In our previous posts, we outlined the core pillars of building a robust nodal power market simulation. They include:
- Leveraging probabilistic inputs.
- Inspecting model attributes to understand gaps and pitfalls with input data (e.g., unit bid curves, load forecasts and outages).
- Utilizing modern, cloud-native OPF tools.
- Executing backtests to tune the model, including point-in-time-correct and perfect-information backtests.
Now, it’s time to put these best practices to work. In this post, we’ll dive into a series of Distill simulation results in the ERCOT market.
The Only Certain Thing Is Uncertainty
Power markets and energy grids are not static systems and cannot be solved with a single deterministic forecast. Fundamental drivers, from shifts in available supply and participant bidding behavior to fuel markets and consumer demand, interact dynamically, meaning no individual feature dictates the outcome. While short-term volatility is often driven by weather and pushed to extremes by Winter Storm Uri and record-breaking heatwaves, the challenge lies in the structural unpredictability of the grid itself. Transmission system outages, demand response, and swings in renewable availability all introduce non-linear shifts within the system.
When we expand our analysis past the short term, drivers of volatility become even less predictable, and additional layers of uncertainty arrive. Policy changes, technological advancements, varying capacity build-out scenarios, large load centers, and shifting political tides all rewrite the rules of grid development. Market price fluctuations are driven by the continuous accumulation of these incremental changes.
Our goal with the price model, especially over a longer time horizon, is to ensure we account for this uncertainty. To anchor our variations in reality, we align our inputs with key market indicators from ERCOT and the EIA.
- Demand Growth: ERCOT forecast projects approximately 367,790 MW of demand in the ERCOT Region by 2032
- Resource Mix: ERCOT primary sources of new capacity tracked include natural gas, solar, wind and batteries
- Load Flexibility: ERCOT approves batch load process with a pathway to connect for large load customers who agree to let ERCOT curtail demand when the grid needs it most
- Gas Prices: The June 2026 STEO indicated slightly rising natural gas prices (but lower projected price increases when compared to the January 2026 report)
- Transmission Expansion: Transmission development included alongside load development in the Batch Zero Process
Modeling the Spectrum of Outcomes
Our approach yields a distribution of outcomes enabling us to inspect where risk and opportunities lie based on potential future shifts.
Our simulations included both single-lever analysis (e.g., isolating a gas price increase while keeping topology constant) and multi-lever analysis (e.g., simultaneously increasing renewables, storage, and transmission capability). We compare these outputs to a baseline of how the market solves today with recent fundamental, allowing us to identify which system changes impact future pricing the most and which ones effectively cancel each other out.
(Note: While our results are presented here as an aggregation of hub pricing, the underlying engine solves at the nodal and bus levels, which, as expected, prove to be much more volatile.)
The graphic above illustrates a handful of our analytical starting points. We selected 80 representative days to cover varying seasonal and hourly patterns, then updated input sets to tackle incremental growth over a 1-to-36-month horizon (ranging from 5% to 25% increases). We layered on more renewable energy, isolated load growth, modeled an increase in gas capacity, and finally combined these effects to reflect a more realistic, blended future. The graphic below breaks these simplified starting points into hourly results. Our gas prices reflect the range and small perturbation on the values published in the EIA STEO report.
Each input alone significantly shifts pricing. The potential for substantial reliability risks increases when the inputs are not combined. The sections below discuss the impact of each input and the eventual ideal result where a reliable, low-cost grid is maintained even through growth and uncertainty.
The Future is a Distribution
The core premise of this analysis, backed by ~6,000 individual solves, is clear: the same day's hub price can land anywhere from below-zero to scarcity pricing depending on how inputs shift.
While ~51% of our plausible futures sit above today's price with a long high-side tail, this should not be interpreted as a 51% chance of higher prices. Because these scenarios are unweighted by real-world likelihood, this distribution represents the possibility space, not a statistical probability.
Below, we break down the factors driving these directional swings. High load remains the primary catalyst for higher prices, especially during summer and winter peaks, while gas prices lift the baseline across the board. Conversely, adding capacity in nearly any form has the potential to suppress prices. However, if new generation is built without corresponding demand, transmission, and storage, heavy curtailment will result. True grid resilience depends on building new capacity alongside the right combination of storage and transmission to move electrons where they need to be in both time and space.
Renewable Energy: Timing & Geography Matter
Solar and storage dominate ERCOT’s interconnection queue. While most proposed megawatts target commercial operation dates within the next three years, only a fraction have finalized interconnection studies and signed an Interconnection Agreement (IA). Recognizing that the full queue is unlikely to materialize based on historical data in the Generator Interconnection Status (GIS) report, our analysis models a more pragmatic growth range of 5% to 25% over the current fleet size.
As expected, expanding renewables significantly reduces energy prices, but the timing of their generation dictates their market impact. When analyzed as an isolated lever, increased renewable penetration drives a 25% average price reduction. However, this impact is heavily constrained by the time of day; when modeled without corresponding demand growth or storage, it results in significant curtailment.
- Solar: Heavily concentrated midday. Because solar output peaks when the system already experiences a structural surplus, additional solar capacity accelerates midday curtailment. However, its predictable generation profile is effective at absorbing early afternoon air-conditioning load before the sun sets.
