[secabeen]

Peak load without solar is not midday. Here's an NYT article from 1975 about introduction of Time of Use billing describing peak rates being in the morning and evening:

> Mrs. Wells changed her housework habits because for part of the year it costs her more than six times as much to use electricity from 8 A.M. to 11 A.M. and 5 P.M. to 9 P.M. as it costs during the rest of the day.

https://www.nytimes.com/1975/06/29/archives/experimenting-wi...

Current CAISO data shows that overall demand still peaks in the late afternoon to early evening. I picked a day in mid-august, and demand at 7pm is 40% higher (39GW) than at solar noon of 1pm (29GW).

https://www.caiso.com/todays-outlook

[epistasis]

You're making a blanket statement about peak time which is incorrect.

Historically in Califorinia, peak load has been in the afternoon, which I count as midday. At least, it's when solar panels are still pumping out a ton of power:

https://www.caiso.com/Documents/CaliforniaISOPeakLoadHistory...

You're posting a random day in winter in California, where overall consumption is low even at its highest, because there's very little demand for cooling. True peak for the California grid is ~50GW, not 25GW like today. You're also omitting all the residential solar that never gets on the grid that drives down midday demand in that graph.

Texas also has midday peaks, here's today and you'll see that even though its winter and very little AC is needed, peak is midday:

https://www.ercot.com/content/cdr/html/loadForecastVsActualC...

My statement was qualified with "most places." There will undoubtedly be some places with other peaks for which solar will not shave the peak. But in most places distributed solar will shave the peak.

[secabeen]

> You're posting a random day in winter in California, where overall consumption is low even at its highest, because there's very little demand for cooling.

You're just seeing the data for today. You can select any day you want.

Let's look at a really generous day for you, the peak annual usage from 2020: 47,121 MW on August 18 @ 15:57. On this day, the peak was indeed at 15:57. However, the demand remains high for hours past that. Demand is above 99% of peak until 5:30pm and above 90% of peak until almost 9pm. Solar production is down to under 1000MW by 6:45pm. Thus we have over 2 hours of near-peak demand when solar is not helping at all. No amount of additional solar (without batteries) will ever cover that 6:45-9pm period of high (if not peak, but it's close) demand.

https://www.caiso.com/todays-outlook

[vel0city]

> peak load has been in the afternoon, which I count as midday

Mid day is the middle of the day, as in noon. You might as well be arguing that you define three as five.

5PM is not "mid day". So you're cherry-picking time frames, making up definitions for things, and still not showing a mid day peak energy use, you're showing a late afternoon energy use.