Battery energy storage revenues in the ERCOT market have undergone a fundamental transformation in the past year.

​An oversupply of battery energy storage competing for fixed Ancillary Service (AS) volumes has prompted operators to pivot toward arbitrage-focused strategies. The safety net of AS has disappeared. Optimizers are now being tested on their ability to extract value from the volatile Day-Ahead and Real-Time energy markets.

Capture rates are becoming an increasingly important metric as arbitrage contributes a larger proportion of the revenue stack.

​Capture rates measure how close a battery comes to some definition of “perfect” performance. They’re calculated by dividing revenue by a theoretical maximum. In this analysis, we define that ideal measure using Real-Time Top-Bottom (TB) price spreads.

One-hour duration batteries are measured against the Real-Time TB1 spread at their node, and two-hour batteries against Real-Time TB2s. The Real-Time Market was chosen for the TB spreads - rather than the Day-Ahead Market - because batteries in ERCOT consistently earn a higher proportion of their revenue through arbitrage in Real-Time.

A 100% capture rate through Real-Time arbitrage alone is therefore impossible. It would require perfect market foresight and the ability to guarantee dispatch. In practice, operators improve capture rates by layering in Day-Ahead offers and opportunistic Ancillary Service participation, while maximizing their Real-Time revenues where possible.

The maturation of the ERCOT market has formed a new landscape for batteries. Consistently high capture rates are now harder to achieve as lucrative alternative revenue streams have become less accessible.

​Exceptions to the rule: lessons for battery energy storage systems from high capture months

​When Day-Ahead Market (DAM) volatility increases, Ancillary Service prices tend to rise alongside it. Higher AS prices reduce the opportunity cost of committing capacity to Ancillary Services. Under these conditions, batteries can often earn more revenue from stacking AS commitments in multiple hours than from chasing spreads in the Real-Time Market - or can effectively supplement energy arbitrage revenues with Ancillary Service revenue, rather than being reliant solely on arbitrage.

​Seasonal opportunities can increase Ancillary Service revenue opportunities for battery energy storage systems.

Thermal resource outages and higher reserve procurement can decrease competition for batteries to participate in certain Ancillary Services. For instance, in May 2025, spring maintenance season and high procurement volumes raised prices for Ancillary Services - particularly, Non-Spinning Reserve - as ERCOT had to rely more on batteries to fulfill reserve requirements.

Similarly, periods when peak Day-Ahead prices exceed peak Real-Time prices are valuable opportunities for optimizers to exceed the opportunity represented through Real-Time TB spreads.

A diversified revenue stack can improve capture rates by offering marginal opportunities beyond Real-Time energy arbitrage.

Subscribers to Modo Energy’s ERCOT Research can read the full report to learn:

• what capture rates can reveal about a battery’s underlying revenue strategy,
• how we built a naive benchmark to compare asset performance against a reproducible baseline,
• and how to identify what made some batteries more successful than others in different revenue environments - through a case study using the naive benchmark.

​In a nodal market, there is no universally winning strategy for battery optimization

Most optimizers struggle to maintain consistently high capture rates due to the large number of factors at play. 'High' capture rates also vary greatly month-to-month. So far in 2025, the average capture rate for batteries in ERCOT has ranged from as low as 38% in January to 85% in May.

Price spreads, and consequently, capture rate potential, are heavily influenced by transmission congestion patterns, which influence the value of protecting transmission equipment across the system through variance in Locational Marginal Prices (LMPs) at different nodes across the network - relative to the system average.