Algorithmic Power Trading in Germany

Executive Summary

Power trading for renewables has split into two distinct activities, and the split has direct consequences for how German asset owners and IPPs structure commercialisation in 2026.

The first activity is execution. Direktvermarkter who handle a 4 GW portfolio across Germany run roughly 50,000 trades per day in normal market conditions, more in high-volatility windows. The trades clear across Day-Ahead, the three pan-European intraday auctions (IDA1, IDA2, IDA3), continuous intraday (XBID), and the Regelenergie product family (aFRR, mFRR, FCR). The work happens 24/7, with algorithms continuously optimising positions and traders supervising in fast-moving conditions. Done well, it captures 8–15% more revenue per MWh than partial-stack alternatives. Done poorly, the same value disappears into imbalance settlement and missed intraday windows.

The second activity is strategic portfolio trading. Wind and solar assets carry long-tenor revenue exposures, Marktprämie, PPA tranches, merchant tail. Each tranche has its own duration, basis risk, and forward curve. Active futures and forwards trading turns those exposures into a managed book: long-dated hedges, calendar spreads, basis trades against the German Phelix benchmark and the front-month spot complex. Most Direktvermarkter route this work to a parent treasury, an external bank, or simply do not run it. The few who run it in-house treat the asset book as a tradable portfolio rather than a passive revenue stream.

The strategic implication for asset owners and IPPs in 2026: the choice of Direktvermarkter has become a choice between providers who execute and providers who execute and manage the strategic layer. The first protects your floor. The second moves the ceiling.

This article maps both layers, sets out the infrastructure that makes them work, and lays out a framework for asset owners and IPPs evaluating providers over the next eighteen months.

Pure Energy · an integrated direct marketer (Direktvermarkter), top-tier algorithmic power trader and originator across the renewable energy value chain, manages a 4.2 GW renewable portfolio in Germany and trades 7.5 TWh annually across Day-Ahead, Intraday and Regelenergie markets, with a layer of futures and forwards trading on top. From Düsseldorf and Monheim am Rhein, the firm sits at the intersection of asset owners, wholesale markets and C&I energy buyers, ranking among Top-20 in European intraday (XBID), Top-15 Direktvermarkter overall and Top-10 in wind-energy-only. The analysis below reflects what we observe inside our own trading operation in mid-2026.

01 · Two species of power trading

Trading for renewables in Germany comes in two species, and asset owners do well to recognise the difference.

The first species is execution. Production from a 4 GW portfolio of wind, solar and BESS assets clears across roughly twenty thousand market intervals every day, quarter-hour Day-Ahead products, continuous intraday flows, the three pan-European IDA auctions, and Regelenergie capacity bids. Each interval requires a price, a volume, a counterparty, and a reconciliation step. Across a year, the portfolio produces millions of clearing events, each of which represents either captured value or value left on the table. Execution is the work of getting that work done, well, fast, and at scale.

The second species is strategic portfolio trading. Behind every renewable asset sits a financial position: a Marktprämie revenue stream tied to monthly Marktwerte, a PPA tranche tied to a specified delivery profile, a merchant tail tied to wholesale spot. Each position has duration. Each carries basis risk against the futures complex. Each can be hedged, partially hedged, or left open depending on the asset owner's view and the forward curve's shape. Strategic portfolio trading is the work of managing those positions across the futures and forwards markets, typically German Phelix futures, EEX forwards, and increasingly cross-border products against neighbouring zones.

Most Direktvermarkter operate only in the first species. They handle execution at varying levels of quality, from passive Day-Ahead-only commercialisation through to integrated multi-market optimisation. Strategic portfolio trading they leave to the asset owner, to a parent treasury, or to an external bank. The reason is mainly capability: futures and forwards trading requires different infrastructure, different risk management, different REMIT discipline, and different people from the intraday execution desk. Building both, under one roof, is uncommon.

The split matters for asset owners because the two layers compound. An execution book that captures 11% above median, hedged with a futures book that locks in attractive forward curve dislocations, produces a different ten-year revenue profile from execution alone. The combination is the strategic case for an integrated provider.

