What is Expected Value?

In trading, profit alone can be misleading.

Expected Value (EV) describes how much a trading model earns per trade on average, taking both wins and losses into account. It answers a simple but powerful question: if this model keeps trading under similar conditions, what should we expect it to produce over time?

A model can make money for a while — and still be structurally fragile. Expected Value looks at the statistical foundation underneath the results, not just the equity curve.

Two models can show identical total profit — but one might rely on a few lucky outliers, while the other generates steady, repeatable outcomes. Expected Value helps expose that difference.

Why Expected Value matters

When models compete, the goal is not just to find the one that made the most money last month. The goal is to find models whose edge is structurally sound.

A model with high win rate but rare catastrophic losses, one with low win rate but oversized winners, or one with wildly inconsistent trade distribution — all can show temporary profitability.

Expected Value cuts through this noise. It balances win probability, average win size, and average loss size into a single expectation per trade.

If that expectation is positive and stable, the model has a mathematical edge. If it fluctuates wildly, the edge may be fragile.

Expected Value in adaptive systems

Markets are not stationary. Edges decay. Regimes shift.

In adaptive environments like darwintIQ, models continuously compete under recent market conditions. What worked six months ago may not work today.

That's why Expected Value is evaluated over rolling windows, not static backtests. The question isn't whether the model was profitable in the past — it's whether it currently shows a stable statistical edge.

Stability matters more than peak return.

How darwintIQ uses Expected Value

darwintIQ ranks models by overall fitness — a composite evaluation of performance and robustness.

Expected Value contributes as one core component of that evaluation. It is never viewed in isolation. Instead, it is considered together with profitability, drawdown behavior, distribution consistency, and stability across recent data.

A model with a smooth, consistent Expected Value under current market conditions tends to rank more reliably than one driven by unstable outliers.

This allows darwintIQ to distinguish durable adaptive behaviors from short-lived statistical accidents.

Final thoughts

Expected Value is not about optimism. It's about probabilistic reality — and in non-stationary markets, probabilistic reality is everything.

It's the difference between a model that got lucky and a model that has an edge. That distinction is what systematic trading is built on.