Risk Controls
KEK reduces trading risk through structured validation, simulation, and monitoring — not through custody or forced execution controls.
Risk controls in KEK are designed to:
- Identify unsafe strategies before execution eligibility
- Constrain behavior and surface failure modes during validation
- Make risk visible and measurable before capital is deployed
This page explains where risk controls apply, how constraints are used during validation, and where responsibility remains with the user.
What this page covers
- Where risk controls exist in the system
- How constraints are applied during validation
- How users define risk boundaries
- What KEK does — and does not — control
Where risk controls apply
Risk controls in KEK primarily apply during:
- Backtesting (historical validation)
- Optimization (robustness-first parameter testing)
- Paper trading (live simulation without real capital)
- Monitoring & refinement (drift detection and continuous evaluation)
During live execution, KEK does not enforce risk limits on-chain or force trade outcomes. Execution remains user-authorized, and responsibility remains with the user.
Validation-level controls
KEK applies structured constraints during validation to expose risk behavior and failure modes before a strategy can be considered for optional deployment.
Position sizing constraints
Strategies are evaluated against configurable position limits such as:
- Maximum position size relative to a portfolio
- Concentration caps across assets
- Exposure ceilings per strategy or regime
These constraints help identify strategies that rely on excessive concentration or unstable sizing behavior.
Exposure analysis
KEK evaluates total exposure to detect hidden portfolio-level fragility, including:
- Aggregate capital deployment
- Correlation and clustering risk (multiple positions behaving as one)
- Sensitivity to adverse moves under certain regimes
Strategies exceeding defined thresholds are flagged during validation and may be rejected or forced into refinement.
Leverage modeling
Leverage is modeled during backtesting and paper trading to observe:
- Liquidation / margin sensitivity
- Drawdown amplification
- Volatility dependence and convexity risk
Default assumptions are conservative and configurable to prevent overly optimistic validation.
Drawdown and stress controls
Strategies are evaluated under adverse and extreme scenarios, including:
- Sudden volatility spikes
- Liquidity drops / widening spreads
- Adverse regime transitions
Stress testing is a standard risk-management technique used to evaluate stability under adverse conditions.
Drawdown behavior is treated as a primary risk signal because it reflects how strategies fail, recover, and persist under pressure.
Monitoring and alerting
Once strategies enter paper trading (and later, post-deployment observation):
- Performance metrics are continuously tracked
- Deviations from expected behavior are detected
- Drift signals can trigger structured refinement cycles
Monitoring informs users and refinement systems — it does not autonomously enforce execution changes.
User-defined risk parameters
Users may define preferences that influence:
- Strategy selection and eligibility
- Validation thresholds and pass/fail gates
- Monitoring sensitivity and drift triggers
Examples include:
- Preferred position sizing limits
- Conservative leverage assumptions
- Maximum drawdown tolerance
- Rebalance sensitivity / churn tolerance
- Exposure caps per asset or portfolio segment
These parameters guide validation and monitoring behavior so strategy evaluation aligns with user intent.
What KEK does not do
KEK does not:
- Force liquidation
- Override user-approved execution decisions
- Halt trades autonomously
- Guarantee loss prevention
- Eliminate market volatility or unexpected events
Risk controls reduce exposure to unsafe strategies — they do not remove risk.
Simulations and backtests are valuable, but they are not equivalent to live trading realities and cannot guarantee outcomes.
Safety-by-design system limits
KEK applies "least authority" design thinking across components so no system layer has unnecessary power.
The principle of least privilege means each system component should only have the minimum permissions required to perform its function, limiting the blast radius of failures or misuse.
This is a core reason KEK can provide validation and monitoring intelligence without becoming an execution authority.
Why this matters
By separating validation from execution, KEK:
- Reduces the chance of deploying unsafe strategies
- Makes risk observable before capital is involved
- Encourages systematic discipline over impulsive deployment
- Preserves user autonomy and non-custodial control
Risk is managed through structure and evidence — not enforcement.