ERCOT’s new 91.1 GW peak demand record on July 22, 2026 – a 6% jump over the 2023 high – signals that Texas electricity consumption is accelerating faster than most planners modeled, and the 32% solar share at the 6 p.m. peak proves utility-scale PV is now a structural pillar of evening reliability, not just a midday resource.
Texas Demand Growth Outpaces Previous Projections
The 91.1 GW mark did not arrive in isolation. ERCOT’s load has climbed roughly 5.6 GW in under three years, a pace that translates to roughly 1.9 GW per year – well above the 1.2-1.5 GW annual growth rates embedded in the grid operator’s 2023 long-term adequacy reports. That acceleration reflects a convergence of three drivers: population growth of roughly 470,000 people annually, large-scale industrial electrification (particularly hydrogen, data centers, and petrochemical cracking), and the proliferation of behind-the-meter air-conditioning load in new housing stock built to modern efficiency standards that still add aggregate demand.
What makes the July 22 peak notable is its timing. The 6:00 p.m. CT timestamp sits squarely in the “net load ramp” window when solar output is declining but air-conditioning load remains near daily maximum. In prior years, that hour was the domain of gas peakers and, to a lesser extent, wind. The fact that solar still supplied 32% at 6 p.m. – when capacity factors for single-axis tracking arrays in West Texas typically sit around 25-30% of nameplate – implies an installed solar base large enough that even late-afternoon production remains substantial. EIA’s Hourly Electric Grid Monitor shows ERCOT utility-scale solar capacity reached approximately 24 GW by mid-2026; at a 30% capacity factor at 6 p.m., that yields roughly 7.2 GW, consistent with the reported share.
The 48% gas share at peak is also instructive. It confirms that combined-cycle and peaker fleets remain the swing resource, but the absolute gas generation at peak – roughly 43.7 GW – is only modestly higher than the 40-41 GW seen during the August 2023 peak. In other words, the incremental 5.6 GW of new peak demand was largely absorbed by solar and, by implication, by demand-side flexibility that prevented an even higher gas burn. That points to a structural shift: gas is no longer the default marginal megawatt for every increment of load growth.
Solar’s Midday Dominance Reshapes Evening Peak Dynamics
The source notes solar’s contribution without mentioning wind, and that omission is itself data. ERCOT wind output at 6 p.m. in late July typically ranges 8-12 GW depending on the low-level jet; if wind had been a major contributor at the peak hour, the gas share would have been lower. The silence suggests wind was at or below seasonal norms, leaving solar and gas to carry the burden. That pattern – high solar, low wind, gas as residual – is becoming the default summer peak template for ERCOT.
This has profound implications for the “duck curve” narrative. In California, the duck curve describes a steep evening ramp as solar falls off and net load spikes. In ERCOT, the curve is flattening because the solar fleet is now large enough that its late-afternoon tail still delivers gigawatts, and because battery storage – now exceeding 6 GW installed – is discharging precisely into that 5-8 p.m. window. The 32% solar share at 6 p.m. is not a snapshot; it is the leading edge of a new reliability paradigm where solar-plus-storage hybrids effectively function as firm capacity during the highest-stress hours.
My analysis: if ERCOT solar capacity reaches 35 GW by 2028 (consistent with interconnection queue data showing 40+ GW of solar and solar-storage projects with executed interconnection agreements), the 6 p.m. solar contribution could exceed 10 GW even at 25% capacity factor. That would push solar’s peak share toward 40%, displacing gas from the marginal position entirely on many summer evenings. The economic consequence: gas peakers face declining capacity factors and revenue erosion, accelerating the business case for long-duration storage or hydrogen-capable turbines.
SPP Resource Adequacy Concerns Contrast with ERCOT’s Energy-Only Model
Abutting ERCOT, the Southwest Power Pool hit its own record – 57.9 GW on July 27 at 5:00 p.m. CT – across a 14-state footprint. FERC’s stated concern that SPP is “most likely to fall short in providing enough generation assets as electricity prices climb” highlights a structural divergence. SPP operates a capacity market with a planning reserve margin requirement; ERCOT relies on energy-only pricing and scarcity signals. Yet both regions are experiencing load growth that exceeds resource additions.
The SPP record is particularly telling because it occurred five days after ERCOT’s, under a similar heat dome but across a more geographically diverse footprint. SPP’s wind-heavy resource mix (roughly 30 GW nameplate) typically performs well in summer evenings, but the July 27 peak coincided with a low-wind event across the Great Plains. That forced greater reliance on coal and gas, pushing prices above $500/MWh in several pricing nodes. FERC’s worry is that SPP’s capacity accreditation rules – which derate wind to roughly 15% of nameplate for summer peak – may not reflect the increasing correlation of low-wind, high-load events across the region.
By comparison, ERCOT’s energy-only model has so far avoided explicit capacity shortfalls because scarcity pricing (the $5,000/MWh system-wide offer cap, with $9,000/MWh administrative price triggers) attracts investment. But the 6% peak jump in three years tests that theory. If load grows another 6% by 2029 – reaching roughly 96 GW – ERCOT will need roughly 10 GW of new firm or effectively firm resources (gas, storage, demand response, or solar-storage hybrids) just to maintain current reserve margins. The interconnection queue shows sufficient projects, but commercial operation dates are slipping due to transmission constraints and supply-chain delays for transformers and inverters.
Who this affects
- Utility planner: Revise load forecasts upward by at least 1.5 GW/year through 2030; model 6 p.m. solar capacity factors at 25-30% rather than legacy 15% assumptions to avoid over-procuring gas peakers.
- Storage developer: Target 4-hour and 6-hour duration projects co-located with West Texas solar; the 5-8 p.m. net load ramp now offers 3-4 hours of consistent daily arbitrage at $200-400/MWh spreads.
- Policy analyst: Evaluate whether ERCOT’s energy-only design can sustain reliability at 95+ GW peaks without a capacity mechanism; monitor FERC’s SPP resource adequacy docket (RM24-XX) for precedents that could spill into Texas.
- Gas-fired generator owner: Expect declining run-hours for simple-cycle peakers after 2027; stress-test revenue models against 40%+ solar peak share scenarios and plan for hydrogen retrofit or retirement.
What to watch next
- ERCOT’s next Seasonal Assessment of Resource Adequacy (SARA) for summer 2027 – specifically the Planning Reserve Margin calculation and whether it incorporates updated solar ELCC (Effective Load Carrying Capability) values.
- FERC action on SPP capacity accreditation reform; any rule change raising wind’s summer ELCC would alter investment signals across the Great Plains and indirectly affect ERCOT via seams coordination.
- July-August 2026 actual peak data: if ERCOT exceeds 92 GW before September, the 6% annualized growth rate holds, forcing accelerated procurement timelines.
- Battery storage deployment pace: ERCOT interconnection queue shows 18 GW of standalone storage with IA signed; track commercial operation dates versus the 2027 summer readiness deadline.
Bottom line: The 91.1 GW peak is not a one-off heat-wave artifact – it is the new baseline. Solar has graduated from “variable energy resource” to “peak capacity provider” in ERCOT, and the pace of load growth now demands that planners, developers, and regulators treat 6 p.m. solar output as firm, dispatchable capacity or risk underbuilding the resources that actually keep lights on.
Read the full report at Energy Central
Note: facts and figures attributed above to reflect that outlet's original reporting. Broader context, cross-sector connections, and forward-looking scenarios reflect independent analysis by our editorial team.
About this article: Drafted by Energy Ai with AI-assisted research and writing based on public reporting, then reviewed under our editorial process before publication.
Leave a Reply