After a decade of rapid solar and wind deployment, capital is now shifting toward battery energy storage systems (BESS). Yet valuation still relies heavily on forecasts – raising the question of whether past mispricing of risk is being addressed, or repeated.
The clean energy industry has largely been built on forecasts. Highly sophisticated techno-economic models – often developed by large analyst teams – combine demand and supply assumptions, technology cost curves, fuel prices and interconnection constraints to simulate day-ahead price outcomes in complex power markets. These models became the standard tool for investment decisions: internally consistent, widely accepted, and often bankable.
No one claims these models predict the future. But they serve a practical purpose. Billions of euros in debt and M&A were underwritten on their outputs – not because they were “right”, but because they were the best available framework to unlock capital in a nascent market.
However, realized outcomes diverged. Models struggled to capture downside risk and volatility. As solar and wind penetration increased, prices and capture rates declined more than expected, leading to downward revisions in asset valuations and, in some cases, stranded pipelines and projects. The industry responded by turning to flexibility, in the form of BESS.
Today, BESS investment decisions are once again underpinned by forecasts – but with even greater complexity. Now that the same pioneers are building BESS, we must ask two things: are we repeating the same patterns, and can we afford to do so?
Why BESS revenue forecasts are likely to be wrong
Forecasting BESS revenues is inherently difficult, and most forecasts will probably prove wrong. Unlike solar or wind, BESS value is not tied to a single market, but to simultaneous participation across day-ahead, ancillary services, intraday trading, balancing mechanisms and capacity payments, each with distinct and shifting dynamics.
At the same time, the market is young and evolving, with limited historical data from operating assets to anchor assumptions. Second, the outlook for many BESS revenue streams is structurally unstable. Ancillary service markets are highly prone to saturation: small increases in capacity can lead to disproportionate price declines. Intraday and activation in balancing markets are even more difficult to model, as value emerges in real time and is highly path-dependent on short-term system conditions. Forecasts must also incorporate technical constraints such as battery degradation, cycling strategies and state-of-charge management – that materially affect outcomes but vary across assets.
The result is a forecasting framework built on multiple uncertain, interdependent assumptions. As complexity rises, so does the risk of error, making any single forecast a fragile basis for valuation.
A different approach – from forecasts to market consensus
With its BESS Price Intelligence, Pexapark takes a different approach. Keep the algorithms to a minimum. When increasing complexity leads to increasingly synthetic outputs, do the opposite. Instead of forecasting complexity, we focus on a simpler objective: identifying the market consensus on BESS revenue expectations.
For this, we need standardized and comparable price signals from active market players across different contract structures and tenors – tolls, floors, swaps – based on trusted, quality-controlled data sourced directly from the market.
Like forecasts, markets are not perfect predictors of where prices will go. They can be wrong. But on average, they tend to be less wrong than any individual forecast, as they reflect the aggregated expectations of multiple market participants at a given point in time.
The concept is not new. Most other commodities are priced off market signals rather than forecasts. Oil has Brent. Gas has TTF. Interest rates have SOFR. These observable benchmarks serve not just as underlyings, but as market consensus for value. In this sense, clean energy remains an outlier.
Pexapark has been applying this approach for solar and wind. Since 2022, our systematic market data collection process has gathered over 25’000 pricing data points from more than 100 developers, utilities, traders and optimizers, across some 20 markets. All data is independently verified by our team of price reporters.
Applying this to BESS is more challenging. Deal flow is even more limited, contract structures are still evolving, and contract standardization is nowhere in sight. As a result, market signals are harder to isolate.
Our ambition is not to predict the future. It is to enable our clients to evaluate all BESS structures with consistent data that reflects today‘s market as closely as possible. We are broadcasting the present. That is the foundation of Pexapark’s BESS Price Intelligence.
What the market tells us
Consider a practical example. According to Pexapark’s reference prices, a 3-year German tolling agreement for a 2-hour BESS asset starting in 2027 is currently priced at 133.27 kEUR/MW/year. Extending the tenor to 10 years, the price decreases to 99.83 kEUR/MW/year.
This difference reflects the term spread – how the market prices the same asset across time horizons. In this case, the downward shift and compression indicate how counterparties discount long-term value and price duration risk in real time.
Applied consistently across contracts, these observations allow for the construction of an implied forward curve for flexibility: year-by-year, market-based estimates of value.
We do not attempt to replicate the granularity of bottom-up forecasts that decompose each revenue component. Instead, we capture something different: the aggregated, observable consensus of market participants – how BESS value is priced in practice rather than assumed in theory.
A broadcast of the present
Our goal is simple: to provide a consistent, market-based view of BESS value across structures – tolls, floors, swaps, and merchant exposure.
Not everyone will welcome it. It challenges established approaches that have underpinned tens of billions in investment in the solar and wind space. And it will often be uncomfortable – especially when market data diverges from internal views.
But that discomfort is the point.
The last investment cycle showed what happens when valuation drifts too far from market reality. BESS now faces the same risk. If we continue to rely on increasingly complex and uncertain forecasts, we risk repeating the same valuation trap.
Is that a risk worth taking without mitigation? We don’t think so. Pexapark’s BESS Price Intelligence is designed to offer a consistent, market-based reference point for valuing flexibility across structures.
Pexapark’s BESS Pricing Benchmarks
Designed to offer a consistent, market-based reference point for valuing flexibility across structures.