Dogmargin converts historical market data into risk-adjusted entry points, giving UK students a disciplined way to study crypto markets before committing capital.
Crypto price action is dense with short-term signals that carry little statistical weight. New investors often mistake this noise for meaningful trend, then act on emotional bias rather than evidence — buying into rallies late and exiting during drawdowns out of fear.
Dogmargin was built around a simple constraint: no position is recommended unless the underlying pattern has been tested against multiple historical market cycles. We replace the instinct to react with a model that has already seen the scenario before, or something statistically close to it.
Dogmargin is built for students who want to understand crypto markets analytically rather than gamble on them. Our models are trained on UK and global historical cycles, then validated through a staged process before any strategy reaches a live account.
We publish our methodology rather than testimonials, because a transparent process is more useful to a student audience than a success story with no verifiable data behind it.
The platform ingests price, volume, and on-chain data continuously, then runs it through models trained to detect statistically significant shifts in volatility and correlation. When a shift is detected, exposure is adjusted automatically rather than left to discretion.
This is not a signal service. It is a layered system where each recommendation carries a documented probability of success, derived from how similar conditions resolved historically.
Rather than asking for trust, we show the stages a strategy must pass before it is deployed. Every model is trained and tested against UK and global market cycles going back through multiple bull and bear phases, not a single favourable period.
Price, volume, and liquidity data are pulled from multiple exchanges and reconciled for consistency.
Raw data is converted into statistically relevant variables, such as volatility clusters and correlation shifts.
Strategies are run against historical cycles to measure drawdown, win rate, and risk-adjusted return.
Validated models trade on live market data without real capital, confirming behaviour outside historical conditions.
Only models that pass all prior stages become available for use with real, fractional positions.
Positions can be sized in small, fractional amounts, so a strategy can be studied and tested without committing a meaningful share of a student budget. Risk parameters are visible before any position is opened, not after.
Every recommendation is accompanied by the reasoning behind it — the correlation, volatility measure, or historical precedent that informed it. This is designed to build financial literacy, not to replace the user's own judgement.
Monitoring volatility manually around a lecture timetable is impractical. The system tracks conditions continuously and adjusts exposure within predefined limits, reducing the need for constant manual attention.
Market data is sourced from established exchange APIs and aggregated price feeds covering major UK and global trading venues. On-chain data is drawn from public blockchain records. All data is reconciled before being used in model training.
No backtested model can predict an unprecedented event with certainty. Instead, the system uses drawdown thresholds that automatically reduce exposure once volatility exceeds historical norms, limiting the scale of loss during abnormal conditions rather than attempting to forecast them.
Crypto-asset activity in the UK currently sits outside Financial Services Compensation Scheme protection. We make this clear to every user before onboarding, and we do not describe any position as risk-free.
Account access requires two-factor authentication, and sensitive data is encrypted both in transit and at rest. We do not hold user funds directly; execution occurs through connected exchange accounts under the user's own control.
Yes. Strategies can be paused or closed from the account dashboard at any time, and fractional positions can be unwound without needing to exit an entire portfolio at once.