Energy was last better‑than‑free: yesterday Next

Research notes

Figures are for London unless they say otherwise.

What happened

• Fri 3 Oct: one marginal negative half-hour at 23:30 (−0.1p).
• Sat 4 Oct: the cheapest day of the period. London averaged 2.2p, maximum only 16.5p, below zero for 9 hours (low −0.4p).
• Sun 5 Oct: below zero for 8 hours, including an unbroken 7 hours from midnight, low −1.3p, average 4.8p. All 14 regions negative all three days.

Likely causes

• Storm Amy, the first named storm of the 2025/26 season. Named by the Met Office on 1 Oct, it hit on 3 to 5 Oct with a 96mph gust on Tiree, a provisional Northern Ireland Oct record of 92mph, and a new UK Oct low-pressure record of 947.9hPa. (confirmed by source: Wikipedia, Storm Amy, and Met Office, Storm Amy named)
• Bloomberg reported on 3 Oct that the UK day-ahead price for Sat 4 Oct settled below zero at −£0.42/MWh on Epex Spot, with Germany (−€0.29/MWh) and the Netherlands also negative. It said the UK had not had a full day average below zero since 2023. (confirmed by source: Bloomberg via Energy Connects, 3 Oct 2025; original at Bloomberg)
• Elexon-metered wind averaged 13.8GW (Fri), 14.0GW (Sat) and 14.2GW (Sun), peaking at 16.5GW on Sat evening. Gas averaged 3.5GW on Sat and 3.3GW on Sun. (confirmed by source: Elexon FUELHH, 3 to 5 Oct)
• Weekend demand was low (total supply about 22GW on both days, dipping to about 16GW at 04:00 on Sun while GB exported over 3GW). (confirmed by source: same Elexon query)
• A Montel analyst, Fintan Devenney, put the negative prices down partly to grid bottlenecks: "There's very little grid infrastructure to move all that power down from Scotland". (confirmed by source: West Midlands News, 7 Oct 2025)

Things to be careful about

• No wind record was set. Bloomberg's piece said wind was forecast to reach 24GW against a record of 22.5GW, and several sites repeated that as if it had happened. NESO's next confirmed record was 22,711MW on 11 Nov 2025, and that announcement gives the previous record as 22,523MW on 18 Dec 2024, so Amy did not beat it. (RenewableUK, 19 Nov 2025; Electrek)
• The "17 straight hours below zero" claim appears in West Midlands News without a source. Agile's half-hourly London prices for 4 Oct do not show an unbroken run like that: they hovered either side of zero from about 01:00 to 15:30, with 18 of 48 half-hours below zero and 27 at or below 0.5p (checked against the Octopus API). The claim may refer to a different (hourly) auction. Treat it as unverified.
• About 365,000 homes lost power during the storm (Wikipedia, above). Worth remembering if a post sounds too cheerful about it.

Wider context that week

• The Ofgem price cap rose 2% to £1,755 on Wed 1 Oct, two days before Agile customers were being paid to use power. (Ofgem)
• NESO published its Winter Outlook on 9 Oct, forecasting a 6.1GW margin, the strongest since 2019/20. (NESO)
• Nuclear was low throughout, at about 2.5GW (Elexon query above).

Confidence

High. Named storm, a Bloomberg report of the negative day-ahead settlement, a Montel analyst quote, and matching generation data.

Gaps

I could not read the Bloomberg original (paywall), only the Energy Connects syndication. No NESO statement about Amy was found. Octopus may have posted about it on social media but I found nothing citable.