Seven indexes. Documented selection.
Custodios Index's catalog of seven indexes was selected from 100 candidate positional configurations. Six met our strict performance criteria; a seventh — CI-WORLD500 — is published as our broadest, lowest-volatility positional benchmark.
One hundred candidates. Seven survived.
We analyzed 100 candidate positional indexes. Six met our strict performance criteria (Sharpe above 0.70 with meaningful alpha). We publish a seventh — CI-WORLD500 — as our broadest, lowest-volatility positional benchmark. It prioritizes diversification and stability over maximum risk-adjusted return, serving investors who want the widest positional exposure. The performance criteria applied to the six were:
- 01. Outperformance. 15-year CAGR must exceed the S&P 500 (13.60%) by a meaningful margin.
- 02. Risk-adjusted return. Sharpe ratio (using 4% risk-free rate) must exceed 0.70.
- 03. Structural defensibility. Methodology must be purely positional, equal-weighted with reasonable caps, and capable of being defended philosophically across centuries.
CI-WORLD500 is included as a broad benchmark tier — disclosed openly with its 0.67 Sharpe — for investors who prioritize breadth and stability. The complete analysis, including candidates that did not meet criteria, is published transparently in our research section.
Quarterly rebalancing with full-quarter confirmation.
Custodios Index rebalances quarterly, following the global institutional standard. The methodology distinguishes between constituent changes (which require full-quarter confirmation) and weight adjustments (which restore equal-weighting at every rebalance).
Quarterly rebalance dates
Third Friday of March, June, September, and December.
Rebalancing occurs on the third Friday of March, June, September, and December — the same dates used by S&P 500 and other major institutional indexes. Changes are announced five trading days before the effective date.
This schedule aligns Custodios Index with the broader index ecosystem and ensures operational compatibility with potential ETF replication.
Constituent changes — full-quarter confirmation
Constituent changes require confirmation across a full quarter. A new position must hold its target rank for the entire quarter before being incorporated into the index at the next rebalance.
This buffer mechanism prevents the index from reacting to temporary fluctuations — what we call “positional wobbling.” For example: if Apple briefly overtakes NVIDIA as the world's #1 by market cap mid-quarter but NVIDIA recovers by quarter end, no change is made.
This approach is structurally similar to the buffer zones used by MSCI's Global Investable Market Indexes methodology and reflects industry best practice for reducing unnecessary turnover.
Weight rebalancing — quarterly equal-weight reset
At each quarterly rebalance, weights are reset to maintain equal-weighting with the methodology's defined caps:
- — CI-WORLD1: 100% (single holding, no cap applicable)
- — CI-WORLD3: equal-weighted with 35% cap per constituent
- — CI-WORLD10: equal-weighted with 12% cap per constituent
- — CI-DECILE1: 10% per decile, equal-weighted within deciles
- — CI-WORLD100: equal-weighted with 1.5% cap per constituent
- — CI-FUTURELEADERS: equal-weighted with 2.5% cap per constituent
- — CI-WORLD500: equal-weighted with 0.5% cap per constituent (500 holdings)
This systematic reset captures the documented benefits of equal-weighted rebalancing: contrarian tilt toward underperformers (mean reversion) and discipline against momentum extremes.
Emergency triggers
Outside of regular quarterly rebalancing, Custodios Index applies emergency triggers for material events:
- — Bankruptcy or insolvency proceedings: immediate removal
- — Mergers and acquisitions: target company removed at deal closure; acquiring company evaluated for inclusion at next quarterly review
- — Delisting from major exchanges for more than 30 days: immediate removal
- — Spin-offs of material business segments (greater than 25% of market cap): both parent and spin-off evaluated at next rebalance
- — Sustained ranking collapse: a constituent that falls outside its target range for two consecutive quarters is replaced
Emergency changes are announced as soon as practical, with effective dates determined by the nature of the event.
Industry alignment
Custodios Index's quarterly methodology aligns with established institutional practice:
- — Major US large-cap benchmarks adopted quarterly rebalancing in the 1990s (3rd Friday of Mar/Jun/Sep/Dec).
- — Major US large-cap equal-weight benchmarks reset weights quarterly, with tens of billions in tracking products.
