Before wind and solar arrived at scale—before they were elevated from boutique virtue to structural backbone—power planners lived in a world with a certain rough elegance to it. Not simple, not trivial, but intelligible.
There were two sides to the system, and their roles were clearly understood.
On one side: demand. Restless, erratic, stubbornly human. People consume electricity when it suits them, not when it suits a spreadsheet. They cook when they are hungry, not when a central planner deems it optimal. They shower when the mood strikes, not when a grid operator would prefer a smoother load curve. Consumption has always been jittery—spiking, dipping, shifting with habits, weather, culture, and a thousand small, uncoordinated decisions.
On the other side: production. And here, for a long time, lay the counterweight.
Generation was designed to be stable, predictable, and, above all, obedient. It did not complain about variability; it absorbed it. The entire architecture of modern power systems was built on that asymmetry. A volatile demand side balanced by a disciplined supply side.
To make this work, production itself was structured into two distinct roles.
Baseload: the quiet, relentless backbone. Think of river hydro, large thermal plants, nuclear—systems that produce a steady flow of electricity, day and night, largely indifferent to short-term fluctuations. They fill the bulk of demand, the portion that is always there, the irreducible minimum of modern life.
And then peak generation: flexible, responsive, almost reactive. Gas turbines, reservoir hydro, assets that can ramp up and down as needed, stepping in when demand surges and stepping back when it recedes. They are not always running, but when they are needed, they are there.
Between these two, the system held.
A jittery demand curve was smoothed over by a supply side that had both stability and agility. It was not perfect—nothing at this scale ever is—but it was coherent. Predictable enough to plan, robust enough to absorb shocks.
And then wind and solar entered the stage.
Not as a marginal addition, not as a supplement, but as a growing, increasingly central component of the generation mix. And with them came a subtle but profound inversion.
Demand did not change. It remained as jittery, as human, as inconveniently unpredictable as ever.
But production did.
Suddenly, the side of the equation that was supposed to provide stability began to exhibit variability of its own—often exceeding that of demand. Output became contingent on weather patterns, on wind speeds that can shift within minutes, on sunlight that follows a daily cycle and disappears entirely for half of it, modulated further by clouds, seasons, and geography.
The buffer was gone.
Or rather, it was still there—but it had been reassigned. Conventional generation, once the backbone, was increasingly pushed into the role of shock absorber. It now had to compensate not only for the natural volatility of demand, but also for the imposed volatility of supply.
Two moving targets, one shrinking stabilizer.
And as anyone who has dealt with systems under stress will tell you, that is not a comfortable configuration.
It becomes progressively more complex. More fragile. More expensive.
Because maintaining stability in such a system requires redundancy. Backup capacity that sits idle until needed. Infrastructure that must be built and maintained not for constant use, but for intermittent rescue operations. You pay for assets twice—once for the renewable capacity, and again for the conventional systems that must shadow it, ready to step in at a moment’s notice.
And then there are the less intuitive effects.
A calm, windless day is an obvious challenge: output drops, and other sources must fill the gap. That is easy enough to grasp.
But a very windy day can be just as problematic—sometimes more so. Generation spikes beyond what the system can absorb. Prices collapse or turn negative. Turbines must be curtailed, not because they cannot produce, but because producing becomes destabilizing or even physically unsafe. Infrastructure designed to harness variability finds itself throttling it to avoid damage.
Too little is a problem.
Too much is also a problem.
And threading the needle between the two, in real time, across an interconnected grid, is not a trivial exercise. It is a constant negotiation with physics, economics, and the limits of control.
Wind and solar are variables.
Variables that introduce additional complexity into a system that was originally designed to minimize it. Variables that require compensation, coordination, and, ultimately, cost.
And those costs do not disappear simply because they are politically inconvenient. They are embedded—in grid upgrades, in balancing services, in capacity markets, in the quiet necessity of keeping conventional plants alive even as they are publicly declared obsolete.
The nuisance is that wind and solar never stop demanding accommodation.
A system once defined by the stability of its supply side now spends an increasing share of its effort managing the instability of that same side. And while it can be done—up to a point—it is neither as simple nor as inexpensive as the early narratives suggested.
The grid still works.
But it works harder. And sometimes not at all.
And it charges accordingly.
https://notalotofpeopleknowthat.wordpress.com/2026/03/13/a-windy-day/