02 · What execution-as-a-service actually looks like

Execution-as-a-service is what a Direktvermarkter sells when it takes on a renewable portfolio for direct marketing. The label is straightforward; the operation behind it is not.

A 4 GW portfolio in Germany generates roughly 50,000 individual trades per day across the venues a Direktvermarkter operates. The volume reflects the granularity of the modern market: Day-Ahead clears at quarter-hour intervals since October 2025, continuous intraday trades up to five minutes before delivery, the IDA auctions clear at three discrete gate-closure points, and Regelenergie products clear on their own auction schedules. Most trades are small: a few megawatt-hours, a few hundred euros of revenue, a fraction of a basis point of imbalance exposure. Aggregated across a year, they make up the difference between a top-tier capture rate and a median one.

The infrastructure behind those trades runs on three layers.

The market access layer connects the Direktvermarkter to EPEX SPOT, NordPool, regelleistung.net, and a handful of bilateral counterparties. Connectivity sits on dedicated lines, redundant gateways, and exchange-certified protocols. Latency to the Frankfurt exchange data centre matters more than most asset owners realise, quarter-hour Day-Ahead bidding rewards providers who can act on late market intelligence; continuous intraday rewards providers who can submit, modify, and cancel orders inside hundreds of milliseconds.

The algorithmic layer runs the bidding logic. Day-Ahead optimisers translate forecasts into bid curves. Continuous intraday algorithms reposition during the trading day as forecasts update and prices move. Regelenergie bidders price availability and energy components against the asset's opportunity cost on the spot complex. Imbalance optimisers watch the trading day from the position perspective, closing gaps before settlement crystallises losses. None of this works without forecasts good enough to drive it, wind, solar, demand, and price forecasts updating multiple times per hour.

The human layer sits on top. A 24/7 trading floor monitors the algorithms, intervenes when market conditions break the assumptions the algorithms operate under, and handles the exceptions algorithms cannot, a sudden shutdown of a major nuclear unit in France, a forecast revision that upends the day's positioning, a counterparty default. Algorithms work in normal markets. Traders work the rest.

Done as a coherent operation, execution-as-a-service costs the Direktvermarkter roughly two to four million euros per year for systems, connectivity, data feeds, and people, depending on portfolio size. It costs an asset owner attempting the same in-house considerably more, because the fixed costs are the fixed costs whether you are running 500 MW or 5 GW. The economic case for outsourcing rests on this asymmetry, which is why even sophisticated IPPs with plenty of in-house finance talent increasingly buy execution rather than build it.

03 · The strategic portfolio trading layer

Strategic portfolio trading is the layer most renewable asset owners do not realise they have until a Direktvermarkter shows it to them.

Take a typical mid-tier German renewable portfolio: 200 MW of wind under EEG support with awards across 2022–2024, 150 MW of solar under a mix of EEG and PPA structures, 30 MW of BESS in commercial operation. The portfolio earns Marktprämie payments tied to monthly Marktwerte, PPA cash flows tied to fixed and floating prices, and merchant revenue tied to wholesale spot. Each cash flow has a duration. Each carries basis risk, the risk that the price the asset realises diverges from the price benchmark the contract references. The duration and basis exposure together create a financial position. The position can be hedged across futures and forwards markets. Most asset owners never see the position because nobody shows it to them.

The trading work itself runs on three families of activity.

Long-dated hedging takes the asset owner's expected revenue stream and partially or fully fixes it through forward sales. A 200 MW wind portfolio earning Marktprämie linked to wind Marktwert can be hedged through Phelix futures plus a wind shape adjustment, locking in revenue against forward curve uncertainty. The trade-off is between certainty and upside: more hedge, less variance, less upside if prices rally. The right balance depends on the asset owner's leverage, capital structure, and view on the forward curve. The Direktvermarkter who can structure this trade adds a service layer most providers never offer.

Calendar spread trading exploits dislocations in the forward curve. The German power forward curve carries seasonal shape, winter premiums against summer discounts, driven by demand, wind generation patterns, and conventional plant availability. When the curve's shape diverges from fundamental drivers, calendar spreads become profitable trades. They are also natural hedges for asset owners with seasonal generation profiles: a winter-heavy wind portfolio can monetise winter premiums by selling calendar spreads when they trade rich.

