NEXUM isn't a static strategy — it's an AI-adaptive system that continuously learns from the market conditions it observes and trades in. That's what makes it powerful, but it also means the quality of what it learns matters enormously.
The market behaves very differently off-hours. During overnight sessions and pre/post-market hours, the NQ futures market is characterized by thin liquidity, wider spreads, erratic price swings, and movements driven by news flow and a fraction of the normal participant base. There's no institutional order flow, no algorithmic market-making at scale, and no meaningful price discovery happening in the same way it does during Regular Trading Hours (RTH).
When NEXUM sim trades in that environment, it's studying the wrong classroom. Its AI components are adapting to — and learning patterns from — a market that simply doesn't exist when you go live. The volatility profiles are different, the momentum characteristics are different, and the mean-reversion tendencies are different. NEXUM will tune itself to conditions it will rarely, if ever, encounter once you switch it on during RTH.
The practical risk is this: By the time you go live during regular hours, NEXUM's adaptive parameters may be calibrated for ghost-town market behavior. That misalignment can cause it to misread momentum, misjudge volatility, and apply risk parameters that don't reflect what the live RTH market is actually doing — potentially right from your very first trade.
The right way to sim trade NEXUM is during the same session you intend to trade it live. Run it during RTH, let it observe and adapt to institutional order flow, real liquidity, and the market dynamics it will actually encounter. That way, when you flip to live, the strategy's learned state is a genuine reflection of the conditions ahead of it — not a distorted picture built from the overnight shadows.
Think of it this way: you wouldn't train for a marathon by running on sand if the race is on pavement. The effort is real, but the adaptation is wrong.