Research
The journal of building an honest track in public, the deterministic methodology, and strategies still in research. Transparency is the product.
The Structural Desk · the convergent thesis
The real edge is a structural role, not picking APY
Three deep researches converge on one truth. Two seats are built, one is killed — we publish all three verdicts, including the NO-GO. Start here.
Research desks
Rates Desk — we publish what we refuse
An engine that prices the disagreement between Pendle implied yield, lending rates and perp funding — and refuses books where the high rate is just unpriced tail. The validated 4-sleeve table AND the live refusal log.
RWA Backstop — we lend against Liquidation NAV
The Collateral Safety Board: we measure the real exit value of tokenized-RWA collateral at size ($100k/$1M/$10M), not the marketing $1.00. The honest finding: across the board, none are cash-like on exit; nearly all have ~$0 public on-chain exit (live counts on the board).
Field notes
The method
Methodology
The deterministic engine, LLM-free RiskPolicy gates, and the daily cycle: APY scan → strategies → allocator → risk gate → virtual rebalance → yield accrual → go-live check.
Research-stage strategies
These are research projections, not historical evidence. They accept no allocations until they have point-in-time data and pass a strict backtest.
RS-001
Anti-Crisis Basket
18.2% APY target · projection · moderate-high risk
RS-002
Cashflow / Concentrated LP
29.24% gross target · before IL · aggressive risk
Strategy Lab
The strategy tournament
Every candidate strategy runs a backtest → paper_30d → live pipeline. Promotion requires Sharpe ≥ 1.5, ≥ 7 paper days, APY ≥ 3%, drawdown ≥ -15%. New strategies start advisory (simulate-only) until proven.
Personal research project in paper validation — not investment advice, not a regulated service, not raising capital; results are simulated. Full disclaimer & risk disclosure →