Electricity prices reflect supply, demand, market rules, and network limitations. Spatial aggregation changes which price drivers a model can distinguish—and which risks it may conceal.
What is a nodal price?
In markets using locational marginal pricing (LMP), the price at a node reflects the marginal cost of serving an incremental unit of demand at that location under the market formulation. Energy, congestion, and losses may contribute depending on the market rules and model.
Two nearby nodes can exhibit materially different price behavior under binding network constraints. That matters for assets whose revenues or costs are tied to a specific settlement location.
What is a zonal price?
Zonal forecasts aggregate or represent prices over a broader region. They can be appropriate for strategic market views, regional hedging questions, or cases where granular network observations are unavailable. The downside is that aggregation can obscure local congestion, basis, and asset-specific exposure.
| Question | Nodal view | Zonal view |
|---|---|---|
| Location-specific congestion | Directly relevant where modeled | Often obscured by aggregation |
| Regional energy fundamentals | Useful but data-intensive | Often a natural starting point |
| Asset-level basis risk | Important | May require separate basis assumptions |
| Data and model complexity | Typically higher | Typically lower |
Why does basis matter?
Basis describes the relationship or difference between two relevant prices, such as an asset's nodal settlement price and a regional reference. A regional forecast can be broadly correct while a project earns materially different realized prices because local constraints change. The appropriate basis measure depends on contract and market design.
Forecasting risk, not just averages
For battery storage or merchant generation, a forecast distribution can matter more than its mean: revenue exposure may be concentrated in rare hours, constrained intervals, or scenario-dependent price spreads. A useful analysis asks not only “what is the expected price?” but also “which network conditions or market assumptions would make the investment case fail?”
How to assess quality
Compare forecasts to relevant baselines over genuinely out-of-sample periods; separate forecast vintages from revised historical data; and assess interval coverage and calibration when uncertainty is presented. A complex network-aware model should demonstrate why its added complexity improves the actual decision at hand.
