Measured in points and percentage, daily changes are relative to the previous trading day’s close. All three major indices closed higher, with similar gains: the Dow Jones +0.56% at 53,061.95, the S&P 500 +0.47% at 7,667.45, and the Nasdaq +0.45% at 26,217.83. Sector performance showed much greater divergence than the indices: Utilities, as tracked by the Nasdaq, led gains at +2.26%, followed by Communication Services at +1.47%, while Industrials lagged at −1.03%. Capital-intensive, highly leveraged sectors reacted first to easing rate hike expectations—a dynamic consistent with the day’s rise in precious metals.

The real standout last night was Broadcom (AVGO). Revenue for the last quarter came in at $29.6 billion, up 86% year-over-year; AI semiconductors reached $16.7 billion, up 221% year-over-year; and guidance for the next quarter is $34.8 billion (including $21.7 billion in AI)—all figures exceeded expectations. Yet the stock closed down 0.66% at $367.24, plunged about 6% in after-hours trading, then recovered slightly to a decline of around 3.5%.

The divergence isn’t about revenue, but about profit margins: this quarter’s adjusted operating margin was approximately 68%, and the company guided next quarter’s margin to 66%. Revenue tells you how much was sold; profit margin tells you whether the business is becoming more or less profitable—when high growth is already priced in, even a slightly weaker margin guidance can push the stock price down.

Scored on a scale of 0–100, compared against industry peers and its own one-year historical performance, with the benchmark set at the close on September 2. Hecla Mining (HL), a precious metals producer with a market capitalization of $13.9 billion, closed at $20.77, up 8.69%, adding approximately $1.1 billion to its market value in a single day. Trading volume remained flat compared to its daily average—the price increase was driven by valuation, not volume.

Among its peers, it ranked a perfect 100 in peer ranking and scored 81 in peer relative strength, meaning no other asset in its category outperformed it today. However, its trend position was only 47, indicating its price still sits in the middle of its 52-week range. This combination may seem contradictory, but it reflects two distinct metrics: peer ranking measures “who rose the most today”—a horizontal, single-day comparison—while trend position measures “how far it is from its own high”—a vertical, time-based comparison. Only by interpreting these two dimensions separately can you avoid mistaking short-term, event-driven strength for a genuine trend.

The cause was straightforward: August ADP private employment increased by only 38,000, below the expected 47,000, cooling expectations of rate hikes and causing silver prices to rebound after a decline. Mining companies’ production costs remain relatively fixed over a year, while selling prices fluctuate with metal market rates; thus, a 1% rise in metal prices can translate into several percentage points of gross margin improvement, which is then amplified further in stock prices—this is operating leverage. Today’s numbers illustrate this most clearly: silver rose less than 1%, while Hecla jumped 8.69%.

Remember this key constraint: This rally stems from cooling bets on rate hikes, but before the Fed’s September 15–16 meeting, two more data releases—Non-Farm Payrolls and CPI—are still pending, and bets can shift direction with the next data point.

The figures are in percentage and represent the single-day price movement on September 2, compared to the previous trading day’s closing price. All six companies are listed under the precious metals sector, yet the same employment data triggered vastly different reactions: on the silver side, Hecla +8.69%, Endeavour Silver +8.69%, and Coeur Mining +6.04%; on the gold-silver mixed side, Wheaton Precious Metals +4.13%; on the gold side, Newmont rose just +2.06%, while AngloGold Ashanti actually closed down 0.28%.

The difference stems from silver’s dual identity: it is both a precious metal and an industrial metal, used in photovoltaics, electronics, and soldering. Therefore, data indicating economic slowdown and reduced expectations for rate hikes benefits gold primarily through lower interest rates, but affects silver through both lower interest rates and improved expectations for industrial demand—giving silver inherently greater volatility.

All sectors rose 2.42% today, but this average masks internal disparities—just looking at it, you might assume the entire sector gained over two percentage points. When you notice overall sector movement, break it down first to see exactly which part is rising before deciding which asset to focus on—this step is useful every day.

