Method notes

Every figure on this site is computed from open data by a pipeline, and each kind of figure has one note here stating what goes in, how it is weighted, which window it covers and where it is known to fall short. A view links to the notes it uses. The How it works page has the overall picture; the Data sources page names where the data comes from.

Measured Someone counted or metered it: a register, a meter, a benchmark, a dated price list.

• The canonical hourly grid Every price series is averaged onto whole clock hours, so prices from before and after the quarter-hour switch compare evenly.
• The daily spread (TB1 and TB2) The daily spread (TB1) is a day's dearest hour minus its cheapest; TB2 averages that gap over the two dearest and two cheapest hours.
• Capture price and value factor The average price wind or solar power earned, weighted by how much each produced in each hour, set against the plain average price.
• Residual load Residual load is what is left of demand once wind and solar generation for the hour is taken out, the load other plants and batteries still have to cover.
• The Dunkelflaute spell detector A Dunkelflaute is a spell when wind and solar together nearly stop generating, found by testing how long their combined share of demand stays low.
• Day-ahead forecast error How far the day-ahead wind and solar forecast misses and which way it leans, nationally and per control area (the region one grid operator balances).
• Negative hours and episodes How many hours the day-ahead price fell below zero in a bidding zone, and how long each run lasted.
• The hour-of-day price profile The average price for each hour of the local day over the window, split into weekday and weekend because the demand shape differs between them.
• Monthly mean prices and the divergence The mean day-ahead price per local calendar month for each bidding zone, and how far DE-LU's monthly mean diverges from France's.
• The generation mix How much electricity each of the site's fuel groups generated in the zone over an average local hour of the day, in gigawatts.
• Cross-border flow and congestion How much electricity physically crossed each border each month, and how often the border's capacity limited that flow.
• Renewable share of generation The share of an hour's electricity generation that came from renewable fuels, as a percentage, tracked alongside carbon intensity.
• Each control area's share of wind and solar Each control area's share of Germany's wind and solar output, by day and by hour of the day. A control area is the region one grid operator balances.
• The battery fleet from the register The battery fleet the Bundesnetzagentur's (Germany's energy regulator) register lists, summed by commissioning month and its usable-energy segment.
• The wholesale-to-retail wedge The gap between the wholesale price and a household bill, broken into energy, network, and tax and levy components, in ct/kWh.
• Battery-electric share of registrations The share of newly registered vehicles that run on a battery alone, for the same geography, segment, period and source.
• Road transport's share of electricity use Road transport's share of electricity use is its electricity consumption as a percentage of all final electricity consumption, for the same year.
• The public charge-point count The public charge-point count reads the Bundesnetzagentur's public charge-point register, counting points and the stations that host them separately.

Estimate Someone computed it from assumptions about today.

• Production-based carbon intensity How much CO₂ each kilowatt-hour of electricity carries, weighted by which fuels generated it in that hour and counted where the power is produced.
• The marginal-fuel model A model of when a gas-fired plant sets the wholesale price, comparing its modelled running cost against the observed price.
• Curtailment cost at a reference price What curtailed German renewable energy would have cost to compensate, at one reference rate applied to the curtailed volumes the grid operators report.
• Implied capacity factor Capacity factor: how much of a fleet's possible output at year-end capacity it produced over the year, from Ember's generation and capacity figures.
• Energy per token Estimated electricity per token of model output (about three-quarters of a word), measured at the graphics processor (GPU) and scaled to the facility.
• The north–south congestion index A monthly index of surplus power in northern Germany that the south lacks, inside one price zone, which no border flow or price shows.

Scenario A what-if: these inputs give this output.

• The perfect-foresight arbitrage bound An upper bound on what a battery could earn by charging in the cheapest hours and discharging in the dearest (arbitrage), with tomorrow's prices known.
• The cheapest-hours saving What a daily load that can move to other hours would save each year running only in the cheapest hours of the day, against paying the average price.
• Implied data-centre load The average electricity load an AI provider's tokens, each roughly three-quarters of a word, imply once an energy cost per token is applied.