Collected by someone else, for a purpose that was not yours#

You will spend your career using numbers collected by someone else, for a purpose that was not yours, with a definition you did not choose.

This is the normal condition of evidence — not a regrettable exception.

Four claims, and only the fourth is the one you wanted#

1. Operational demand measures electricity supplied through the grid by generators of a certain size and type.
2. Total consumption is that, plus everything supplied some other way.
3. Assumed: in 2005, that “everything else” was small enough to ignore.
4. Therefore, in 2005, operational demand was close to total consumption.

Step three is holding the whole bridge up#

1. Operational demand measures electricity supplied through the grid by generators of a certain size and type.
2. Total consumption is that, plus everything supplied some other way.
3. Assumed: in 2005, that “everything else” was small enough to ignore.
4. Therefore, in 2005, operational demand was close to total consumption.

Now change the year to 2025, and read step three again.

Minus fourteen megawatts#

South Australia’s Q3 2025 minimum operational demand.

Nobody unconsumed electricity that afternoon. Operational demand is a specifically defined quantity — and by that definition, grid supply briefly ran the other way while rooftop systems carried the load.

Three failures worth telling apart#

• Never sound — the bridge was only ever hope. A different source is needed.
• Sound, then the world moved — the term in the middle grew. The data is as good as ever; the inference built on it is not.
• Sound, but too wide for this decision — say what the proxy supports and what it does not.

A condition, not a caveat#

“Treat with caution” cannot be checked next year. A named breaking condition can:

“This breaks once more than about a fifth of supply is behind the meter, and I would see it first in the summer minimum.”

← Back to lecture