Wide day-ahead spread in France today: 19 €/MWh at 15h, 241 €/MWh at 19h. That is 222 €/MWh from trough to peak (241 - 19). Did not expect French evening prices this high. Wind was weak and gas output on Friday was 5.4x the previous Friday’s, per RTE data cited by Selectra. Whether French gas or neighbours set the peak, this data does not say. https://selectra.info/energie/actualites/prix-electricite/2026-10-17
Capture rate is the ratio between generation-weighted and time-weighted average prices. Usually you compute the capture rate over a specified amount of time, e.g. a certain week, month, or year. You start with the generation-weighted average price your generator captured on the market: Capture price = sum(generation(t) x market price(t)) / sum(generation(t)) This is the average market price weighted by your generator’s output at the time of generation. For instance, your PV-weighted capture price was 56 Euro / MWh over the past year. ...
There’s a direct link between gas prices and wholesale electricity prices. The marginal cost of thermal generators (CCGT, CHP) is directly linked to how much they pay for their fuel. So when wholesale gas prices go up, the same generator will increase their electricity auction offer price. And when these gas-fired generators are marginal in the merit order stack the market clearing price goes up for everyone. Holding all else constant, how does a 1 Euro/MWh gas price increase affect the average daily electricity price? And how does this differ between the European bidding zones? ...
Depending on the asset you optimize and the market you act in, there are at least two different ways to model the effect your orders have on the market: As a price taker, you don’t accept any price - you can still submit limit prices. But you assume that your order does not affect the clearing price. The prices you deal with and optimize against are assumed to be external parameters. ...
Looked through Eurelectric’s 10 AI use cases for markets & trading. What stood out: very little of this is “AI trading” in the sense of a model directly deciding orders. The recurring setup is much more familiar: ML / DL for forecasts → optimisation for constrained decisions → trader or control system. The most interesting examples for me are ESB’s BESS dispatch optimiser, Fortum’s flexibility portfolio optimisation across competing asset uses, and EDP’s seasonal forecasting feeding hydro and trading decisions. ...
R² clicked better for me once I thought about the baseline predictor. Take the WWS vs electricity price example. Before looking at WWS share at all, the simplest prediction for every US state is just the average electricity price across all states. Now fit a line using each state’s wind, water and solar share to predict its electricity price. R² measures how much that fitted line reduces the squared prediction error compared with just predicting the average for every state. ...