Market snapshot · September 18, 2026 close
S&P 500 (SPY) $761.69 −0.13% Nasdaq-100 (QQQ) $721.45 +0.60% Dow (DIA) $515.88 −0.48% Russell 2000 (IWM) $284.10 −0.52% 10-Year Treasury (IEF) $90.80 −0.49% Crude Oil (USO) $153.82 −0.93% Gold (GLD) $401.17 +0.74% US Dollar Index (UUP) $28.39 −0.02% Volatility (VXX) $17.76 +0.31% Semiconductors (SMH) $573.00 +2.17% Silver (SLV) $59.93 +1.65% Emerging Markets (EEM) $67.03 +0.19% Bitcoin (BTC) $81,055.00 +4.51% Ethereum (ETH) $2,630.20 +5.92% S&P 500 (SPY) $761.69 −0.13% Nasdaq-100 (QQQ) $721.45 +0.60% Dow (DIA) $515.88 −0.48% Russell 2000 (IWM) $284.10 −0.52% 10-Year Treasury (IEF) $90.80 −0.49% Crude Oil (USO) $153.82 −0.93% Gold (GLD) $401.17 +0.74% US Dollar Index (UUP) $28.39 −0.02% Volatility (VXX) $17.76 +0.31% Semiconductors (SMH) $573.00 +2.17% Silver (SLV) $59.93 +1.65% Emerging Markets (EEM) $67.03 +0.19% Bitcoin (BTC) $81,055.00 +4.51% Ethereum (ETH) $2,630.20 +5.92%

Opinion · August 2026

Not Every Rally Comes With the Same Kind of Explanation

August 13, 2026 · Analysis, prompted by Thursday’s three-name rally

SanDisk rose 13.67%, Netflix rose 5.43%, and Datadog rose 4.70% Thursday — three green headlines that read the same way on a price chart. They are not the same kind of evidence. Netflix moved on a discrete, dated, falsifiable fact: Bill Ackman’s Pershing Square disclosed it bought the stock. SanDisk moved on a retrospective narrative — two separate articles arguing its prior 45% decline was overdone, built on projections rather than a new fact about Thursday itself. Datadog moved on nothing I can find at all. Treating all three as equally “explained” would be a mistake.

A hard catalyst, a soft narrative, and no explanation

The Ackman disclosure is the cleanest case: a specific investor took a specific, dated action, and the market repriced Netflix in response. Whether or not Ackman turns out to be right about Netflix’s moat, the fact itself is verifiable and happened Thursday. SanDisk and Micron sit in a murkier middle ground. The coverage driving their rally wasn’t a new fact about either company — no new contract, no earnings surprise, no guidance change — it was commentary re-litigating a decline that had already happened, arguing the memory-shortage thesis justifies a higher price than the market had been assigning. That's a real argument worth taking seriously, and it happens to be adjacent to the same AI-infrastructure capacity constraints I examined in the Physical Limits of Compute whitepaper. But an argument published the same day a stock happens to rally is not the same evidentiary weight as a disclosed transaction, even when both show up in the news feed labeled the same way. Datadog, meanwhile, offers no argument and no fact — just a fourth large, unexplained swing in six sessions, following the pattern I flagged Tuesday and first wrote about on August 6.

Where this connects to my own thesis

I hold all three names as Core positions, and the point of sorting evidence this way isn’t to trade around it — it’s to avoid letting a well-written narrative piece feel as conclusive as a disclosed fact just because both arrived on the same day a stock went up. The memory-shortage argument for SanDisk and Micron may well be correct; I already own both and haven’t changed that view. But I should hold that view with exactly the confidence the underlying evidence supports, not with the extra confidence borrowed from a green candle on a chart.

This is an opinion piece reflecting my own interpretation of Thursday’s price action, built on the reporting cited in full on the linked News article. It is not investment advice.