# Style and theme rotation: publication review

Reviewed on 2026-10-17. This page consolidates research navigation; it does not report a new backtest.

## Source scope

The research owner retains six source notes at source revision `f0cb03a7d53e0be9b1ebf7c741e56df4bf35055e`. Sources are private; the fingerprints below identify the reviewed documents without publishing private reproduction locations, input panels, strategy implementations or fitting settings.

| Source note | SHA-256 |
| --- | --- |
| weekly_etf_rotation_v1.md | `97b056adf8d3600f30d459515148f3513cfb704718e10bca15c01304c7b567f9` |
| weekly_etf_rotation_risk_cap_exploration_20260916.md | `dc929a47cafe13467a6b286f5d5b7f98458617aaafd197c862c2bb76ba924a80` |
| weekly_etf_rotation_pool_expansion_matrix_20260916.md | `06b374f1df2e2834b40bd3101a816beddfe62c423e9ddb33996363f9de247393` |
| weekly_etf_rotation_holdover_exploration_20260916.md | `a4f348aff1f5e20162fb2182f050111daa086e84f2a7b2772d88bdcc1c0fc89f` |
| weekly_etf_rotation_ridge_exploration_20260916.md | `96e65ec6dd411dd172da580cc8f000aff80715a4b4b366727ebb78adbf1baa67` |
| weekly_etf_rotation_expanded_universe_exploration_20260916.md | `2a7553a186bf880f50fa8e79e0ceaf191b94f03b8f92c833320af2e972657278` |

## Public projection

Allowed fields: descriptive experiment identifier, research group, bilingual display label, annualized return (percent), Sharpe ratio, maximum drawdown (percent), average bond fallback weight (percent, when reported), and executed rebalance count (when reported). Missing values remain absent rather than zero.

The page exposes 24 configurations: 9 risk-cap/allocator configurations, 8 defensive candidate-pool configurations, 3 holding-continuity configurations, and 4 fixed-slot score configurations. They retain separate comparisons. A/B/C are display identities for the three Ridge challengers; private fitting settings are omitted.

The broader industry-universe result is summarized qualitatively, outside the core-pool chart. Legacy ETF and hotspot figures remain bound to the existing digest-validated aggregate asset and its separate review.

## Interpretation and limits

Input coverage is 2015-01-05 through 2026-08-24 for the retained weekly experiments. This does not establish a common first portfolio-return date. Weekly runs assume next-observed-session close execution and 5 bps on gross traded notional. Candidate-pool rows use defensive HRP allocation and a 75% total risk cap; risk-cap rows compare each allocator against its own baseline. Holding-continuity and score rows use fixed slots. Sharpe conventions are not independently recomputed across all notes.

Figures are rounded, transcribed note aggregates. This review checks transcription and publication scope, not the raw ledger, data availability, corporate actions, capacity or executable performance. All experiments reuse historical data. Adding candidates or inspecting annual slices is not new independent evidence. Bond fallback is not cash. A theme-to-ETF mapping advantage remains unvalidated.

No daily NAV is reconstructed from annualized summary values. No private holdings, security-level results, raw vendor data, production settings or live-trading claims are published.