The revenue stack for battery energy storage systems in ERCOT is changing. Ancillary Service clearing prices are declining, relative to Energy prices. Additionally, more batteries are becoming commercially operational every month - meaning more competition in those markets.
As a result of these changes, more battery energy storage capacity is available to perform Energy arbitrage. And the proportion of revenues from Energy has increased to 26% in 2024 - as highlighted in our H1 revenue round-up.
By the end of 2025, installed battery energy storage capacity is expected to increase by 3x - and potentially even more. Meanwhile, procured Ancillary Service volumes will remain relatively constant.
As such, Energy arbitrage will make up an increasingly large proportion of the revenue stack. So, how can operators ensure that their systems are earning reliable revenues efficiently while using an arbitrage-focused strategy?
Energy arbitrage can be done in a few different ways, such as:
- Buying and selling Energy during the lowest and highest-priced hours - exclusively in the Day-Ahead Market.
- Buying and selling energy during the lowest and highest-priced hours - exclusively in the Real-Time Market.
- Or operating across both markets - by buying at the lowest hours in either the DAM or the RTM, and then selling across the highest hours in either market.
So, let’s look at how these different strategies might impact revenues.
Battery energy storage revenues across Energy arbitrage strategies
In the first half of 2024, two-hour battery energy storage systems in ERCOT earned an average of $38/kW.
They did this while cycling an average of 0.45 times per day - equivalent to 81 total cycles over the time period.
But what if a two-hour battery had decided to only pursue revenues via Energy arbitrage? How would its revenues compare to the Index?
Perfect foresight of Day-Ahead vs. Real-Time prices
Imagine a battery energy storage operator has perfect foresight of prices in the Day-Ahead and Real-Time Markets.
The operator is trading a two-hour battery energy storage system. They structure the battery’s bid and offer curves to buy and sell energy in whichever of the Day-Ahead or Real-Time Markets produces a larger two-hour price spread. (And they forego Ancillary Services entirely.)
If this approach had been taken every single day through the first half of 2024, the battery would have earned $49/kW. This is 31% higher than the $38/kW earned by the average two-hour battery in ERCOT, according to the Modo BESS Index.
Of course, this would’ve required cycling the battery every day.
So, let’s imagine that the battery performs the same number of cycles during this period - 0.45 per day, or 81 in total - as the actual average.
With perfect foresight, this would’ve meant capturing the 81 best days of spreads. And this still would’ve resulted in revenues 11% higher than the Index.
This perfect foresight DA/RT strategy - while opting to perform 0.45 cycles per day - would’ve led to earnings of $0.52/kW/cycle.
However, capturing the better of the Day-Ahead or Real-Time spread on every operating day would be nearly impossible. But this indicates that the opportunity to earn competitive revenues - relative to the Modo BESS Index (which includes revenues from Ancillary Services) - by opting for an Energy arbitrage-only strategy does exist.
So, how might an operator attempt to capture this revenue realistically?
Day-Ahead Energy arbitrage
Imagine our two-hour battery decides only to buy and sell Energy in the Day-Ahead Market. It structures its bid and offer curves to buy and sell energy during the two lowest and two highest-priced hours every day.
This would’ve resulted in cumulative battery energy storage revenues of $33/kW during this period.*
To achieve these revenues, the battery would have to cycle every day, meaning per-cycle revenues would be just $0.18/kW/cycle.
However, not every spread is worth capturing. If a two-hour battery cycled the same amount as the average battery in the Index - capturing the 81 most lucrative Day-Ahead spreads - per-cycle revenues would’ve increased to around $0.35/kW/cycle.
Real-Time Energy arbitrage
Now, let’s look at performing Energy arbitrage in the Real-Time Market instead.
Capturing the best available spreads in the Real-Time Market every day would’ve resulted in cumulative battery energy storage revenues of $37/kW. This is just below the two-hour Modo Index.
But, this still requires cycling the battery every day, often to capture minimal spreads. As a result, per-cycle revenues would have been 56% lower than the Index.
Reducing the battery’s cycling rate to the average two-hour battery cycling rate (0.45 cycles/day) over the first half of the year would’ve reduced cumulative revenues to around $31/kW. However, per-cycle revenues would’ve been $0.40/kW/cycle.
This is still below the Index - but the Ancillary Service revenue opportunities that existed in the first half of 2024 won’t necessarily be around for much longer.
Modo’s ERCOT subscribers can read the rest of the article below. Learn about the mechanics of implementing a Day-Ahead/Real-Time optimization strategy that:
- provides daily revenue security,
- hedges against the volatility risks of the Real-Time Market,
- doesn’t require daily cycling of a battery energy storage system,
- and achieves per-cycle revenues on par with the H1 2024 two-hour index.
Implementing a DA/RT optimization strategy for a battery energy storage system
Maximizing wholesale revenues is one of the main focuses when operating any type of resource. However, it’s important to do this while keeping in mind other factors, such as: