SigmaAxion
Quantitative · Systematic · Disciplined

Capital where the probabilities are right.

SigmaAxion is a quantitative trading platform for systematic event and equity strategies. Research, execution, and risk in one operating surface — built for desks that measure everything.

NAV 1,000.00 USD Sharpe 30D 1.84 Max DD -4.2% Hit rate 58.3% Avg edge +3.1% Latency p50 62ms Uptime 99.97% Venues Kalshi · Alpaca Strategies 14 live NAV 1,000.00 USD Sharpe 30D 1.84 Max DD -4.2% Hit rate 58.3% Avg edge +3.1% Latency p50 62ms Uptime 99.97%
Decision latency
< 100 ms
Signal to order, round trip
Daily reconciliation
100%
Position, cash, P&L
Risk layers
4
Hard · Soft · Strategy · Operator
Auditability
Every event
Append-only log

Systematic edge, repeatable process.

01 · DATA

High-fidelity market data

Normalized feeds from regulated exchanges and prediction venues, time-stamped at ingest, with microstructure snapshots preserved for replay and backtest parity.

02 · MODELS

Mean-reverting, event-driven, relative-value

A portfolio of uncorrelated strategies — mispricing against a consensus, cross-venue arbitrage, and calibrated event pricing — each subject to its own capacity and cost envelope.

03 · EXECUTION

Low-impact, low-variance

Post-only makers, liquidity-weighted sizing, Kelly-fractional capital allocation, and strict pre-trade checks. No order fires without a live risk greenlight.

Operating principles.

Process > prediction
We measure decisions, not just outcomes.
Every opportunity, fill, and hedge is logged with its pre-trade assumptions and its post-trade reality, so the process keeps improving.
Survival first
Quarter-Kelly sizing, hard daily stops.
Geometric growth beats explosive growth. We cap drawdown before we cap upside, and we don't override the risk engine mid-session.
Transparency
Every cap, every breach, every override.
An append-only audit log records every state-changing action with actor, IP, and reason. Risk limits are tunable; history is not.
Small edges, honest math
We trade the arithmetic, not the story.
Fees, slippage, and fill probability are modeled before the first dollar moves. If the expectation is negative after costs, the order never prints.