Adaptive Risk, Explained — A Plain-Language Companion
Companion post. Plain-language summary of the Adaptive Risk Engine whitepaper. For the full methodology and limitations, read the whitepaper at /whitepapers/adaptive-risk-engine.
What "Adaptive Risk" actually means
Most retail risk tools are static: a fixed stop-loss percentage, a fixed position cap, a fixed margin buffer. Treeova's Adaptive Risk Engine is deterministic math the agent pulls levers on — never arbitrary arithmetic done by a language model. The agent decides when to tighten or loosen a lever; the math itself is fixed code, audited, and version-controlled.
Two tiers, one principle
- Standard tier — sensible defaults: trailing stops, hard floors, penny-option guards, and basic position caps.
- Adaptive tier — the standard tier plus regime-aware adjustments: tighter stops in high-volatility regimes, looser stops in established trends, stricter caps when correlations spike.
Both tiers share the same core principle: the agent never invents a number. It selects from a pre-approved set of throttle levels, and the math runs deterministically from there.
The trailing stop state machine
The trailing stop is implemented as an explicit state machine — not a floating heuristic. It moves through defined states ("armed", "ratcheted", "triggered") with audited transitions. The whitepaper walks through the states and the transition rules.
Safety floors
A safety floor prevents the engine from sizing into positions where the worst-case loss exceeds an account-level threshold. The exact threshold is account-tier-specific and intentionally not published — but the existence and behavior of the floor are documented in the whitepaper.
Audit log
Every risk decision is written to an append-only audit log: which lever, which throttle level, what regime, and what the position state was at the time. This makes the system reviewable — critical for both compliance and debugging.
Limitations
- Adaptive Risk reduces avoidable loss; it does not eliminate market loss.
- Penny-option guards and safety floors are best-effort under fast market conditions.
- Throttle thresholds are conservative on purpose.
Read the methodology
Adaptive Risk Engine: Two-Tier Protection Model →
Past performance does not guarantee future results. See /legal/risk-disclosure.