TideAlphaTheme Index Breakout
Three themes whose whole index has just made a six-month high, ranked by six-month index return, re-ranked every month.
What it holds
Each theme has an equal-weight index of its holdable members. A theme qualifies at a month end when that index made a 126-day closing high inside the last 10 sessions. The book holds the three qualifying themes with the highest 126-day index return, a third of the book each, split evenly over the members of the theme. A slot no theme fills holds cash. The rule buys a theme only when the whole group has broken out together, so it refuses the months when one or two members carry a theme that a momentum rank would still pick. On 2026-08-21 the live book held T:SHIPPING, T:LUXURY and T:AGING over 10 stocks.
The numbers
CAGR and worst loss on month-end values, so the worst loss is a drawdown and not one month. Sharpe on monthly returns at 0% risk-free. One basis for every row: members holdable after 126 sessions and $1M of daily dollar volume, decisions at the close, fills at the next close, 10 bps a side, idle cash at the 3-month bill.
| Book | CAGR | Sharpe | Worst loss | Window | Basis |
|---|---|---|---|---|---|
| Theme Index Breakout this book | 36.4% | 1.00 | -38% | 2003-12 to 2026-08 | 150 themes, three sleeves |
| SPY buy and hold | 11.1% | 0.80 | -51% | 2003-12 to 2026-08 | SPY, one position |
| Equal-weight theme basket | 17.5% | 0.89 | -51% | 2003-12 to 2026-08 | 150 themes, every holdable member |
- Worst loss without rounding: -37.8%. Annual volatility 38.2% and Calmar 0.96 on the same basis.
- Against 300 random three-theme books on the same calendar the row sits at the 100th percentile for CAGR and for Sharpe; the random median is 14.4% a year. Return over the basket is 20.9 points a year, t 3.14, beta 0.93 over 273 months (results/themestrat_more_d.json).
- Source: results/themestrat_more_d.json, read through crashdash/results/oracle_audit.json (rule_books, generated 2026-08-27), which carries the row of each source file named below.
- Every row is in sample. The rules were chosen with the full history in view, and the theme list is the 2026 list, so themes that died without a surviving member are missing. Live results from 2026-08-31 are the only holdout.
What it means for an investor
- In a bad year it falls about as far as the market, from a much higher level. It lost 29% in 2008, 15% in 2011 and 3% in 2022. Its two worst months are -24.0% (2026-07) and -22.6% (2022-06), and the worst loss on month-end values is -37.8%.
- It is a monthly job. Turnover is 15.1x a year, 50 to 90 trades a year, and the 3.0% of cost is already inside the numbers above.
- It runs no overlay and it does not cut size. Nothing in the limit-drawdown study cut its drawdown without cutting its return. Its worst falls are theme reversals inside rising markets (2018-Q4, 2023-Q3), which no index overlay reaches.
- Volatility is 38.2% a year, about 2.6 times SPY. Size the position so a 40% fall is survivable.
- The row is in sample. The rule was chosen with the full history in view and the theme list is the 2026 list. The live weeks are the only holdout.
Live record
The rule books run on the production server every Saturday at 08:00 New York. Each run writes a dated report with the holdings of every book, the rankings and a scorecard against the equal-weight basket. The book has no overlay, so it holds its three sleeves for the whole month.
Newest run: report_rules_2026-08-21.html. The run reports need an account.
Read more
- Forward test design. Why this book was picked to run forward, and on what basis.
- Theme rotation, more strategies. The study that built the rule (results/themestrat_more_d.json).
- Theme rotation study. The theme rotation study: the universe, the basket and the first engines.
- Oracle audit. The Oracle rule set on the same basis, for comparison.
- The Oracle: rules, evidence and research. The technical page: the stage model, the rules and the papers.
The research reports need an account. The Oracle page at /oracle is public.