German Battery Storage Earned Less as Power Prices Rose 22%

German electricity prices climbed 22% month-on-month in August 2026, and most battery storage strategies earned less than they did in July. The numbers move in opposite directions, and the gap between them is the whole story.

That contradiction matters now because autumn is closing in. Operators and investors are making Q4 allocation decisions, and August handed them a clear warning: the divide between static single-market strategies and dynamic cross-market optimisation widened sharply, and it is likely to widen further as daylight hours shorten and renewable output turns more variable.

Here is what the August data actually tells you about which strategies are built for a high-price environment and which are quietly bleeding value, so you can position with a sharper read on the autumn trading conditions ahead.

Why August’s price surge did not translate into stronger battery returns

The headline numbers looked like a gift for storage. According to Enspired Trading, the EPEX day-ahead auction average reached €127.67/MWh in August, a 22.3% jump on July. The intraday continuous volume-weighted average price (VWAP) climbed even harder, up 27.4% to €134.1/MWh.

On paper, that is exactly the environment batteries are supposed to love. Higher prices, more volatility, and on 28 August an intraday hourly spike to roughly €500/MWh, with quarter-hour intervals blowing through anything a normal week produces.

An exceptional event, not a strategy input On 28 August, individual quarter-hour intervals spiked above €3,000/MWh. This was a one-off dislocation driven by a generation shortfall, not a repeatable revenue line to model against.

So why did returns soften? The answer sits in the shape of the price curve, not its height. Battery arbitrage depends on the spread between the cheapest and most expensive hours, and the cheapest hours come from negative-price windows when renewables flood the grid.

Battery storage revenue mechanics across day-ahead, intraday, and ancillary markets interact in ways that make headline price moves a poor proxy for actual earnings, which is why the August divergence between rising power prices and falling strategy returns is less paradoxical than it initially appears.

Those windows compressed. Day-ahead prices fell below zero for just 55 hours in August, down from 79 hours in July. Fewer negative hours means fewer moments to charge for almost nothing, which erodes the floor of the arbitrage trade.

The weekly baseload trajectory tells the same story of easing conditions rather than sustained tension. Epignosis Insights tracked a steady decline through the month:

• Week of 3 August: €129/MWh (peak, during a heatwave)
• Week of 10 August: €108/MWh
• Week of 17 August: €96/MWh
• Week of 24 August: €91/MWh

Here is the mechanism, and it is the single most useful takeaway of the month. When the average price rises but the floor-to-peak spread does not widen in step, the arbitrage opportunity does not improve. It can actually narrow.

That tells you something uncomfortable if you have been using headline power prices as a shortcut for storage revenue potential. A high-price environment can reduce the exact spread conditions batteries need, which means an elevated monthly average is not a reliable proxy for how much a battery actually earned.

What the revenue data reveals about strategy performance in August

Look at the strategies one by one and a pattern emerges before you reach the punchline. Every single-market spot strategy fell month-on-month, despite prices rising across the board.

Continuous intraday trading held up best among the spot approaches at roughly €12,900/MW per month, though even that slipped around 10% from July. Day-ahead trading came in near €10,800/MW, and the intraday auction landed around €10,500/MW. Stacking wholesale products together lifted the result to about €15,400/MW, but the direction of travel was the same: down.

Ancillary services fared a little better at the top end. Frequency Containment Reserve (FCR), the fastest-responding balancing product that pays batteries to hold capacity ready to stabilise grid frequency, was the strongest individual strategy at roughly €15,300/MW per month. Voltcast’s pure FCR stack estimate put it slightly higher at €15,971/MW.

But that FCR figure carries a warning. Modo Energy found that realised benchmark FCR revenues fell by roughly one-third over the month, because widening day-ahead spreads pulled battery capacity out of FCR and into wholesale energy trading.

The full picture across strategies:

For context, Voltcast’s day-ahead arbitrage index put a 2-hour system at €9,373 for the month and a 4-hour system at €17,048, while Modo Energy’s benchmark German BESS revenue landed near €191,000/MW annualised.

Now the decisive number.

The performance ceiling, not the typical outcome suena energy’s forecast-based cross-market strategy generated roughly €25,300/MW in August, up 10% month-on-month and around 65% above the strongest individual strategy.

That 65% gap is what should concentrate the mind. It tells you that operators locked into a static allocation left a large slice of available revenue unclaimed every single month, and the cost of that rigidity climbs as market conditions shift faster.

The mechanics of cross-market optimisation and why August rewarded it

The outperformance was not luck. August produced an unusually sharp divergence between markets, and that divergence is precisely what a dynamic strategy is built to exploit.

FCR capacity prices barely moved, averaging €21.61/MW/h, down just 1.0% on July, according to Enspired Trading. At the same time, day-ahead spreads widened. When one market sits flat while another opens up, the value lives in reallocating capacity toward the market that is paying more. A static strategy simply cannot make that move.

When forecast accuracy determines who captures the spike

The second edge was forecasting. Anticipated generation shifts barely disturbed prices, because the market priced them in advance.

