IREN
IREN Limited
Portfolio weight
Not currently held
This position has been exited. The analysis below is kept as a record of the reasoning at the time and is not maintained as current research. See the Decision Log for the full decision history.
Original thesis
IREN was held as power-linked digital infrastructure exposure, combining Bitcoin mining economics with an emerging AI cloud pivot. The thesis was that IREN's low-cost power base positioned it well to transition GPU capacity toward AI workloads as mining margins compressed.
What changed
The position appreciated significantly from the August 2025 entry. After a partial trim in April following a 127%+ gain, the remaining position was evaluated against other opportunities. The risk/reward at elevated prices relative to the AI pivot execution stage made reallocation the better portfolio construction decision.
Why the position was exited
Exited IREN fully on April 30 after the position had run up. The exit was a capital-allocation upgrade, reallocating away from a less attractive risk/reward setup toward better opportunities.
What it taught
Significant appreciation from a speculative entry creates a natural reassessment point. When a position has already captured most of the near-term thesis, holding requires a new and distinct reason to remain sized. Reallocating gains into higher-conviction setups is sound portfolio management.
Relative valuation
Forward P/E
Valuation data pending
Direct-peer average
Pending
Versus its own history
Historical forward-P/E series pending. Maintained by hand in data/fundamentals/manual.ts. No free, reliable source exists for consensus estimates, so nothing is inferred here.
Direct peers
used in the average
Similar enough that comparing multiples is meaningful.
APLD
Applied Digital
90% similar
fwd P/E pending
NBIS
Nebius Group
78% similar
fwd P/E pending
CRWV
CoreWeave
76% similar
fwd P/E pending
CBRS
Cerebras Systems
64% similar
fwd P/E pending
Strategic competitors
never averaged
Competing for the same customers or budget, but too different financially for their multiple to say much about this holding.
ORCL
Oracle
46% similar
fwd P/E pending
Excluded from peer average: competes for the same demand, but a different financial profile
AMZN
Amazon.com
43% similar
fwd P/E pending
Excluded from peer average: too diversified for a clean multiple comparison
GOOGL
Alphabet Class A
27% similar
fwd P/E pending
Excluded from peer average: too diversified for a clean multiple comparison
MSFT
Microsoft
27% similar
fwd P/E pending
Excluded from peer average: too diversified for a clean multiple comparison
Selected by weighted similarity across business model, revenue model, products, end markets, customer type, capital intensity, growth, margins and geography. Direct peers score ≥55% and sit within 2 diversification steps; only they feed the peer average.