Basis trading captures the gap between the futures benchmark (Phelix base or peak) and the asset's actual realised price. A wind asset's realised price tends to trade at a discount to the base benchmark because wind generation correlates negatively with price. The discount has a typical level, call it the wind shape factor, and trades around that level as fundamentals shift. Active basis trading takes positions on the shape factor, hedges the asset against extreme shape moves, and monetises mean reversion when the shape moves abnormally far. The work requires a research desk and a market view; most Direktvermarkter do not run it.

For asset owners with portfolios above roughly 100 MW, the strategic portfolio trading layer adds a second source of revenue and a second layer of risk management. The first source, execution, is operational. The second, strategic, is financial. Run together, they treat the asset book as a position to be managed actively rather than a stream of payments to be collected passively.

04 · The infrastructure that makes both layers work

Building both layers under one roof requires more than ambition; it requires specific infrastructure that very few Direktvermarkter own end-to-end.

On the execution side, the requirements come from market structure. Quarter-hour Day-Ahead bidding closes at 12:00 CET each day, with the bid window narrow enough to demand pre-computed bid curves rather than discretionary order entry. Continuous intraday trading runs until five minutes before delivery, with order books refreshing dozens of times per minute. The IDA auctions run on three separate clearing schedules, each with their own gate closure. Regelenergie auctions run weekly, daily, and within-day depending on the product. A coherent execution operation runs all of these on a unified scheduling system, with risk limits, position monitoring, and settlement reconciliation in a single dashboard.

On the strategic side, the requirements come from product structure. Phelix futures trade on EEX in Leipzig, with daily, monthly, quarterly, and yearly products extending out to three years forward. Forward OTC trades clear bilaterally with brokers, with longer tenors and customised structures. Cross-border forwards capture spread positions against the Dutch, French, and Nordic zones. A coherent strategic operation runs market data feeds across all these venues, position management against standard risk metrics (delta, vega for option components, scenario analysis for shape risk), and REMIT-compliant counterparty management.

The connection between the two layers happens at the position level. The execution book takes the day-to-day variation in production and price. The strategic book takes the longer-dated revenue exposure and the basis risk. The two books talk to each other constantly: an execution-side hedge that closes part of the day's exposure changes the strategic book's net position; a strategic hedge that fixes a portion of next year's revenue changes the execution book's working assumption about the wind shape factor. Without integration, the two operations duplicate hedges, miss offsets, and accumulate basis risk neither side measures.

The infrastructure investment for both layers, run together, sits in the low single-digit millions of euros per year for a Direktvermarkter at meaningful scale. The marginal cost of adding strategic portfolio trading on top of an existing execution operation is much lower than the cost of building either from scratch, which is the operational reason that integrated providers, once they have the execution layer, tend to build the strategic layer rather than refer asset owners elsewhere.

05 · The economic case for outsourcing trading

Most renewable asset owners and even quite a few IPPs face a recurring question: should we run trading in-house, or buy it as a service?

The honest answer for almost all asset owners is buy. The arithmetic is clear once you set it out.

A coherent execution-as-a-service operation costs roughly two to four million euros per year in fixed cost, systems, connectivity, exchange memberships, data feeds, risk management software, and a team of forecasters, traders, quants, and operations staff. The fixed cost is the same whether the operation runs 500 MW or 5 GW, because the seats on the trading floor and the data feeds in the rack do not scale linearly with portfolio size. Below roughly 5 GW under management, the unit cost of in-house execution exceeds the typical Direktvermarkter fee by a meaningful margin. For most German asset owners outside the top-tier IPPs, in-house execution does not pay back.

Strategic portfolio trading raises the threshold further. The fixed costs include not only the trading desk but a research function, a risk management function calibrated for forwards and basis exposure, and the regulatory compliance overhead that REMIT and EMIR impose on financial market participants. A strategic portfolio trading operation worth running costs another two to three million euros per year on top of the execution operation. Below roughly 10 GW under management, this rarely justifies the build.