The chart shows the contracted or delivered capacity and corresponding payers for five mining companies; the megawatt figures and lease amounts are contract terms, not current revenue. All five are converting their mining facilities into AI data centers, but the contracted megawatts are only the numerator—the payer is the denominator. For the same long-term contract, who pays determines its true value.

MARA didn’t convert at all; revenue depends entirely on crypto prices: hash rate increased 22% year-over-year in Q2, but revenue fell 27% to $174.9 million, during which it sold 2,213 BTC to bolster cash. Core Scientific (CORZ) was the earliest to convert, having delivered 243 MW under a 12-year lease, but its customer base is nearly exclusively one cloud startup, for which it issued $3.3 billion in project bonds at a 7.75% coupon. TeraWulf (WULF) has signed over 510 MW, with Google backing approximately $1.3 billion in rental obligations. Cipher Mining (CIFR) allocated 600 MW to two payers: Amazon directly signed a 15-year agreement for 300 MW, and Fluidstack signed a 10-year agreement for the other 300 MW, with Google separately guaranteeing $1.73 billion. Applied Digital (APLD) holds 1,410 MW in total, with 1,010 MW backed by investment-grade cloud providers.

They fall together, but diverge when rebounding.

Expressed as a percentage, the value is calculated as (latest closing price − 52-week low) ÷ 52-week low, measuring how much each asset has recovered from its respective low. Cipher Mining rose 114.3%, Applied Digital 89.3%, TeraWulf 72.3%, Marathon Digital 57.2%, while Bitcoin rose 34.7% over the same period, and Core Scientific only 28.0%.

The key is that viewing it from a different perspective yields the opposite impression: from a drawdown standpoint, the six assets are clustered tightly between 38% and 55% below their highs, with minimal variation; but from a recovery standpoint, they range from a 28% rebound to a 114% gain—a fourfold difference. They fall together, but only diverge when rebounding—the volatility has never contracted; what’s often called “decoupling” hasn’t occurred in price, but in contract performance. The market isn’t granting them a label of transformation—it’s rewarding contract quality.

Measured in megawatts, this represents the capacity allocated by CIFR and its corresponding payment parties; the total capacity is 807 MW, with 74% assigned to AI clients. It does not purchase GPUs or sell computing power—only provides electricity, facilities, and grid connection permits—and the entire company has just 66 employees. This sentence is key to understanding it: it doesn’t make money from computing power, but from the power plant.

It split the counterparty into two parts: half, 300 MW for 15 years, directly contracted with Amazon; the other half, 300 MW leased by Fluidstack for 10 years, with Google guaranteeing $1.73 billion in rental obligations. The advantage is that a default by one party won’t collapse the entire arrangement, but the cost is that revenue is locked in by rent—regardless of how much the customer earns from using this power. It’s a business trading certainty for an upper limit.

Putting the five together reveals today’s most valuable takeaway: To assess the quality of an AI contract, first look at who’s paying—whether it’s Amazon, a startup backed by Google, or a startup growing through funding—the credibility varies drastically. This is simply counterparty risk, a well-known concept in crypto, now playing out in a new context. Next, examine concentration: Applied Digital’s 1,010 MW is distributed across investment-grade cloud providers, while Core Scientific’s 243 MW is almost entirely tied to a single one. Finally, consider when the money actually arrives—the contract value represents the total over the next decade, but this year’s cash flow is an entirely different matter.

Units are in billion USD and represent the midpoint of the company’s full-year revenue guidance for fiscal year 2026 (ending October); the leftmost bar is the actual result for fiscal year 2025 at $4.77 billion, while the other three bars are guidance figures, not realized performance. The initial guidance of $5.90 billion was issued in December 2025, raised to $6.10 billion in March, and further increased to $6.30 billion in June, with the midpoint year-over-year growth rate rising from 28% to 32%.

Pre-market earnings for Ciena (CIEN) are scheduled for 12:30 UTC tonight, alongside initial jobless claims and a Federal Reserve official’s speech; the August ISM Services PMI will be released at 14:00 UTC. The true turning point of the week comes tomorrow with the August Non-Farm Payrolls report. The company’s midpoint guidance for this quarter’s revenue is approximately $1.625 billion, up about +33% year-over-year.