- Wind: Tends to produce most strongly off-peak when system demand is lowest, driving down overnight prices. Wind expansion offers an additional system benefit: wind patterns across ERCOT's distinct geographic footprints are largely uncorrelated.
A combined renewable fleet has the most potential to both reduce prices and manage curtailment.
Energy Storage: Balancing Time
An increasing share of the new generation in the ERCOT interconnection queue is shifting toward co-located solar-plus-storage. Operating in tandem, energy storage is critical to mitigating the midday energy gluts and severe negative pricing historically observed when solar builds out in isolation.
We identified storage provides its highest value to the grid when its growth outpaces solar capacity additions. Instead of suppressing prices indiscriminately, batteries reshape the daily price curve—shaving down afternoon net-load peaks and creating an artificial floor during midday surplus hours.
To illustrate this effect, the chart below isolates hub pricing on a mild operational day, demonstrating the marginal impact of a modest solar expansion against three distinct levels of storage buildout (0%, 135%, and 200% of the current fleet).
Managing the Demand Explosion
ERCOT’s latest long-term forecast projects a peak demand nearly four times current records. However, this extreme outlook is unlikely to fully materialize; it stems from a new forecasting methodology and intense speculation around data centers, which account for roughly 87% of that projected growth. Because early 2026 data shows actual demand falling short of prior aggressive projections, we excluded these extreme outliers from our analysis in favor of more modest, realistic growth expectations. Even with these moderated inputs, 90% of our demand scenarios resulted in price increases over the baseline.
The Power of Flexible Load
If ERCOT demand does push toward the higher end of these forecasts, flexible demand will be critical to managing the scarcity tail. We modeled varying levels of price responsiveness for new large loads, accounting for two primary mechanisms: state policies mandating curtailment when reserve margins shrink, and market prices forcing data centers into unprofitability.
Integrating this type of controllable load leaves normal operating hours untouched, but it alters extreme events, effectively collapsing the $5,000 scarcity price cap to approximately $1,600 and erasing outright cap events.
The alternative to flexible load is a costly reliance on emissions-heavy peaker plants and/or transmission + batteries. Instead, managing scarcity through demand flexibility means a high-renewable grid requires significantly less firm fossil-fuel backup to maintain reliability. Large loads like data centers can be successfully absorbed if they curtail just a small slice (17% to 30%) of their own demand on the worst days—representing a mere 3% to 6% of total system load—yielding a clear, consistent price benefit.
Ultimately, we can electrify the economy and support massive data center growth without triggering a wave of new fossil-fueled generation, provided that new load is built to be flexible.
Natural Gas: Two Sides of the Tail
Texas Energy Fund projects are receiving approval for commercial operations, including new dispatchable natural gas plants. While fuel markets put upward pressure on baseline power prices, the addition of dispatchable natural gas capacity can serve as a powerful structural hedge against extreme volatility. Even under aggressive fuel-price expansion scenarios, a robust thermal fleet can actually reduce tail events.
For instance, looking strictly at a high-load profile paired with surging natural gas prices, localized power prices spiked by an average of 58% over baseline. However, when we overlay a scenario that includes incremental gas capacity expansions alongside those exact same fuel and load increases, that price increase is blunted to 46%.
The true value of adding dispatchable thermal capacity isn't found in normal operating hours, but in how it reshapes the tail risk. By expanding the dispatchable supply stack, the market is less likely to exhaust its operating reserves and hit the high-priced peaker backstop or trigger administrative scarcity pricing mechanisms.
The economic proof is in the tail of the distribution: across our high-load simulations, introducing new gas assets flattens the extreme P90 outcomes, collapsing the tail risk from an unsustainable +1,933% spike down to a manageable +229% adjustment.
A critical caveat to our capacity analysis is the structure of these new assets. If new dispatchable gas is built behind the meter as captive generation for data centers, the grid loses its volatility cushion. These units fail to protect the market supply stack from scarcity pricing, while their localized fuel consumption simultaneously drives up the marginal cost of gas for front-of-the-meter generators. In short, behind-the-meter isolation risks shifting the entire price distribution upward.
Transmission & Combined Effects
Up to this point, we have analyzed supply, demand, and storage shifts as localized variables. However, electrons must still respect the physical limitations of the transmission grid. To test system limits, we modeled a coordinated expansion: rapid load growth alongside aggressive renewable buildout, paired with transmission expansion (new build plus rating upgrades) to relieve generation islands and load pockets. With coordinated transmission buildout, average wholesale hub prices remain mostly unchanged versus the baseline.
This outcome highlights the benefits of macro-level grid planning like proposed by the ERCOT Batch Zero Process. When transmission infrastructure scales in tandem with development, it mitigates severe localized congestion, prevents costly renewable curtailment, and safely routes zero-marginal-cost power directly to rising load centers.
Embracing the Distribution
Planning around a single price forecast no longer holds up. As our ERCOT analysis demonstrates, the grid has become far too dynamic and complex to be captured by one curve.
Building a strategy that merely survives the "average" day leaves market participants highly vulnerable to extreme tail risks. Conversely, over-indexing on worst-case scenarios can cause traders and developers to miss out on massive profit opportunities.
In a market where uncertainty is the only certainty, the advantage goes to participants who plan across the full distribution.
If you’re interested in digging further into our grid simulation results (or building scenarios of your own), please reach out.