- — Major developed-market global benchmarks transitioned to quarterly index reviews in 2021.
- — Other major international index families are moving toward more frequent rebalancing schedules through 2026.
By following this convention, Custodios Index ensures operational compatibility with potential ETF replication and reflects what the institutional indexing industry has converged on after decades of iteration.
Discipline over velocity. Disclosure over confidence. Position over name.
Why equal weight.
Custodios Index weights every position equally, subject to caps — not by market capitalization. This is a deliberate choice, and the reason we outperform cap-weighted benchmarks over the long run.
Each position counts as a position
Our philosophy is “Just positions. Nothing else.” If we weighted by market cap, our flagship CI-WORLD10 would be dominated by its three largest holdings, and positions #7 through #10 would barely register. That would contradict the entire premise: the #10 global position is still a position of global leadership, and it counts. Equal weight is the only weighting scheme consistent with positional indexing.
You capture the full upside
A common misconception is that cap-weighting captures more of a winner's rise. It does not. When a holding rallies during a quarter, an equal-weighted index rises with it in full — you hold your allocation in that company through the entire move. You capture all of the intra-quarter upside.
The rebalance is where the advantage compounds
At each quarterly rebalance, equal weighting resets allocations. In practice this means trimming the holdings that ran up and adding to those that lagged — systematically selling high and buying low within the index. This contrarian discipline is automatic, unemotional, and repeated every quarter. It is the documented source of the equal-weight premium observed across decades of market history.
Cap-weight concentrates risk at the top
Cap-weighting does the opposite. As a company rises, its weight rises with it — leaving the index increasingly exposed to a handful of names precisely when they are most expensive. When those names correct, a cap-weighted index falls with its full concentrated weight. Equal weight, having trimmed at the peak, absorbs less of that drawdown.
Caps prevent the extremes
Pure equal weight at 500 holdings would create unnecessary turnover and liquidity strain. Each index therefore applies a cap (for example, 12% on CI-WORLD10, 0.5% on CI-WORLD500). Caps keep the index investable while preserving the equal-weight logic.
Equal weight is not a stylistic preference. It is the weighting scheme that aligns with our philosophy, captures full upside, and compounds a contrarian advantage at every rebalance.
This describes index construction methodology. It is not investment advice. Past performance, including the historical equal-weight premium, does not guarantee future results. See /legal.
The 1924 / 2024 / 2124 test
Custodios Index indexes follow one design rule: they must make sense in 1924, today, and 2124.
- — Country-specific indexes (countries change in relevance).
- — Sector-specific indexes (sectors can disappear or be redefined).
- — Theme-based indexes (themes are by definition temporal).
- — Factor-based indexes (factors can be arbitraged away).
- — Pure positional indexes (“the top N”).
- — Cap-weighted or equal-weighted by transparent rules.
- — Quarterly rebalancing based on market data.
If an index requires the world to look like it does today, we don't build it. If an index works regardless of which companies, countries, or sectors lead at any moment, we build it.
What we mean by “data”.
Custodios Index distinguishes three types of data, each with different epistemological status. We disclose these distinctions because the difference matters.
Real Data
- Long-horizon US large-cap historical returns from publicly available academic datasets (1928–2025).
- Daily market prices calculated using publicly available market data.
- Constituent identification using real company tickers.
- Market cap and fundamentals from publicly available market data and public regulatory filings.
Calibrated Estimates
- MSCI World pre-1969 (constructed from factor models calibrated to post-1969 real data).
- Nasdaq 100 pre-1985 (constructed from factor models calibrated to post-1985 real data).
- S&P 600 pre-1994 (constructed from factor models calibrated to post-1994 real data).
- These are explicitly labeled as estimates wherever shown.
Designed Methodologies
- Expected returns based on documented factor premiums (Fama-French 1992, Novy-Marx 2013, Anderson-Reeb 2003, Mauboussin 2014).
- Forward projections using historical factor returns.
- Index values calculated daily from real constituent prices applied to Custodios Index's defined constituent lists and rebalancing rules.
We disclose these distinctions because the difference matters. Past performance — real or simulated — does not guarantee future results.