The practical conclusion: even quite large IPPs benefit from outsourcing both layers to a Direktvermarkter who runs them at scale. The asset owner retains the strategic decisions, how much to hedge, what view to take on the forward curve, when to opt out of EEG support, while the Direktvermarkter handles the execution and the day-to-day risk management. The asset owner gets institutional-grade trading capability without paying institutional-grade fixed costs.

The control argument that asset owners sometimes raise, that in-house trading gives more control over the asset's commercialisation, survives less well than it sounds. A well-structured Direktvermarkter relationship gives the asset owner real-time portfolio dashboards, hedging strategy input, and audit-grade settlement traceability. The asset owner's strategic decisions get implemented through the Direktvermarkter's operation rather than through an in-house desk, but the strategic decisions themselves remain with the asset owner. The control argument confuses execution with decision rights, and the two are separable.

06 · A decision framework for asset owners and IPPs

Five decisions face asset owners and IPPs evaluating algorithmic power trading capability in 2026.

Decision 1: Insource or outsource. Below roughly 5 GW of portfolio scale, outsourcing execution to a capable Direktvermarkter dominates in-house build. Above 10 GW, the choice narrows but rarely tips toward fully in-house unless the asset owner has strategic reasons (a parent trading house, captive financing entity, or institutional treasury operation already running power trading capacity). Most asset owners and most mid-tier IPPs sit in the band where outsourcing is the right answer.

Decision 2: Execution-only or execution plus strategic portfolio trading. Execution alone covers the day-to-day commercialisation. Strategic portfolio trading adds futures and forwards management on top. Asset owners with portfolios under 100 MW typically benefit from execution alone, strategic portfolio trading at sub-100 MW scale rarely justifies the structuring overhead. Above 100 MW, the strategic layer adds revenue and risk management value that compounds over the asset's life.

Decision 3: Provider scale tier. Top-five providers (above 5 GW under management) offer scale efficiencies and broad operational stack but sometimes commodity-style relationship management. Mid-tier providers (1–5 GW, including Pure Energy) offer integrated capability with closer commercial relationships and more flexibility on structuring. Long-tail providers (below 1 GW) typically lack the strategic portfolio trading layer entirely and run thinner execution operations. The choice between mid-tier and top-tier depends on the asset owner's preference for scale versus relationship intensity; both can deliver capable execution.

Decision 4: Reporting transparency. A capable algorithmic trading provider gives the asset owner real-time portfolio dashboards covering production, market revenues by venue (DA, IDA, XBID, Regelenergie), imbalance attribution, hedge positions, and settlement traceability. Providers who treat reporting as a regulatory minimum signal where else they invest. The dashboard is a reasonable proxy for the underlying operation's quality.

Decision 5: Provider selection criteria. When the gap between providers accounts for 8–15% of attainable revenues on the execution side and another 1–3% per year on the strategic side, route-to-market selection moves out of procurement and into the strategy office. The questions plant operators and IPPs should put to providers:

• Does the provider operate a 24/7 trading floor with proprietary algorithms, or is the algorithmic layer outsourced?
• Does it run all market venues, DA, IDA1/IDA2/IDA3, continuous intraday, Regelenergie, within a single integrated stack?
• Does it offer strategic portfolio trading across futures and forwards, or only execution?
• How many trades per day does the operation run, and across how many venues?
• What infrastructure investment is the provider making per year, and how is it scaling?

Pure Energy operates as a one-stop-shop, a bridge between renewable assets, wholesale markets and C&I energy buyers, anchored by 4.2 GW under management in Germany, 7.5 TWh in annual trading volume, and a Top-20 ranking in European intraday (XBID). The trading operation handles the full execution stack across DA, IDA1/IDA2/IDA3, continuous intraday, and Regelenergie, supplemented by an active futures and forwards book for strategic portfolio trading. For asset owners and IPPs evaluating their commercialisation choices in 2026–2028, the integration we run across these layers is the kind of capability that compounds across an asset's full life.

07 · Outlook 2026–2030

Three theses for the next eighteen to forty-eight months.

Thesis 1: The split between executors and integrated providers widens. The capability gap between Direktvermarkter who run execution only and those who run execution plus strategic portfolio trading should widen as the futures market deepens and as asset owners learn to ask better questions of their providers. The integrated providers who own both layers extend their addressable market into mid-tier asset owners who previously bought execution from one provider and routed strategic trading to a bank or simply did not run it. The execution-only providers compress.

