Storage: the batteries that live off the spread

Grid-scale batteries and pumped hydro charge when power is cheap and discharge when it's dear — soaking up the daily price swing the Spread view measures. Here's how that arbitrage works, how fast storage is arriving, and the honest limits. The optimistic counterpart to Dunkelflaute.

Grid batteries live off the daily price spread, charging cheap at midday and discharging into the evening peak. German grid-scale battery power has grown from about 0.6 GW in 2021 to about 4.5 GW in 2026, on figures curated from published market reports.

As storage grows it erodes the very spread it feeds on — a modelled, illustrative effect here, not a forecast.

A transparent toy battery (1 MW / 2 MWh, 85% round-trip) run over real ENTSO-E day-ahead prices, plus a curated capacity series. The captured figure assumes perfect foresight — an upper bound. Pre-computed and static.

1 · A day in the life of a battery
avg day-ahead price + charge/discharge

The average day-ahead price (line) dips at midday when solar floods the grid and peaks in the evening. A 2-hour battery charges in the cheapest hours (bars below zero) and discharges into the evening peak (bars above) — buy low, sell high.

Upper bound. The captured spread assumes perfect foresight and an 85% round-trip efficiency — the same caveat as the Spread view's arbitrage figure. Real revenue is lower: round-trip losses, cycling wear, and more storage flattening the very spread it feeds on.

2 · Where storage is being built
installed grid-scale battery power · GW

Installed grid-scale battery power has grown roughly tenfold in a few years, led by Great Britain and Germany — the supply-side answer to renewable volatility. Power (GW) is what the market reports track; energy (GWh) differs by each project's duration.

3 · Storage vs the spread
monthly 2-hour spread (TB2) · €/MWh

A battery earns off the gap between the cheapest and dearest hours — the Spread view's TB2. The wider that daily spread, the better the arbitrage case. But the relationship isn't a free lunch: round-trip losses, cycling wear and cannibalisation — more storage flattening the spread as it enters — all pull real revenue below the line.

A battery earns by charging when power is cheap and discharging when it's dear. The figure here assumes perfect foresight and a stated round-trip efficiency, so it's an upper bound on what a real battery captures — the same caveat as the Spread view's arbitrage number. Round-trip losses, cycling wear, and the fact that more storage flattens the very spread it feeds on all pull real revenue below this line.