Outperformance vs BTC: Counting the Winners

Date: 2026-03-18 Author: OLTA Research Abstract: Of 74 indices in the OLTA library, 23 beat the BTC Sharpe ratio of 0.14 over the 2-year window ending 2026-05-21. The overwhelming majority of those outperformers are Equities-family baskets that have effectively zero correlation to BTC by construction, supplemented by the five Live Diversified-family baskets that explicitly mix equities into a crypto base, and a set of recently-listed RWA baskets whose short windows inflate their headline Sharpe figures. The conclusion: in this regime, cross-asset construction is the only reliable path to BTC outperformance. The result is regime-dependent, not universal.

1. The count

Filtering the full backtest library for Sharpe strictly above the BTC reference of 0.14 yields 23 names. The conservative count, which restricts to baskets with a backtest window of at least 365 days, yields 14 names.

2. Family breakdown

The 23 strict-outperformers, by family:

• Equities: 12 of 15 indices.
• RWA: 7 of 8 indices. The eighth has the only full 2-year history in the family and is modestly negative.
• Diversified: 5 of 6 indices. The sixth is in Watchlist pending longer history.
• Strategy: 1 of 12 indices (a low-volatility expression).
• Core, Sector, Ecosystem, ThematicBeta, Curio: zero outperformers.

The pattern is direct: the families that beat BTC are the ones with structurally low or negative BTC correlation. The families that fail are the ones with high BTC correlation.

3. The conservative count

Filtered to baskets with a backtest window of at least 365 days, the 14 strict-outperformers break down as 8 Equities baskets, 5 Diversified baskets, and 1 Strategy basket. The full per-index ranking is documented in the methodology brief on request.

Every Equities outperformer in this filtered set has correlation to BTC near zero by construction (the Equities family averages approximately -0.04). Every Diversified outperformer has correlation to BTC between -0.03 and -0.08.

The conclusion: in this 2-year window, the only baskets that beat BTC's Sharpe were baskets whose construction did not include a meaningful BTC correlation. The crypto-pure baskets (Core, Strategy except the low-volatility expression, Sector, Ecosystem, ThematicBeta, Curio) all underperformed BTC on a risk-adjusted basis.

That is not a finding about the OLTA library. It is a finding about the market regime.

4. The regime fingerprint

The 2-year window of the report covered:

• BTC: +11.2 percent total, max drawdown -49.5 percent, Sharpe 0.14. Effectively sideways.
• S&P 500: roughly +25 percent total, max drawdown around -15 percent, Sharpe in the high-1.0 band over the same window.
• Gold: approximately +50 percent total return.
• Mid-cap and small-cap crypto: -50 to -80 percent drawdowns, repeated. Most baskets ended below their starting NAV.

This is a US-equity-bull, gold-bull, BTC-flat, alt-crash regime. Cross-asset construction with low BTC correlation was the empirically correct positioning for this regime.

A 2017-2018 window would have shown:

• BTC up roughly 17x in 2017, then down -84 percent in 2018. Net positive over 24 months.
• S&P up roughly +20 percent in 2017, then -6 percent in 2018. Net positive but small.
• Gold flat.
• Most alts up multiples in 2017, then down -90 percent in 2018.

In a 2017-2018 window, the Diversified family would have underperformed BTC on absolute return. The Diversified Sharpe might still have been similar (drawdowns were also smaller because of the equity and gold legs), but the relative ranking on absolute return inverts.

A 2020-2021 window would have shown:

• BTC up roughly 6x. S&P up roughly 30 percent. Gold up roughly 25 percent. Most alts up 5-10x.
• The Sector and Ecosystem families would have dominated.
• The Equities family would have lagged.
• The Diversified Sharpe would have been roughly similar, but the absolute return would have been the smallest of the high-Sharpe set because the gold leg flattened the upside.

The honest reading: cross-asset construction works in regimes where at least one of equity, gold, or crypto is doing meaningful work and the other two are not catastrophic. It does not work in a regime where one asset class moves so much that the rest are pure drag.

The 2-year window of this report is the regime where cross-asset construction works best. The conclusion is real and the result holds across four stress windows. But it is not regime-agnostic.

5. The absolute-return contrast

A separate ranking by total return (rather than Sharpe) shifts the leaderboard. The leaders are Equities baskets with full 5-year history, several of which delivered total returns in the +300 to +760 percent range. The Diversified family entries appear in the middle of the return ranking by design: the construction explicitly sacrifices some absolute return for risk control. The Sharpe-rank-leader Diversified basket returned in the high-double-digit range, at roughly half the volatility and half the drawdown of the absolute-return leaders. Detailed return-rank tables are in the methodology brief on request.

6. The implication

The headline of this report is the Sharpe outperformance. The absolute-return ranking tells a complementary story.

Story 1: For a Sharpe-maximised portfolio in this regime, cross-asset Diversified is hard to beat.

The Sharpe leader within the multi-year-window leaderboard is a Diversified basket with a Sharpe in the [1.5, 1.6] band. It is the cleanest cross-asset expression of the highest-Sharpe single-asset positions in the OLTA opportunity set.

Story 2: For an absolute-return-maximised portfolio in this regime, Equities-only is the path.

The high-return Equities baskets are pure-equity expressions with no diversification beyond the within-family equity correlation structure. The drawdowns are larger (typically 30 to 80 percent) but the absolute returns are several multiples of BTC's.

Story 3: BTC by itself is in the middle of the pack in this regime.

A 100 percent BTC allocation produced 0.14 Sharpe and +11.2 percent total return over 2 years. It is outperformed on both Sharpe and absolute return by every basket in the strict-outperformer set, and it is outperformed on drawdown by more than half of them. The takeaway is that BTC-by-itself is no longer the highest-conviction crypto-related position in 2026; the cross-asset wrapper or the cross-equity wrapper are.

7. Where this breaks

The result reverses in a hypothetical 2017-2018 window, a 2020-2021 window, or any future window where BTC's drawdown ends in a strong recovery within the basket's holding period. The report does not claim that cross-asset will beat BTC in all regimes; the claim is specifically that in this window (2024-05 to 2026-05) it did, by a margin that holds across:

• Multiple rebalance strategies.
• Four crisis stress windows.
• The full 2-year holding period.

A future window with a different BTC return profile could and likely will produce different relative rankings. The Diversified family is constructed to absorb both directions (it has a meaningful crypto leg), but it will not outperform a pure-BTC position if BTC delivers a 17x bull run.

8. Construction commonality across the outperformer set

The outperformer set is unified by a single construction principle: low BTC correlation, real-world-asset or equity-asset backing. The family labels differ but the structural positioning is identical at the daily-return horizon. Tokenised commodity ETFs and tokenised treasuries behave like the Equities family at the daily-return horizon despite carrying an RWA family label.

9. Caveats

• Window dependence. Already discussed. The Sharpe outperformance is real but window-specific.
• RWA short windows. Several outperformers have 57-67 day windows. Their Sharpe ratios are statistically less reliable than the multi-year baskets and could regress materially over the next 6 months of data.
• Survivorship in the constituent set. Equities and Diversified baskets use the constituent universe that exists today. Underlying companies that went private or delisted in the 5y window (rare for the underlying universe) are excluded by construction. This biases the comparison slightly in favour of the equity-leg families.
• Risk-free rate. Sharpe is reported at zero risk-free. At a 3.8 percent USD overnight rate (mid-2026), the BTC Sharpe shifts from 0.14 to roughly -0.04 (the basket's annualised return falls below the risk-free rate). The Diversified family Sharpes shift down by approximately 0.16, leaving the Sharpe leader at roughly 1.38 versus BTC's roughly -0.04. The outperformance gap widens at a positive risk-free rate. The ranking is unchanged.
• Synthetic comparison. The "Diversified would have done in 2020-2021" thought experiment in section 4 is qualitative, not numerical. The engine does not synthesise that period because the crypto-data window starts in 2024.
• Issuer concentration. As noted in 03-diversified-family-paper.md, the equity legs across Diversified route through Dinari for single stocks and Backed Finance for ETFs, settling on Base via Uniswap v3. Operational risk at the issuer level is a real residual that no backtest captures.

10. References

• Per-index Sharpe and vsBTCSharpe tables: methodology brief on request
• Engine: lib/backtest.js
• Cross-reference: 02-current-state-analysis.md for full library breakdown
• Cross-reference: 05-sharpe-decomposition.md for the driver analysis on the top names
• Cross-reference: 03-diversified-family-paper.md for the Diversified construction