Battery trading performance: Demystifying normalised revenue and percent of perfect foresight

• Where should I connect?
• Which configuration should I choose?
• What revenue strategy best fits my risk-return profile?
• How much revenue can I capture from the merchant market?

Being able to confidently answer any of these questions typically means relying on a combination of different industry-accepted evaluation metrics. In theory, using the ‘right tool for the right job’ sounds simple enough but in reality, one of the highest risks you can take is to make decisions based on an evaluation metric without fully understanding its limitations or level of accuracy.

In this blog, we discuss what the main performance metrics are for batteries, how to calculate them and what their limitations are.

Two main battery storage trading performance metrics

There are several metrics to assess battery trading performance and revenues.

The normalised revenue and percent of perfect (PoP) capture rate are the two main metrics that stand out in assessing how attractive a battery’s revenues are as driven by their operational performance.

As a general rule, both of these metrics should be applied to battery gross margins (net revenue considering charging costs). For specific sites or locations, you can also account for the impact of Marginal Loss Factors and network tariff charges

Normalised revenue

What is it used for?

Normalised Revenue is a simple way to compare battery returns in a standardised way in AU$/MW or AU$/MWh. It can be particularly useful for:

1. Understanding how returns vary across different regions or markets (e.g. New South Wales vs Victoria, NEM vs ERCOT);
2. Comparing returns under different commercial models for the same asset (e.g. fully merchant vs partially merchant vs fully contracted)
3. Determining potential trade-offs in different trading / operating strategies
4. Eliminating variance in returns driven purely by differences in asset size

How is it calculated?

Normalised Revenue (AU$/MW/year) = Gross Margin (AU$/year) / Installed Capacity (MW)

or

Normalised Revenue (AU$/MWh/year) = Gross Margin (AU$/year) / Energy Throughput (MWh)

For example, a 100MW/200MWh battery storage system with a gross margin of AU$25 million in a year, while averaging one discharge cycle per day, would have a normalised revenue of AU$250,000 per MW per year, or AU$342.5 per MWh energy throughput per year.

What are its limitations?

Normalised Revenue is valuable for providing insight into how different variables can influence returns, but it can also provide a false sense of confidence that you are truly comparing ‘apples to apples’.

These metrics should be used with extreme caution when comparing battery storage systems across different locations. Particularly, they do not explicitly account for asset-specific operating limitations or network constraints, which can materially affect realised revenues. Also, you will rarely have all of the information to determine whether:

1. the observable revenues are representative of the entire value; and
2. if the underlying objectives driving observed outcomes apply to your organisation.

To put this in context with a few examples:

• A battery used as part of a wider portfolio hedging strategy will see a very different result to one focused solely on maximising its own wholesale electricity arbitrage revenues.
• A simple average of Normalised Revenues across all assets operating in a given region or market is not a good indication of the returns of your asset, as it does not account for site-specific differences like network constraints.
• The choice of normalisation basis can materially bias comparisons: using AU$/MW revenues will make longer-duration storage (e.g. 4-8 hours) look more attractive, while AU$/MWh typically favours shorter durations.

Percentage of perfect capture rate

What is it used for?

PoP is used to benchmark how well an asset performed relative to the best performance it could have achieved.

How is it calculated?

At first glance, PoP seems straightforward:

PoP Capture Rate (%) = Actual Gross Margin (AU$) / Maximum Possible Gross Margin (AU$)

The theoretical maximum in this calculation is the ‘Perfect’ part of PoP, or the maximum revenue the battery could have made if it were traded perfectly.

For example, if a battery earned AU$750,000 over one year and the maximum it could have earned was AU$1 million, its PoP capture rate would be 75%.

The tricky part here is determining what ‘perfect’ actually is. There has been growing debate recently about the utility of PoP, mostly stemming from live batteries falling short of the PoP capture rates assumed (especially on more volatile days). Some have even commented that PoP is not reliable or useful, especially on volatile days, where a low PoP capture rate doesn’t mean performance was poor if it was an outcome of trading in line with your risk profile.

This is a completely valid conclusion, and a useful point to keep in mind once you are actually trading a battery. Unfortunately, it doesn’t help much if you are trying to evaluate an optimiser’s performance claims or justify prospective returns to your lender.

