Top-Bottom (TB) spreads serve as a benchmark for the revenue that a battery can earn from energy arbitrage. By comparing TB spreads across durations and locations, stakeholders can identify optimal strategies and sites for battery energy storage deployment.

Energy arbitrage involves charging batteries when electricity prices are low, and discharging when prices are high. When Ancillary Services become saturated with battery energy storage capacity, energy arbitrage tends to make up the vast majority of merchant revenues for batteries.

This “buy low, sell high” strategy capitalizes on the daily fluctuations in power prices to generate revenue.

A TB spread is a simple but powerful way to quantify this opportunity. It measures the difference between the highest and lowest electricity prices on a given day.

• A TB1 spread is the highest hourly price minus the lowest.
• A TB4 spread is the sum of the four highest hourly prices minus the four lowest.

TB spreads act as a benchmark for the maximum possible daily revenue a battery could earn from arbitrage.

How does a TB spread differ between the Day-Ahead and Real-Time markets?

TB spreads behave differently in Real-Time and Day-Ahead markets, depending on the region and its structural characteristics.

In ERCOT, TB spreads are typically higher in the Real-Time market due to the system’s high volatility. Real-Time prices in ERCOT respond to sudden changes with large price swings and thus high spreads.

In contrast, CAISO often produces higher TB spreads in its Day-Ahead market, the Integrated Forward Market (IFM). This can reflect more predictable constraints priced into the Day-Ahead schedule, such as renewable curtailment and anticipated congestion.

TB spreads measure opportunity across time and space

TB spreads offer diminishing returns as duration increases. TB1 captures the most lucrative opportunities by reflecting peak price volatility, while each subsequent hour adds progressively less value.

Year-over-year tracking of TB spreads helps identify evolving market conditions and volatility.

Markets like ERCOT, which lack a capacity market, tend to exhibit more price volatility compared to regions like PJM that rely on more structured market mechanisms.

Within a single region, differences in spreads between individual nodes also arise. This is primarily due to transmission congestion. These nodal patterns can highlight grid bottlenecks and emerging opportunities for storage investment.

P-values help measure the consistency of revenue

P-values refer to percentiles within the distribution of recent TB spreads. They summarize how often different revenue outcomes have occurred historically.

• P50 (50th percentile): the median value. 50% of days in the measured sample had higher revenue opportunity, and 50% were lower. P50 can represent the expected outcome.
• P10 (10th percentile): 10% of days achieved lower revenue. This reflects a reliable expectation for daily revenue.
• P90 (90th percentile): 90% of days achieved lower revenue. This represents a high revenue scenario.

P-values help stakeholders gauge the volatility of arbitrage revenue at different locations or under different strategies.

In June 2025, ERCOT and CAISO, for example, had a similar P50 of about $90/MW-day, but ERCOT had a wider distribution, with a lower P10 and a higher P90. CAISO revenues were more consistent, while ERCOT had a wide range of potential revenue in a given day. There was more upside on some days, but other days had extremely low revenue opportunities.

What influences a TB spread?

Most of the US electricity market is nodal, so prices are influenced by supply and demand at a particular location.

These prices and their resulting TB spreads are influenced by three components:

1. Energy: The baseline cost to supply an additional unit of electricity to the whole system.
1. Influenced by fuel costs, renewable generation, and demand peaks.
2. Congestion: The added cost due to limitations in the transmission system. If certain lines are congested and cheaper electricity cannot reach a part of the grid, higher prices result at nodes behind the constraint.
1. This leads to intra-regional price separation, which can result in sharp increases in prices at some nodes.
3. Loss: The cost that results from energy lost during transmission over distances.
1. Not all ISOs in the US include a loss component, and in general, this is only a minor contributor to price.

Understanding how much of a TB spread at a node is driven by energy versus congestion helps evaluate the sustainability of arbitrage revenues.

TB spreads in action

We can explore the drivers of TB spreads by examining a day’s price components at different nodes.

Plano Storage 4 is a battery energy storage system in Monterey Bay, California - sited alongside transmission lines that bring power up and down the California coast. On July 15, 2025, it had a TB4 spread of $68/MW-day. Mustang BESS 1 is located 110 miles to the South-East, near multiple solar generation resources. Its TB4 on July 15 was nearly double at $121/MW-day.

Reviewing the hourly price data, the elevated spread at Mustang BESS was driven by a negative congestion component throughout the afternoon and evening. This congestion was the result of excess solar that caused transmission lines in the local area to be at risk of becoming overloaded. This resulting export constraint caused the minimum prices for the day at Mustang’s node to be lower than others across the system, like at Plano, thus increasing its spread.

Understanding TB spreads offers a valuable lens into the arbitrage revenue potential for battery energy storage operators. As market conditions evolve, tracking TB spreads and their drivers remains essential for navigating volatility and maximizing the value of storage in modern power systems.

Click here to see TB spreads across US ISOs and Europe in the Modo Energy Terminal.

If you have any questions regarding the content of this article, reach out to sam@modoenergy.com in the US Research team.