I assume that there are a number of different stakeholders* who are each conducting their own retrospectives of large-scale** battery operations in South Australia over the recent windless (and quite volatile) days.

* we’d see these stakeholders separating into three broad categories:
Category 1 = those aligned with particular commercial interests looking at ‘market’ outcomes (e.g. the participants themselves, the autobidder vendors*** and the battery OEMs and supply chain).

Category 2 = those aligned more with looking at ‘system’ outcomes (e.g. from a security of supply standpoint … both now, and into the future)

Category 3 = various other spectators and commentators (like us at Global-Roam Pty Ltd, the providers of this WattClarity® service).

** remembering that small-scale batteries are one more aspect of the NEM that is invisible.

*** Way back ~5 years ago, when we prepared GenInsights21, we included two appendices covering bidding behaviours:

Appendix 22 reviewed ‘bidding busyness’:

(a)  Which we’re curious to revisit and update now, given the AER review of bidding guidelines that’s currently underway.

(b)  In Appendix 22, we identified more than a dozen different vendors (or internal developments) of auto-bidder and/or technology assisted bidding.  I wonder how many we’d find if we had a look now, because there seems to be new names popping up quite frequently?

 Appendix 23 reviewed bidding price bands.

Any review we do will be in our ‘spare time’ (or when the itch of curiosity becomes too great to resist!) and in dribs-and-drabs.

Take, for instance, the two charts below, which were produced from the same ‘Trends Engine’ query in ez2view on Wednesday morning 24th June 2026 (so with some data updated only until the 04:00 ‘next day public’ data cut-off.

The most recent 10 days in South Australia

In the first chart, we see the most recent 10 days of operations of the battery fleet in South Australia:

For those who’d like to access their own copy of this query, (where they have a licence) they can open their own copy of this query here.

I’ve highlighted on the chart the approximate ‘Aggregate Stored Energy’ at 12:00 on each of the 9 days that had fully elapsed in the chart (i.e. not yet Wednesday 24th June 2026).  With this we can see:

1)  A slight reduction across those two points from Friday 19th June 2026 to Saturday 20th June 2026;

2)  But then a steep reduction to 12:00 Sunday 21st June 2026;

3)  Which is what makes me wonder (as noted on the chart) if ‘the goose was cooked’ (in terms of stored energy) even before the ‘Market Demand’ rose, and volatility emerged that evening?

Over ~50 days in South Australia

Winding the chart back to a start at the start of May 2026 (i.e. so covering more than 50 days) we see the following:

Whilst the volatility (on Sunday 21st June 2026 and Monday 22nd June 2026) clearly stand out in terms of price outcomes, there are more periods we see where the maximum daily point for Aggregate Stored Energy was only modest.

… were these just occasions where the behaviour did not just come back to bite?