PoP is a reliable and useful performance benchmark but the problem is the lack of standardisation and transparency in the assumptions that underpin it.

There are a variety of approaches to calculate the PoP and a commonly used approach is a Quality Ranking of A that denotes high quality, B for medium quality and C or D for low quality.

Is your ‘Perfect’ the same as mine?

1. Price Spread Method | Quality Rank: D

This approach estimates the maximum possible energy revenue from perfectly arbitraging daily price spreads, typically a single full cycle per day. It’s simple, intuitive, and fast to compute, making it popular for feasibility studies or quick performance snapshots.

However, it’s also highly idealised. It ignores real-world constraints like grid limitations, energy capacity, partial cycling, and participation in multiple markets. Price spread PoP can therefore be misleading when assessing actual operational performance.

2. Perfect Foresight Optimisation  | Quality Rank: C

A more rigorous method runs a full optimisation model for battery dispatch across available markets over a chosen period (e.g. a month or year) using actual market prices. The model assumes the battery has perfect knowledge of future prices and operates within defined limits (power, energy, cycles, etc.).

This gives a more realistic “perfect” benchmark but still assumes an unrealistic setup: the model is optimising over an entire period with perfect foresight, something no real trading system can do. Actual trading decisions must be made sequentially and in real time, without visibility of future prices.

3. Perfect Information Simulation | Quality Rank: B

Simulate sequential 5-minute decision-making. At each step the battery optimiser has perfect information over a limited look-ahead window (typically 24-48 hours), but only the immediate 5-minute dispatch is binding. After each interval settles, the horizon rolls one interval forward and the optimisation is re-solved with updated information.

This better mirrors real operations in a 5-minute spot market like the NEM: you optimise with the best available (here, “perfect”) near-term information, not omniscient foresight.

4. Perfect Information with real-world limitations | Quality Rank: A

Builds on the Perfect Information approach to include all the real-world limitations and characteristics of BESS:

• Asset: downtime, ramping limits, operating SoC bounds, degradation/throughput constraints, parasitic load, etc.
• Network: outages, curtailment, grid constraints, etc.
• Commercial objectives (markets, contracts, etc.)

When these factors are properly modelled, this approach provides a credible upper bound on achievable revenue.

Normalised Revenue and PoP are two commonly used metrics for evaluating a BESS’s trading performance. While both metrics are powerful tools, they can be applied or misinterpreted incorrectly without an appreciation of their inherent limitations.

Normalised Revenue offers a clean way to compare returns across projects, strategies, and markets, but it can obscure important differences in asset objectives, durations, portfolios, or commercial arrangements.

PoP is conceptually elegant but fraught with variation: the definition of “perfect” differs widely across the industry, and simplified methods can materially distort evaluation, particularly on volatile days. High-quality PoP benchmarking requires transparent modelling choices and realistic representation of physical, market, and operational constraints.

Ultimately, no single metric tells the whole story, and a robust performance evaluation requires understanding what each metric can and cannot reveal, and ensuring that the assumptions behind the numbers align with the decisions being made.

The Energy Storage Summit Australia 2026 will be returning to Sydney on 18-19 March. It features keynote speeches and panel discussions on topics such as the Capacity Investment Scheme, long-duration energy storage, and BESS revenue streams.

To secure your tickets and learn more about the event, please visit the official website. ESN Premium subscribers receive an exclusive discount on ticket prices.

About the Authors

Sahand Karimi, CEO/co-founder of OptiGrid, a spin-out from the University of Adelaide. He completed his PhD research with the University of Adelaide and CSIRO, focusing on optimisation solutions for hybrid battery plants.

An Electrical Engineer by training, Sahand has over eight years of experience in power systems analysis and modelling, with a focus on optimising renewable energy and battery storage operations. Alongside his research, he worked for two years at AEMO on the Frequency Performance Payments framework.

Henry Swisher, director of market strategy at OptiGrid, brings over a decade of experience advising government entities, investors, retailers, developers and large consumers on how to effectively manage risks and create opportunities in electricity markets.

As a subject-matter expert in electricity market risk management, route-to-market strategy, and commercial due diligence for energy businesses, Henry is well known and regularly consulted by leading organisations in the global energy sector.

In his role as director of market strategy, Swisher leads business development and market expansion efforts, building strong partnerships and positioning the organisation for growth.