Its position is the second leg in AI capital spending: money first flows to chips, then to optical channels that connect computing power—and Ciena specializes in coherent optical transmission and routing/switching equipment between data centers, as well as in metro and long-haul backbone networks. Focus on full-year guidance rather than quarterly revenue because quarterly results can be distorted by delivery timing and recognition points; the midpoint of full-year guidance reflects the company’s own assessment of annual demand—whether it changes or not is a stronger indicator of order visibility than whether a single quarter was met.

Drill down: Which line item supports this financial statement?

The figures are expressed as percentages, representing year-over-year growth rates for each business line in the previous quarter. Total revenue for the previous quarter was $1.57 billion, comprised of $1.10 billion from Optical Networking, $179 million from Global Services, $174 million from Routing and Switching, $94 million from Platform Software and Services, and $23 million from Blue Planet Automation.

Optical networking drove the earnings report, generating $1.1 billion, up 42.2% year-over-year, accounting for 70% of total revenue. However, the fastest-growing segment was routing and switching, at +87.9%, with revenue of just $174 million—about one-sixth of optical networking’s scale. The two segments serve different purposes: optical networking connects data centers, while routing and switching handles the ingress and egress of data from enterprise campuses, riding the wave of AI cluster traffic—smaller in base size but with greater elasticity.

Read the growth rate and scale together: A doubling from a small base won’t move the overall financials, but it signals where demand is spreading; a 42% increase from a large base is the real driver of this quarter’s numbers. Tonight, beyond the total, watch whether routing and switching can continue approaching a doubling, and whether Blue Planet’s −16.4% has stabilized. Also, last quarter’s adjusted gross margin was 44.9%, and the company has guided this quarter to 45%—when revenue grows through direct bulk purchases from large customers, gross margin is the measure of true value.

Q1: Silver prices changed less than 1% today—why did Hecla Mining (HL) rise 8.69%?

Since miners’ extraction costs remain relatively fixed over a year, while selling prices fluctuate with metal prices, small changes in metal prices translate almost entirely into gross profit, which is then further amplified in stock prices—this is operating leverage. By placing “metal price movements” and “stock price movements” side by side, the ratio between them reveals the degree of leverage.

Q2: Both are precious metals, so why is the silver price rising 6% to 9%, while gold is hovering around 0%?

Because silver is both a precious metal and an industrial metal, it is used in photovoltaics, electronics, and soldering. Data indicating an economic slowdown and reduced expectations for rate hikes benefit gold primarily through lower interest rate pressures, but benefit silver through both lower interest rate pressures and improved expectations for industrial demand—giving silver inherently greater elasticity. On that day, the sector rose an average of 2.42%, a figure that masks the disparity between the two ends.

Q3: Why do "Industry Ranking 100" and "Trend Position 47" appear simultaneously in the five-dimensional rating?

Because they measure two different things. Peer ranking compares who rose the most today among similar assets—a horizontal, single-day comparison. Trend position measures how far the price is from its 52-week high—a vertical, time-based comparison. A stock can easily be the top gainer of the day while still trading at only half of its own 52-week range.

Q4: If mining companies have shifted to AI, why do their stock prices still follow cryptocurrency prices?

Because the decoupling did not occur in price. Looking at the rebound from the 52-week low, Cipher Mining rose 114.3%, while Core Scientific only rose 28.0%—a fourfold difference. But when measured as a pullback from the high, all six assets clustered between 38% and 55%. They fell together, and only diverged when rebounding—the volatility range hasn’t narrowed; what’s different is the contract quality, not price correlation.

Q5: When reviewing an AI long contract, which number should you look at first?

It’s not about megawatts—it’s about the payer. The contracted megawatts are just the numerator; the payer is the denominator. A 10-year lease with Amazon, a startup backed by Google, or a startup expanding through funding represents vastly different quality. Next, examine concentration—whether the risk is concentrated with one party or diversified across several. Finally, consider when the money actually arrives—the contract value represents the total sum over the next decade, but this year’s cash flow is an entirely different matter.