SHOC: More Aggressive And Concentrated Chip ETF With Upside Ahead

SHOC-rated Strive US Semiconductor ETF is a “Buy” for concentrated exposure to leading semiconductor names tied to the AI data-center buildout. The portfolio shows ~21% weighted-average price target upside and an EPS growth forecast of 53% for 2027 (PEG 0.76x). Planned weight caps will trim NVDA’s allocation in September and likely lift Micron, while TSM is absent—factors that could shift near-term relative performance across the group.

Analysis

This is more a flow-and-factor rotation than a pure AI call. The important mechanism is that the basket is being nudged away from the highest-multiple GPU leader and toward memory, which means incremental capital would support the names with the strongest near-term re-rating potential rather than the already-owned consensus winner. That makes the ETF potentially useful for catching second-order beneficiaries of AI server spending, but it is a less clean expression of data-center capex than broader semis products because it lacks the foundry lever and is more exposed to one leg of the stack.

The real winner is MU if HBM pricing and capacity remain tight; memory is one of the few places where earnings can inflect faster than estimates because supply discipline tends to lag demand by quarters. The loser is relative performance in NVDA, not necessarily absolute fundamentals: weight caps create forced selling pressure against a name that still dominates index ownership, so any disappointment in gross margin or Blackwell timing would matter more in the next 1-3 months than the ETF itself. Excluding TSM also removes a key hedge against AI infrastructure growth broadening into advanced-node wafers and packaging.

Contrarianly, consensus is assuming AI spend is a single-factor trade, when the more durable trade may be the bottleneck that moves next in the chain. If GPU demand stays strong but memory and interconnect become the constraint, MU can rerate further while SHOC underdelivers versus a more complete semi basket like SMH/SOXX. The thesis is falsified if NVDA guides up sharply again or if memory pricing rolls over before the September rebalance; in that case the ETF becomes a late-cycle momentum product rather than a differentiated AI basket.

Key Decisions for Investors

• Long MU / short NVDA into the September weight-cap reset; use this as a 1-3 month relative-value expression on mechanical ETF demand and AI-stack reallocation.
• Prefer SMH or SOXX over SHOC for core AI semiconductor exposure over the next 6-12 months; SHOC looks more like a tactical satellite than a benchmark-quality beta vehicle.
• If already long NVDA, trim a portion ahead of the rebalance and redeploy into MU or a broader semi basket; the risk/reward is better in the name with under-owned earnings upside.
• Watch the next memory-price and HBM commentary cycle as the key confirmation point; if MU margins fail to inflect, fade the SHOC/MU momentum trade quickly.
• Avoid chasing SHOC solely on the rating upgrade unless ETF flow data confirms demand; without that, the upside is more likely to accrue to constituent relative performance than to the ETF wrapper itself.