Core concepts
Raxx enforces the structure you defined using your own rules and your own data. This page explains the four concepts that underpin how the platform works.
Rule-based strategy definition
A strategy in Raxx is a rule set you write. It has three required parts:
- Entry conditions — the criteria that must be met before a position is opened (price levels, indicator values, time-of-day restrictions, or any combination you specify).
- Credit thresholds — the minimum credit or premium your structure requires. If a candidate trade does not meet the threshold you defined, Raxx does not act.
- Exit conditions — the rules that close a position: a profit target, a stop level, an expiry horizon, or a combination. Raxx enforces the exit you defined regardless of how the trade looks in the moment.
You write the rules. Raxx enforces them. The platform has no strategy of its own — it has no opinion about what you should trade, when you should trade it, or whether your rules are well-designed. It executes what you set.
Strategies are stored as named structures in your account. You can hold multiple strategies and activate the ones relevant to your current approach.
Structure enforcement
The central problem Raxx addresses is not a skill gap — it is a structure gap. Most traders know their own rules; the difficulty is executing those rules consistently when emotion, fatigue, or opportunity-cost anxiety intervenes.
Raxx removes that intervention point. When your entry conditions are met, Raxx acts. When they are not, it does not. There is no manual confirmation step that lets emotion override the rule.
The enforcement model has three gates:
- Entry gate — checks every entry condition in your strategy before a position is opened. All conditions must pass. A single failing condition blocks the entry.
- Credit gate — validates that the credit or premium at the moment of evaluation meets your defined threshold. If the market has moved such that the credit is no longer sufficient, the entry is skipped.
- Exit gate — monitors open positions against your exit conditions and acts when the conditions are met. The exit fires on your rules, not on an in-the-moment judgment call.
These three gates run in order, deterministically, every time a candidate trade is evaluated. No gate is optional.
AI augmentation
Raxx includes AI-assisted features in its reporting and analysis layer. These features are retrospective only — they describe what happened on your own historical data. They do not predict, recommend, or propose trades.
What AI surfaces in Raxx:
- Pattern recognition on your history — the AI reviews your past trades and surfaces recurring conditions: which entry setups appeared in your winning and losing trades, which credit levels your strategy captured, where your exits fired early or late relative to your defined structure.
- Historical markers — the AI annotates your trade history with relevant historical events (earnings announcements, stock splits, macro data releases) that occurred during open positions. These annotations describe context for what already happened.
- Retrospective descriptions — AI-generated summaries describe the behavior of your strategy across a historical date range. Every sentence is past-tense and sourced from your own data.
AI in Raxx does not propose trades, score your strategy against others, or describe what your strategy might do in the future. It does not generate recommendations from external models or third-party signals.
The AI layer exists to help you understand what your own strategy did. Understanding is retrospective. Execution is yours.
Deterministic execution model
The order-firing path in Raxx is rule-based and operates the same way every time it runs. When your strategy conditions are evaluated, the outcome is determined by your rules alone — not by probabilistic models, external signals, or AI inference.
- Reproducibility — given the same market data and the same rule set, Raxx produces the same decision. You can backtest against historical data and trust that the live execution path applies the same logic.
- Auditability — every execution decision is logged with the input conditions and the rule that triggered or blocked the action. You can review exactly why Raxx acted or did not act on any candidate trade.
- No autonomous AI action — the AI layer observes and describes; it does not write to the execution path. A pattern surfaced by AI retrospective analysis does not automatically modify your strategy or queue an order.
The separation between the AI layer and the execution layer is architectural, not just a policy. The execution path evaluates your rules. The AI path reads completed history. They do not intersect.