As an individual then yes you can reduce your peak load on the grid by reducing power in the evening. But that is only the case because other people have already massively reduced the peak during the day by installing solar, see this graph they provide:
https://www.iso-ne.com/about/what-we-do/in-depth/solar-power...
What I think. Demand metering in a solar grid would likely motivate people enough to cut late afternoon and early evening usage. So I think policy and rate setting can match infrastructure and it's costs to demand.
The demand isn't greater in the evening, the relative demand, after you subtract all the energy provided by solar, is in the evening, during periods of the year when demand is low but solar supply is high.
It's like someone cured cancer and then someone came up with the llama curve, showing that some of the people who survived cancer got pushed off cliffs by llamas ten years later and everyone was like "Oh no, that cancer cure has caused the llama curve, what will we do?"
But some of that capacity is slow to respond and need time to ramp up, which may be longer than the time taken for solar to ramp down.
The simple solution to this is to start those slow to ramp sources a little earlier and throw away their excess energy output.
Obviously that's wasteful, but as long as the generation mix is overall cheaper and less carbon intensive with solar added, which it is, then it's not a big deal. And obviously as soon as you add storage then they can help these sources ramp more efficently at first and expand to replace them entirely over time.
All in all it's one of the least problematic problems that ever got its own name.
In general, yes, but if you look for a term called the duck curve which characterizes that frequently the load comes up after peak output, or even after sunset. For coastal CA wind often helps fill that offset as winds frequently come up near sunset, but it's not always a guaranteed thing.
So there does need to be some sort of offsetting storage - even if it's fairly short term. But, that gives us a nice set of market niches for renewable storage to grow up though with early smaller uses growing to larger uses. ie. peak offsetting for daily variation at small capacity, to larger uses of multi-day/weekly variation compensation, maybe extended weather variation offset, to massive uses for scales like seasonal offset.
https://en.wikipedia.org/wiki/Duck_curve
PS: and oddly enough, some of the old recommendations for efficiency are backwards for a newer renewable grid: e.g. getting a thermostat that schedules lower temperatures into the evening can push power use to later, vs pre-cooling during the day when you might consume the electric coming off panels w/o the need to store it in a battery.