Episode Summary
Executive Summary: The episode examines Texas/ERCOT’s grid-scale battery boom: rapid buildout has crushed merchant revenues by saturating ancillary services and arbitrage, yet looming data-center load growth and thermal retirements could restore volatility and battery value around 2029-2030. The core tension is whether developers can survive the current trough long enough for demand growth to reset the market.
Main Topics: ERCOT battery buildout and market saturation (Priority: 5/5): Texas storage expanded from a few hundred MW at the start of the decade to well over 17 GW operational/approved, creating a far larger fleet than the market’s shallow ancillary-service needs. Collapse in merchant battery revenues (Priority: 5/5): As batteries crowded into ancillary services and then arbitrage, spreads compressed and average returns fell sharply; current revenue levels are below what most projects need to finance new builds. Weather, volatility, and battery duration (Priority: 4/5): Extreme weather still creates spikes, but milder periods plus more battery dispatch have reduced volatility. Longer-duration batteries are becoming more valuable because they can survive multi-hour or multi-day events. Financing strategies amid low revenues (Priority: 4/5): Developers increasingly need tolling, revenue floors, swaps, or other offtake structures to bridge the low-revenue period, though these deals are hard to price and can limit upside. Data-center load growth as the demand-side reset (Priority: 5/5): Massive projected load from data centers in Texas could outpace current oversupply and bring volatility back, but timing is uncertain and behind-the-meter generation could mute the grid impact. Queue slowdown and project attrition (Priority: 4/5): Signs of pullback are emerging: queue growth is flattening, withdrawals are rising, and project finance is getting harder, even though many gigawatts remain in the pipeline. ERCOT as a template for other markets (Priority: 3/5): The same cycle—ancillary-service saturation, then arbitrage cannibalization—can play out in other storage-heavy markets such as PJM, CAISO, and the UK, albeit with different market structures.
Key Arguments: ERCOT storage revenues have been “cannibalized” by rapid deployment: the more batteries that enter, the more they suppress the very price spreads and ancillary-service value they were built to capture. The market’s current weakness is structural rather than just weather-driven; most of the revenue decline is due to supply growth and market saturation, though mild weather has also reduced volatility. A two-hour ERCOT battery generally needs roughly $100/kW in revenue to meet financing hurdles, while recent realized revenue near $30/kW is not financeable on a merchant basis. Despite current losses, battery owners may stay afloat through self-financing, balance-sheet strength, or older projects that enjoyed better early-year revenues. The next major battery upcycle depends on whether data-center-driven demand shows up fast enough to tighten ERCOT before today’s depressed revenue period lasts too long. Behind-the-meter generation/storage at data centers could reduce observed grid peak demand, meaning headline load growth may overstate the actual lift to ERCOT prices. ERCOT’s likely re-tightening window is around 2029-2030 in the model discussed, but a hotter summer or winter storm could bring volatility back sooner. The same lifecycle seen in ERCOT—shallow market saturation followed by migration to deeper energy markets—will likely repeat in other regions as battery penetration rises.
Data Points: ERCOT grid-scale battery capacity at start of decade: ~200 MW - Approximate installed storage capacity in Texas at the beginning of the 2020s ERCOT grid-scale battery capacity by late June of this year: 16.5 GW / nearly 29 GWh - Host’s opening framing of Texas battery growth ERCOT capacity figure cited by guest for 2026: 17-18 GW fully approved; 20+ GW completed construction/near commercial operation - Guest’s current system snapshot New capacity added last year: 6 GW - Host’s opening summary of recent buildout Revenue decline: ~84% less in 2025 vs 2023 - Average ERCOT battery earnings compared across years Days with high earnings in 2023: 58 days - Days when average battery earned at least $0.50/kWh Days with high earnings in the prior year: 3 days - Contrast with 2023 high-price days Installed-kW revenue in 2023: nearly $200/kW - Strong summer/ancillary-service period Installed-kW revenue in last 12 months: ~$30/kW or slightly less - Recent merchant revenue level in ERCOT Rule-of-thumb revenue needed for a 2-hour ERCOT battery: ~$100/kW - Approximate financeability threshold Ancillary service share of battery revenues in 2022-2023: 90%+ / around 80-90% - Early decade revenue mix for batteries Current revenue mix: ~80% from energy arbitrage - Shift as ancillary services saturated ERCOT peak load: 91 GW - Current system peak referenced in the discussion Battery project capacity in interconnection queue: 150-160 GW - Queue has stalled at this level Battery withdrawals from queue in first half of year: 13.7 GW - Host’s opening remark on withdrawals Queue withdrawals after interconnection agreement last year: 13-15 GW - Guest’s estimate of projects backing out New capacity expected by end of year: 5-6 GW - Projected annual ERCOT battery additions Data-center project interest in Texas: 400-500 GW - Headline stated interest to connect (guest notes this is inflated) Likely portion of nameplate data-center load that becomes actual grid draw: 50-70% - Guest’s rough long-run rule of thumb Modeled return of meaningful volatility: 2029-2030 - Guest’s internal production-cost model timing PJM regulation market size: ~600-700 MW - Used as comparison for shallow ancillary market depth PJM battery fleet size: ~500 MW - Comparison market where regulation prices are currently elevated Battery duration trend in ERCOT: ~1 hour to nearly 2 hours average; some 2.5-4 hour projects - Shift toward longer-duration storage Commercial operation timing of new storage: 60-100 days for VPPs; batteries take years - Referenced through ad copy as comparison for flexibility assets
Pivotal Quotes: "“high prices cure high prices”" — Brant Vermillion: Describing cyclical power-market dynamics and why current low revenues may eventually attract fewer new batteries and more load tightness "“the revenue opportunity for storage has collapsed is a strong word. But I mean, ultimately, that’s kind of what it has done”" — Brant Vermillion: Summarizing how battery economics in ERCOT deteriorated as deployment scaled "“we’re starting to see volatility re-emerge in our model around 2029, 2030”" — Brant Vermillion: Guest’s timing estimate for when ERCOT battery revenues could recover meaningfully
Implications: ERCOT storage is in a cyclical trough: developers need patience and offtake structures, while investors should focus on duration, load growth, and interconnection risk. The next upside likely hinges on data centers, but behind-the-meter power could delay the rebound.