Crypto index comparison, tier 3
CRIX Alternative: CCi30 vs CRIX
The CCi30 Cryptocurrency Index is the rules-based alternative to the CRIX. This page reviews the CRIX, The CRyptocurrency IndeX under the eight-criterion CCi30 test and compares both indices on universe, weighting, independence, track record, and investability.
What is the alternative to the CRIX?
The CCi30 Cryptocurrency Index replaces the CRIX for investors who need a whole-market benchmark. The CCi30 holds the 30 largest cryptocurrencies by smoothed market capitalization, weights them by the square root of that figure, excludes stablecoins by rule, and has published live values since 1 January 2015.
- 30 constituents
- Square-root weighting
- Stablecoins excluded by rule
- Live since 1 January 2015
- Independent, fully rules-based
What is the CRIX?
CRIX, developed at Humboldt University Berlin (Härdle, Trimborn et al.) and later commercialized with Royalton Partners, is the academic community’s flagship crypto index: a market-cap-weighted composite whose distinguishing feature is a dynamic constituent count, selected by statistical information criteria (AIC-family model selection) so that the number of members expands and contracts with market structure.
How is the CRIX built?
At each reconstitution, CRIX chooses how many constituents to include by testing which index size best approximates the total market according to an information criterion, then cap-weights those members. The design is published in the econometric literature and is, by academic standards, exemplary in transparency.
Where the CRIX falls short statistically
The most intellectually serious competitor, and a revealing contrast
CRIX and the CCi30 were born in the same intellectual moment (2014–2015), both from academic mathematics, both asking the right question: how does one represent an entire novel asset class with a tractable portfolio? They diverged on a deep philosophical point, and the divergence teaches something.
CRIX optimizes statistical fit to the total market, accepting whatever constituent count and turnover that requires. The CCi30 optimizes statistical fit subject to investability: a fixed count of 30, derived as the point where both constraints bind, the minimum for statistical significance (the top 30 represent the whole market at 99% confidence with a 1.11% margin of error, capturing ~90% of true-crypto capitalization) and the maximum before the liquidity cliff, beyond which added constituents generate fees without performance. Square-root weighting maximizes effective diversification within that portfolio; EWMA smoothing acknowledges that real rebalancing happens in real markets vulnerable to real manipulation. CRIX’s information-criterion machinery is elegant, but it re-derives at every reconstitution what the CCi30 solved once, in closed form, in public, and a fixed answer is what replication requires.
The consequences of CRIX’s choice: a fluctuating membership count that makes replication a moving target; cap-weighting that leaves the “optimal fit” dominated by Bitcoin anyway (fitting a BTC-dominated total market with a BTC-dominated portfolio is not difficult, merely uninformative); and, because the total market it fits includes whatever the data vendor counts, the pegged-asset contamination problem enters through the target itself. An index calibrated to approximate a total-market series that contains stablecoins is calibrated, in part, to approximate fiat.
Where CRIX deserves respect:
it includes assets by data, not by custody committee, privacy coins have appeared in its membership as their capitalization warranted. On the censorship criterion it stands with the CCi30 against the entire commercial field.
Can the CRIX be replicated by an investor?
This is where the academic design fails outright. Variable constituent counts, cap-weight turnover, and indifference to execution cost make CRIX a superb object of study and a poor object of investment. No significant investable product tracks it, the market’s own verdict on theoretical elegance without implementation discipline.
Method and sources
Methodology facts on this page come from the published documents of the provider; constituent lists change and should be re-verified before citation. The CCi30 rules are published in the methodology manual. The full comparison set is on the crypto index comparison hub, and the allocation calculator shows the CCi30 basket for any amount.