The clearest example was the partial solar eclipse on 12 August. Everyone knew photovoltaic output would drop, so participants adjusted ahead of time and the price impact stayed muted.

Contrast that with 28 August, when weaker-than-forecast solar and wind left a generation gap exceeding 10 GW around midday. Nobody had positioned for it, and that is exactly why prices exploded to roughly €500/MWh hourly and above €3,000/MWh at the quarter-hour level.

Only fast-responding, unlocked capacity could capture that window. The lesson is structural: forecast-based optimisation is a prerequisite for capturing extreme intraday events, not an explanation applied after the fact.

The third mechanism is the one most likely to catch operators off guard.

• Stable FCR clearing prices sat alongside widening spot spreads, creating a clear reallocation signal.
• Unanticipated generation shortfalls rewarded capacity that was free to respond rather than locked into a single product.
• A growing pool of competing assets quietly compressed realised FCR returns even as headline prices held.

That last point deserves attention. Modo Energy documented realised FCR revenues falling by roughly a third while clearing prices dropped only 1%. Germany’s FCR demand is finite, sitting at 584 MW of the ENTSO-E system-wide cap of 3,450 MW, and by late September intraday marginal capacity prices ranged from €15.55/MW to €172.03/MW across daily blocks.

What this tells you is blunt: a stable-looking clearing price is not a reliable signal of stable realised revenue when the asset pool is growing. Model future FCR returns off headline auction prices and you will systematically overstate what the market will actually pay.

Three competing philosophies for the autumn trading environment

August did not settle the strategy debate. It sharpened it. Three credible positions are circulating among industry researchers, and each reads the same evidence differently.

1. Dynamic arbitrage. Modo Energy argues for flexible pivoting, chasing wholesale spreads when they widen rather than anchoring to static ancillary positions. August’s data is the strongest exhibit for this view.
2. Ancillary anchoring. Aurora Energy Research counters that because Germany has no capacity market, FCR and aFRR remain essential sources of predictable cash flow, with wholesale trading acting as a secondary top-up.
3. Product stacking. IndexBox highlights that adding fast-responding reserve products such as Momentanreserve to an existing stack can lift project IRR by 1-2 percentage points, subject to product design and regulatory eligibility.

The genuine tension sits underneath all three. ESS News estimates merchant optimisation can theoretically reach around €200,000/MW per year for a two-hour system, but that number swings wildly month to month.

Energy transition volatility creates fundamentally different risk profiles across power markets at different stages of renewable buildout, and Germany’s August experience, where a 22% price rise coexisted with softer battery arbitrage, illustrates how transition-phase market structure can invert the intuitive relationship between price levels and storage returns.

A theoretical upper bound, not a forecast ESS News puts merchant optimisation potential near €200,000/MW per year for a two-hour system. That ceiling is subject to severe month-to-month volatility and should not be read as a bankable expectation.

That volatility is the problem. It makes traditional project financing harder even when headline earnings look attractive, which is why the German market is turning toward battery tolling contracts, where operators trade away spike upside in exchange for bankability and lower revenue risk.

Because Germany offers no capacity market to absorb merchant risk, this becomes a real financing decision rather than an operational preference. The choice between chasing optimisation upside and locking in tolling certainty is where the strategy debate actually bites.

Germany’s new capacity market legislation, introduced through the Electricity Supply Security and Capacity Act in July 2026, materially changes the financing context the article describes; operators assessing the tolling versus merchant optimisation trade-off now face a market structure that did not exist when most existing project finance models were built.

What August’s data actually changes heading into Q4

Two findings carry into autumn, and both reshape how you should read a monthly result.

First, high average prices do not lift battery returns unless spread conditions improve alongside them. August proved that a 22% price rise can coincide with softer arbitrage when negative-price windows shrink from 79 hours to 55 hours.

Second, the gap between static and dynamic strategies is now large enough that strategy choice outweighs asset-level decisions in sheer magnitude. A 65% revenue difference between the best single strategy and cross-market optimisation is not a rounding error, it is the decision that dominates.

The autumn setup points toward these conditions persisting rather than reversing. Solar output falls with the season, so expect negative-price hours to compress further, spreads to stay wide, and FCR cannibalisation pressure to build as more assets compete for a capped pool.

Germany’s carbon pricing framework directly influences the price floor dynamics that shape storage arbitrage conditions, because carbon costs embedded in fossil generation marginal prices affect both the absolute level of peak prices and the spread between dispatchable and renewable generation hours.

This does not mean abandon FCR or wholesale arbitrage. It means the relative value of each will keep moving, and the infrastructure to reallocate dynamically now matters more than the initial allocation itself.

Watch these indicators through Q4:

• Negative-price hour count each month, as a direct read on arbitrage floor conditions.
• Day-ahead TB2 spread width, the clearest gauge of wholesale opportunity.
• FCR clearing price versus realised revenue, to catch cannibalisation the headline number hides.
• Momentanreserve regulatory progress, given its potential IRR uplift.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions. Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors, and forward-looking statements are speculative and subject to change based on market developments.