Methodology

How the Forecast Works

The forecasts on this site are generated using machine learning models trained on historical electricity-market, power-system, and weather data. The forecast extends up to 14 days ahead and is refreshed several times each day.

Two-stage architecture

First, dedicated models forecast wind generation, solar generation, and electricity demand. These are gradient-boosted decision tree models, fitted separately for each area. Their outputs are then passed to a downstream price model together with market and power-system features.

One shared neural network model uses these forecasts and power-system data to predict hourly prices for all supported areas together. This allows conditions in one area to influence price forecasts in connected markets.

Training and weather data

Historical training data currently extends back to 2023. Weather inputs use ECMWF forecasts delivered through Open-Meteo, including archived forecasts for model training, together with market data from SMARD, Elering, ENTSO-E and other publicly available sources.

Model Inputs

The downstream price model combines several types of information:

• Forecast fundamentals: electricity demand, wind generation, and solar generation.
• Power-system conditions: each area's own nuclear availability, hydrological conditions, and forecast cross-zonal transfer capacity.
• Temporal and market context: hour, weekday, season, public holidays, direct weather summaries, and recent price patterns.

What the Model Learns

The model identifies patterns that have historically been associated with electricity prices. For example, higher renewable generation generally puts downward pressure on prices, while stronger demand and tighter supply conditions tend to support prices. Weather can also affect prices indirectly through demand and generation availability.

Rather than explicitly simulating the physical operation of the power system, the model learns these statistical relationships from historical data.

Forecast Uncertainty

The central forecast is a point estimate rather than a guarantee. Where available, the Weather Ensemble page runs the same trained models across alternative ECMWF weather scenarios. This illustrates uncertainty arising from the weather forecast, but it does not include uncertainty from outages, market behavior, fuel prices, data errors, model error, or other unmodeled developments.

Geographic Coverage

Forecasts are currently available for:

• Austria (AT)
• Belgium (BE)
• Czechia (CZ)
• Denmark (DK1, DK2)
• Estonia (EE)
• Finland (FI)
• France (FR)
• Germany (DE)
• Latvia (LV)
• Lithuania (LT)
• Netherlands (NL)
• Norway (NO1-NO5)
• Poland (PL)
• Portugal (PT)
• Slovakia (SK)
• Spain (ES)
• Sweden (SE1-SE4)

Limitations

Statistical rather than physical modeling

The approach is statistical rather than a full market simulation. It does not explicitly optimize generation dispatch, reservoir operation, transmission flows, bids, or unit commitment. Instead, it learns relationships present in historical market and power-system data. Factors that are unavailable or not explicitly modeled, including fuel and emissions prices, outages not known when the forecast is issued, and participant behavior, can therefore cause actual prices to differ materially from the forecast.

Transmission

Cross-zonal transmission is represented using forecast transfer capacities published by ENTSO-E. This captures changes such as planned maintenance, but it does not reproduce the full market-coupling and network-constraint calculation used in actual market clearing. Congestion and cross-border effects are therefore approximated rather than physically simulated.

Hydro treatment

Hydro generation is not currently an operative price-model input or a forecast.

Hydrological state variables such as reservoir levels, river flow, snow, and precipitation remain separate price-model inputs where configured. The approach does not simulate reservoir optimization or hourly hydro dispatch, so it cannot anticipate changes in operating strategy or genuine shifts in hydro generation.

Unfamiliar and extreme conditions

Both model families learn from the historical record rather than from physical first principles, so both are strongest when future conditions resemble combinations already represented in their training data. The gradient-boosted trees that forecast the fundamentals cannot produce a value outside the range they were trained on at all. The neural price model can, but a price far beyond anything in the record is an extrapolation rather than a learned relationship. Extreme price spikes, negative-price events, and other unusual market situations can therefore be difficult to predict, particularly when similar events are rare in the historical record.

Structural market change

The electricity system changes over time as generation capacity, demand, transmission, and market rules evolve. Models are retrained and inputs such as installed capacity are updated, but historical relationships may still become less representative as the system changes.

Data Sources

The live forecasting system currently uses the following public and third-party data sources. The forecasts, derived features, and visualizations are calculated by this site. Source attribution and market-data reuse conditions are described on the Data Licences page.

• Bundesnetzagentur | SMARD.de provides day-ahead prices for DE, AT, BE, CZ, DK1, DK2, FR, NL, NO2, PL and SE4, plus German and Austrian wind and solar generation, demand and installed wind capacity. German fundamentals cover the Germany/Luxembourg bidding zone; Austrian fundamentals cover Austria. These data are licensed under CC BY 4.0. Hourly aggregation, derived features and forecasts are this site's calculations.
• Elering AS provides day-ahead prices for FI, EE, LV and LT. These are Nord Pool prices redistributed by the Estonian transmission system operator, retrieved in the market's native intervals; hourly aggregation and forecasts are this site's calculations.
• ENTSO-E Transparency Platform provides electricity prices for the remaining areas, plus generation, demand and installed capacity outside the SMARD markets, nuclear outages and forecast transfer capacity for every area.
• Open-Meteo ECMWF API provides operative wind, temperature, solar, and precipitation forecasts. Its Historical Forecast API provides model history. Its Ensemble API provides weather scenarios.
• Fingrid Open Data provides Finnish wind actuals, forecasts, and capacity, plus near-term demand forecasts.
• SYKE Hydrologiarajapinta provides Finnish precipitation and snow-water conditions.
• NVE reservoir statistics provides Norwegian reservoir fill levels and stored energy.
• SMHI Vattenwebb provides Swedish modeled river-flow conditions.
• Nager.Holidays provides national public-holiday calendars.