Capridgan continuously reads market and on-chain data, then recalibrates portfolio exposure as your stated risk tolerance and the surrounding conditions change. The model adjusts; the strategy remains yours to approve.
Underlying inputs: order-book depth, on-chain settlement flow, volatility clustering, and macro liquidity signals — synthesised into a single risk-adjusted allocation view.
Crypto markets generate a volume of data that exceeds what manual analysis can usefully process: price ticks, order-book changes, on-chain transfers, derivatives positioning, and sentiment proxies arrive continuously and at variable reliability.
Capridgan treats most of this as noise by design. The system is built to isolate the subset of signals that have demonstrated predictive relevance to risk-adjusted return, while discarding transient price action that does not alter the underlying risk picture.
The objective is not to react to every fluctuation. It is to distinguish a genuine shift in market structure from short-term volatility, so that portfolio adjustments are made for the right reasons.
Three stages run continuously. None of them commits capital without a defined authorisation boundary set by the investor.
Market, liquidity, and on-chain data are ingested continuously and normalised against historical baselines, so the model always evaluates current conditions against established context rather than isolated data points.
The model compares current market risk against the investor's declared tolerance and existing allocation. Deviations are flagged before they are acted upon; the investor's thresholds, not the model's preferences, define acceptable boundaries.
Where calibration indicates adjustment is warranted, rebalancing is executed within parameters set in advance by the investor. The architecture of the strategy — what is permitted, and to what degree — remains under the investor's control throughout.
Rather than present outcomes, this section describes the reasoning and testing discipline behind them.
The system combines predictive time-series modelling with Bayesian inference to update risk estimates as new data arrives, rather than treating each forecast as a static, one-off calculation. This allows confidence in a given signal to be revised as evidence accumulates, instead of discarded or over-trusted outright.
Strategies are stress-tested against distinct historical market cycles, including periods of sharp contraction and low liquidity, rather than a single favourable sample window. A model that performs only in benign conditions is treated as unproven, regardless of its average return.
Client data and account permissions are handled under access-control and encryption practices consistent with UK financial services expectations. Execution permissions are scoped and auditable, and are never granted beyond what a given strategy requires.
The same underlying model supports materially different risk postures, because the parameters — not the engine — define the strategy.
For mandates prioritising drawdown control, the model weights liquidity and correlation stability more heavily than short-term return potential. Rebalancing frequency is deliberately constrained, and allocation shifts require a higher confidence threshold before execution is proposed.
This approach suits investors and asset managers for whom consistency of risk-adjusted return matters more than capturing every upward move.
Where the mandate tolerates greater variance in pursuit of return, the model's confidence thresholds are set lower and rebalancing can respond more frequently to detected shifts in market structure. The risk ceiling is still defined by the investor, not the model.
This configuration is intended for investors who have already accepted higher volatility as a condition of participation, and who want the strategy's responsiveness to match that mandate.
A consultation with Capridgan begins with a review of your current risk tolerance and allocation, not a sales pitch. From there, we outline how the calibration parameters would apply to your portfolio.
The value of crypto-assets can fall as well as rise, and you may get back less than you invest. Past performance and backtested results are not a reliable indicator of future performance. Capridgan does not provide personal financial advice; this material is provided for informational purposes and should be assessed in light of your own circumstances. Crypto-assets are largely unregulated in the UK and offer limited recourse if things go wrong.