Qual é a latência média entre a deteção de uma variação de mercado e o ajuste do stop-loss?

A latência depende da fonte de dados ligada e da carga do mercado no momento, mas o processo de recálculo interno foi desenhado para operar em escala de milissegundos a partir da receção de novos dados. Em períodos de alta volatilidade extrema, a prioridade do sistema é a consistência do ajuste, mesmo que isso implique um ligeiro aumento de latência.

O Amphojel integra-se diretamente com a minha corretora?

A integração depende da corretora utilizada e da disponibilidade de uma API de execução de ordens compatível. Nos casos em que a ligação direta não é possível, o sistema pode funcionar em modo de recomendação, indicando o novo nível de stop para confirmação manual.

Os ajustes de stop-loss são testados antes de serem aplicados em conta real?

Sim. Cada alteração à lógica de ajuste passa por backtesting contra períodos históricos de diferentes regimes de volatilidade antes de estar disponível para utilização em contas reais. Resultados de backtesting são indicativos e não constituem garantia de desempenho futuro.

Posso definir limites máximos que o sistema nunca pode ultrapassar?

Sim. Os limites de risco por operação e por conta são definidos pelo utilizador antes de qualquer automatismo ser ativado, e o modelo não está autorizado a propor ou aplicar ajustes fora desses limites.

O que acontece se a ligação de dados falhar durante uma posição aberta?

Em caso de falha de ligação, o sistema mantém o último nível de stop-loss validado e emite uma notificação, em vez de remover a proteção ou de tentar estimar valores sem dados atualizados.

Data-driven risk management

Dynamic stop-loss supported by AI predictive models

Amphojel analyzes volatility, liquidity and historical patterns in real time to recalculate your stop-loss levels as market conditions change, reducing exposure to drawdowns without replacing your entry strategy.

Explore the Demo See the Methodology

Volatility Monitored at short intervals, per asset

Correlation Cross between assets and benchmarks

Automatic adjustment From the stop level, within the limits defined by you

Results depend on the risk profile, the asset traded and market conditions in each session.

The problem

Intraday volatility tests the discipline of even the most experienced traders

A fixed stop-loss, set at the start of the session, rarely reflects how volatility behaves throughout the day. In moments of greater uncertainty, the same stop level may be too tight and generate premature exits, or too wide and allow losses greater than planned.

The decision to manually adjust a stop-loss in real time is also an emotional decision: it involves pressure, limited time and, often, confirmation bias. Amphojel replaces this manual adjustment with a continuous analytical process, based on market data and not on momentary reaction.

Amphojel data analysis team reviewing market indicators

Technology

How the predictive model decides when to move the stop-loss

The system combines three layers of analysis — volatility, liquidity and correlation between assets — to continuously recalculate the ideal distance between the current price and the stop level, always respecting the risk limits defined by the user.

Volatility Reading

Measures price amplitude over short windows to distinguish normal noise from relevant directional movement.

Depth of liquidity

Considers the volume available in the order book before tightening or widening the stop, avoiding executions in thin market conditions.

Correlation between assets

Compares the asset's behavior with related reference indices, signaling isolated versus systemic movements.

User-defined limits

Each automatic adjustment respects the risk tolerance and maximum limits previously configured, without exceeding these parameters.

The model does not predict future price with certainty — it reduces the likelihood of a static stop-loss becoming misaligned with actual market conditions by recalculating your position with each new set of data received.

Methodology

From raw data to adjustment decision: the four-step process

Transparency about how data is handled is a central part of how Amphojel is built and validated.

Market data collection

Prices, volumes and order book depth are received from data sources linked to your brokerage account, without changing the original values.

Normalization and pattern detection

Data is normalized by asset and compared with historical series to identify relevant deviations from usual behavior.

Continuous backtesting

Each rule adjustment is tested against historical periods of high and low volatility before being made available in a real environment.

Execution within defined limits

The new stop-loss level is only applied if it remains within the risk tolerance and capital rules you set up.

Note on data sources: The market data used comes directly from your broker connection or from market data providers that Amphojel integrates with. Data from unidentified third parties or external predictions without traceability are not used.

Practical application

How Different Trading Styles Utilize Dynamic Stop-Loss

The model's behavior adapts to the time horizon and risk tolerance of each trading style.

Scalping

Short interval adjustments for minute operations

In operations of very short duration, the model prioritizes reading immediate volatility and depth of liquidity, recalculating the stop several times within the same position to follow fast movements without excessively increasing the risk per operation.

Swing trading

Stops that track multi-day trends

For positions held over several days, the model gives more weight to the correlation between assets and medium-term trends, gradually widening the stop on consistent movements and tightening it when signs of reversal appear in correlated assets.

Algorithmic Trading

Integration as a risk layer on top of existing strategies

Traders who already operate with their own automated systems can use Amphojel only as a stop-loss management layer, receiving the levels recalculated via API and keeping the entry and exit logic of their own strategy unchanged.

FAQ

Technical questions about latency, integration and data

What is the average latency between detecting a market change and adjusting the stop-loss?

Latency depends on the connected data source and the market load at the time, but the internal recalculation process is designed to operate on a millisecond scale upon receipt of new data. In periods of extreme high volatility, the system's priority is tuning consistency, even if this implies a slight increase in latency.

Does Amphojel integrate directly with my broker?

Integration depends on the broker used and the availability of a compatible order execution API. In cases where direct connection is not possible, the system can work in recommendation mode, indicating the new stop level for manual confirmation.

Are stop-loss adjustments tested before being applied to a real account?

Yes. Each change to the adjustment logic is backtested against historical periods of different volatility regimes before being available for use in real accounts. Backtesting results are indicative and do not constitute a guarantee of future performance.

Can I set maximum limits that the system can never exceed?

Yes. The risk limits per operation and per account are defined by the user before any automation is activated, and the model is not authorized to propose or apply adjustments outside these limits.

What happens if the data connection fails during an open position?

In case of connection failure, the system maintains the last validated stop-loss level and issues a notification, instead of removing protection or trying to estimate values without updated data.

Do you have other technical questions about the platform? Consult the support center for additional details on integration and configuration.