Author · brief 2026-06-22

@DratchCap DratchCap

Mechanistic AI-compute-stack bull who models unit economics inside a disinflationary-bubble macro frame

Posts original, reasoning-dense theses on the AI infrastruct

trader score
-2.78
hit rate
30%
mean α
-2.10%
signals 14d
104

Grade = how their written analysis reads (A best). Trader score = how their last-20 timestamped calls performed vs SPY. · Analyst brief as of 2026-06-22.

Their picks, scored

Across their last 20 scored bets: 30% hit rate, -2.10% mean alpha, trader score -2.78. Their last-14d mentions, direction-adjusted, have moved -5.4% since posting (mean over 53 mentions with price data).

Bullish AI infrastructure, memory shortage, and select platforms

DratchCap is concentrated in AI infrastructure and adjacent scarcity trades, especially FRMI/NBIS power-cloud exposure, memory supply tightness via MU/SNDK, and selective mega-cap AI platforms. His distinctive read is that AI cost declines and commercialization constraints expand usage rather than cap the cycle, while power, compliance, and memory become bottlenecks. No clear flip appears in the window; conviction stays bullish with one explicit NVDA hold disclosure and two long calls tied to FRMI.

Themes4analyst read · 2026-06-22
AI power and compliance infrastructure bottlenecks
bullintensifying14 signals
⚠ 43% of theme signals are FRMI — flag pump risk
Memory shortage and take-or-pay cycle reset
bullintensifying8 signals
⚠ 62% of theme signals are NVDA — flag pump risk
AI cost deflation and capex durability
mixedNEW6 signals
⚠ 83% of theme signals are NVDA — flag pump risk
Meta AI product monetization optionality
bullconsistent3 signals
⚠ 100% of theme signals are META — flag pump risk
Direction this week

The author stayed consistently bullish on AI infrastructure, especially FRMI and NBIS, while adding a new memory-shortage basket around MU/SNDK and revisiting META optionality. There is no clear flip or trim in the signals; the only explicit position disclosure is holding NVDA rather than selling. FRMI is the highest concentration and carries pump-risk given repeated bullish and meme-style posts.

Position disclosures1skin in the game
NVDA Implied long NVDA at 13x 2027, holding because selling risks missing upside. held
Best hypotheses1their highest-scoring claims in our index
Hertz valuation may understate non-traditional mobility optionality because investors still price it mostly as a rental-
bullMEDIUM2 co-supporters
Recent signals15receipts included
date (PT)tickerauthorsentwhat they saidsince thenreceipt
2026-07-02META@DratchCap+0.35Says Alex Wang indicated latest Meta model uses much compute and is on par with GPT-5.5.
2026-07-02META@DratchCap+0.75Bullish long-form thesis that AI agents and distribution give Meta major upside.
2026-07-02BE@DratchCap-0.10Sarcastically lists recent AI positive newsflow as if fundamentals were improving.
2026-07-02AWS@DratchCap-0.10Sarcastically lists recent AI positive newsflow as if fundamentals were improving.·
2026-07-02GOOGL@DratchCap-0.10Sarcastically lists recent AI positive newsflow as if fundamentals were improving.
2026-07-02·@DratchCap·Cites historical 8% median bounce for momentum over T+5 after Momo Red Wedding.·
2026-07-02META@DratchCap-0.45Says TMT Momentum unwind is worst since 1999.
2026-07-02MU@DratchCap-0.50Says TMT Momentum unwind is worst since 1999.
2026-07-02NBIS@DratchCap-0.45Says TMT Momentum unwind is worst since 1999.
2026-07-02·@DratchCap·Says innovation and demand are being limited by lack of GPU supply.·
2026-07-01META@DratchCap+0.45Says Doomberg reported nothing new and shakeout is easy for META.-4.9%
2026-07-01META@DratchCap+0.35Implies excess capacity demand is strong enough to absorb META supply.-4.9%
2026-07-01·@DratchCap·Notes momo July performance is already near worst in 15 years.·
2026-07-01AMZN@DratchCap+0.25Argues excess compute concern is inconsistent if cloud majors are not down big.+0.4%
2026-07-01GOOGL@DratchCap+0.25Argues excess compute concern is inconsistent if cloud majors are not down big.-0.4%

Grade is our human read-worthiness rating; trader score is a rolling 20-bet hit-rate/alpha composite — different things, often disagreeing. “Since then” is direction-unaware in the table; the summary line above adjusts for which way they leaned.