The energy shock is a sharper test for discretionary demand than for cloud software

The market tell

The current geopolitical shock is not producing one clean “risk-off” trade across this basket. As of the September 16 close, SNOW rose 2.49% and RH rose 1.35%, while WSM fell 1.48%; in post-market trading, SNOW was $332.50 at 19:59 ET and RH was $128.50 at 19:59 ET, both above their 16:00 ET closes. DDOG was $231.80 at 19:58 ET, also above its close. These are snapshots, not causal proof, but they are consistent with investors still distinguishing durable software demand from more exposed consumer and freight channels.^[1]

The geopolitical catalyst is unusually concrete. Reuters reported on September 16 that vessel crossings through the Strait of Hormuz remained in the single digits, while separate reporting described weakening buffers in the energy market.^[2][3] CNBC reported that higher oil prices and Treasury yields were squeezing U.S. households, and that diesel prices had climbed above $6 per gallon.^[3] The key question for this group is therefore not whether demand is “resilient” in the abstract. It is which parts of demand can absorb higher energy, logistics, and financing costs without forcing customers to optimize usage or defer large-ticket purchases.

Why the software signal still matters

Datadog’s Q2 FY2026 results provide the cleanest evidence for the positive side of the hypothesis. Revenue grew 36% year over year to $1.12 billion, and management described broad strength across customer sizes, spending bands, and industries. Enterprise new-logo annualized bookings more than doubled from a year earlier, while the company’s AI customer group reached 750 customers.^[4][5]

There is a necessary qualification. Datadog’s management also said that AI customers had moved from validating outcomes toward focusing more on cost optimization. That is a healthy adoption pattern if observability becomes part of the cost-control stack; it is a risk if usage-based customers trim workloads faster than new products and new logos compensate.^[5] SNOW belongs in the same broad debate, but this pass did not establish a comparable, current transcript datapoint for Snowflake, so the software conclusion should not be generalized mechanically across every name.

Why consumer and home furnishings face a different test

RH’s latest reported quarter showed GAAP revenue of $922.2 million, up 2.6% year over year. Its adjusted EBITDA margin included a $55.1 million tariff benefit, or 600 basis points, which makes the headline margin strength less clean as a read on underlying demand.^[6] Management has also described a housing market that has been unusually weak and warned that tariffs and sourcing transitions affect large categories such as furniture, lighting, and rugs.^[7]

That distinction matters for RH, WSM, LZB, LESL, and TPX. A consumer can continue paying for a mission-critical software platform while delaying a sofa, mattress, flooring project, or other discretionary purchase. Higher diesel and shipping costs add another pressure point, while higher yields can slow housing turnover and raise the hurdle for remodeling. The latest tape is mixed rather than conclusive: RH held up better than WSM on September 16, LZB ended slightly lower, and LESL posted a large percentage gain from a very low quoted price. Those moves should be treated as observations, not proof that the underlying demand regime has changed.^[1]

ETH and the limits of a single basket thesis

The requested scope also includes ETH. The quote feed returned a $22.94 16:00 ET close, down 0.48% on the day, but the symbol-to-asset mapping is not sufficiently clear here to use that figure as a clean read on spot Ether or to connect it to the geopolitical catalyst. TPX likewise returned an old timestamp rather than a current September 16 quote. Those data-quality limits are important: a basket thesis is only as strong as the identifiers and freshness behind each constituent.

More broadly, crypto and software may respond to liquidity, rates, and risk appetite differently from furniture and mattresses. Combining them under “resilient demand” risks hiding rather than explaining the drivers. The stronger version of the hypothesis is conditional: recurring enterprise workloads can remain resilient, while rate- and freight-sensitive discretionary demand needs evidence of improving housing activity and stable input costs.

What would confirm or weaken the thesis

Evidence that would confirm it:

• Continued enterprise expansion, new-logo bookings, and AI workload growth at DDOG and SNOW without a corresponding acceleration in usage optimization.
• Consumer companies reporting positive traffic, orders, and conversion that persist after tariff benefits or other one-time margin offsets fade.
• Stabilizing Hormuz traffic, energy prices, diesel, and freight costs, reducing the chance that the geopolitical shock passes through to household budgets.
• A cleaner housing and financing backdrop that releases deferred demand for furniture and home-related purchases.

Evidence that would weaken it:

• More software customers reducing consumption or delaying deployments as AI-cost controls become budget cuts rather than optimization.
• Discretionary brands cutting promotions, lowering guidance, or reporting weaker written sales as energy and financing costs rise.
• Prolonged disruption around Hormuz or additional attacks on energy infrastructure, which would make the input-cost shock broader and harder to absorb. Reuters reported that shipping traffic had already fallen sharply, so duration—not just the first price reaction—is the central variable.^[2]

What to watch next

1. Shipping and energy: daily vessel transits through Hormuz, crude and diesel prices, and any restoration or rerouting of cargo flows.
2. Rates and households: Treasury yields, housing turnover, and commentary on large-ticket affordability.
3. Software usage: DDOG and SNOW disclosures on consumption growth, enterprise bookings, AI workloads, and cost optimization.
4. Furniture and home demand: written sales, traffic, order backlogs, inventories, tariff pass-through, and margin performance excluding refunds or other temporary benefits.
5. Data integrity: confirm the intended instruments for ETH and TPX before drawing portfolio-level conclusions from this basket.

The base case is not that geopolitics has invalidated the demand thesis. It is that the thesis has split in two: software can continue to benefit from operational complexity and AI deployment, while discretionary companies must prove that demand can outrun energy, freight, tariff, and financing pressure. The next few operating updates will matter more than a single day’s price move.