Thesis 2: AI-driven dispatch optimisation tightens the execution layer further. Machine learning models for short-term price forecasting, asset-specific generation forecasting, and intraday repositioning have moved from research to production over the past three years. The capture-rate gap between providers running modern AI dispatch and those running rule-based optimisation should widen by another 2–4 percentage points by 2028, with most of the gain concentrated in the BESS and hybrid asset segments where dispatch flexibility is highest.

Thesis 3: Cross-border strategic portfolio trading becomes meaningful. German power markets trade increasingly against Dutch, French, Polish, Nordic, and Baltic zones, with structural shifts in interconnection capacity, renewable penetration, and market design making cross-zone basis trading more valuable. Direktvermarkter with active cross-border operations, Pure Energy operates across Germany, the Netherlands, Sweden, Finland, Estonia and Japan, gain a positioning advantage as cross-border products mature. Domestic-only providers fall further behind.

08 · Closing

Power trading for renewables used to mean execution. In 2026 it means execution plus strategic portfolio management, and the two together demand capability most providers do not own end-to-end.

The arithmetic for asset owners works in one direction. Building both layers in-house requires fixed costs that pay back only above ten gigawatts of portfolio scale. Buying execution-only from a capable Direktvermarkter captures the operational layer but leaves the strategic layer unmanaged. Buying both layers from an integrated provider captures the full revenue stack at a cost that even mid-tier IPPs can justify.

For plant operators with renewals coming up, and for IPPs structuring portfolios for the next decade, the question to ask a provider has expanded. Execution capability is the floor. Strategic portfolio trading is what moves the ceiling. Asset owners who ask both questions, and pick providers who can answer both, will compound revenue advantage across the maturity cycle. Those who ask only the first will discover, year after year, that their portfolios earn less than they could.

Methodological notes and sources

All market data in this article are publicly published. No internal Pure Energy positions, forecasts, or trading statistics were used. The article discusses public market structures and trading practices. It does not constitute investment or trading advice. Specific decisions about route-to-market structuring require an asset-, regulatory-, and contract-specific case-by-case assessment.

Primary sources:

• EPEX SPOT. Day-Ahead and Intraday market specifications; quarter-hour Day-Ahead auction (since October 2025); IDA1/IDA2/IDA3 product specifications; XBID continuous intraday data.
• EEX (European Energy Exchange). Phelix futures and forwards specifications; calendar product structure; cross-border product launches 2025–2026.
• NordPool. Cross-border market structure; Nordic-Baltic interconnection data.
• Regelleistung Online & regelleistung.net. aFRR/mFRR/FCR auction data and product specifications, 2025–2026.
• Bundesnetzagentur (BNetzA). Direct marketing regulation, REMIT enforcement, market monitoring publications.
• ACER & ESMA. REMIT and MIFID II/EMIR guidance for power and gas market participants.
• Modo Energy. German power market analytics; intraday capture-rate research; BESS and hybrid asset commercialisation reports, 2025–2026.
• Capstone DC. Europe Battery Storage Edge, Insights 2025; renewable trading operations research.
• Wood Mackenzie / S&P Global Commodity Insights / Argus Media. Power market commentary and forward curve analytics.
• Industry media. Energie & Management (E&M), Montel News, Energate Messenger, ICIS, pv magazine Deutschland.

About Pure Energy

Pure Energy is an integrated direct marketer (Direktvermarkter), top-tier algorithmic power trader and originator across the renewable energy value chain. Based in Düsseldorf and Monheim am Rhein, Pure Energy GmbH manages a 4.2 GW renewable portfolio in Germany, trades 7.5 TWh annually, and ranks among Top-20 in European intraday (XBID), Top-15 Direktvermarkter overall and Top-10 in wind-energy-only. With operations across Germany, the Netherlands, Sweden, Finland, Estonia and Japan, and expansion underway in France, Italy and the United Kingdom, Pure Energy serves as a one-stop-shop, a bridge between renewable energy assets, wholesale power markets, and C&I energy buyers.

Power for Good · pureenergy.de