Who appears in supply, who receives accepted Balancing Mechanism actions, and how much volume was accepted from the units NESO labels as batteries.
London settlement days 1-31 October 2026 (31 days; 25 October has 50 settlement periods because of the clock change). The archive selection was fixed on 4 November; later source revisions are not substituted.
Accepted-action and accepted-volume figures come from Balancing Mechanism records (BOALF and Elexon's settlement accepted volume) only. Balancing done outside the Balancing Mechanism, such as frequency response and other contracted services, is not visible in them.
Share of the signed recorded generation sum, including net interconnector and pumped-storage values.
GB settlement dates: 1-31 October.
UTC interval: 30 September 23:00 to 1 November 00:00 UTC.
Share of accepted actions by the fuel label NESO gives each BM unit in its monthly files (latest month labelled, August to October 2026). Units NESO's files list without a usable label are shown as Unlabelled, units absent from those files as Not in NESO's files. Each accepted action is counted once.
London settlement days 1-31 October.
An action is counted when its instruction begins inside 30 September 23:00 to 1 November 00:00 UTC.
Supply is not an action-count denominator. These distributions cannot be subtracted to measure who contributes more. Bar lengths are scaled separately within each distribution; printed percentages name the values.
197 BM units carry NESO's BATTERY label in its files for August to October 2026; 194 have accepted actions on these settlement days. The label is NESO's, mapped to Elexon identifiers through the BM unit register; it is not a census of GB battery assets or proof of physical identity.
On these settlement days, BM units with no NESO fuel label, the Unlabelled and Not in NESO's files bars together, received 0.6% of accepted actions (1,999 of 351,175). These actions stay in the denominator; whether those units are batteries is not determined.
Group accepted actions / all accepted actions
74.0%259,827 / 351,175
An accepted action is BMU + acceptance number, counted when its instruction begins inside these settlement days; the file of the UTC day before is read for that. A count of instructions, not energy.
Group accepted bid MWh / every BM unit's accepted bid MWh
35.9%213,085.1 / 594,188.1 MWh
Elexon's indicative settlement accepted bid volume (DISPTAV, Original), against each unit's final physical notification, as a magnitude. Not metered delivery.
Group accepted offer MWh / every BM unit's accepted offer MWh
8.3%175,832.9 / 2,130,613.4 MWh
The same settlement record for accepted offers. Bids and offers are each over their own total and are not netted.
For comparison, the name-rule list frozen on 2026-06-12 (69 high-confidence units, not every battery) received 27.9% of accepted actions, 15.2% of accepted downward volume and 3.3% of accepted upward volume on the same settlement days.
Three separate denominators and two units, instructions and MWh. No addition, subtraction, contribution ranking or combined score follows from placing these percentages together. An instruction count is not energy, and accepted volume is not metered delivery.
Each value uses its own day's denominator. A monthly share is a ratio of monthly totals, not the average of these percentages. The instruction share and the volume shares can move apart on the same day; neither explains the other.
Solid line: share of accepted actions. Dashed line: share of accepted downward volume (bids). The upward-volume share is in the table.
| Day | Periods | All accepted actions | Actions | Bids | Offers |
|---|---|---|---|---|---|
| Oct 01 | 48 | 8,615 | 76.1% | 22.6% | 3.4% |
| Oct 02 | 48 | 10,664 | 66.0% | 41.4% | 4.1% |
| Oct 03 | 48 | 7,990 | 74.9% | 51.5% | 11.6% |
| Oct 04 | 48 | 10,360 | 60.6% | 22.8% | 7.3% |
| Oct 05 | 48 | 11,574 | 64.9% | 10.0% | 7.2% |
| Oct 06 | 48 | 8,176 | 83.4% | 42.2% | 4.4% |
| Oct 07 | 48 | 8,432 | 85.0% | 37.0% | 3.3% |
| Oct 08 | 48 | 11,125 | 88.8% | 52.4% | 10.2% |
| Oct 09 | 48 | 14,097 | 73.9% | 11.4% | 21.1% |
| Oct 10 | 48 | 12,623 | 70.2% | 40.2% | 10.3% |
| Oct 11 | 48 | 8,255 | 88.5% | 38.1% | 5.2% |
| Oct 12 | 48 | 8,618 | 60.2% | 44.4% | 5.3% |
| Oct 13 | 48 | 13,286 | 60.0% | 67.1% | 9.1% |
| Oct 14 | 48 | 13,085 | 50.1% | 84.4% | 5.2% |
| Oct 15 | 48 | 10,276 | 66.1% | 35.3% | 6.4% |
| Oct 16 | 48 | 9,644 | 87.2% | 48.6% | 3.3% |
| Oct 17 | 48 | 8,337 | 86.3% | 29.2% | 7.6% |
| Oct 18 | 48 | 8,817 | 87.6% | 22.4% | 3.8% |
| Oct 19 | 48 | 10,837 | 82.9% | 31.7% | 8.9% |
| Oct 20 | 48 | 12,968 | 68.5% | 15.1% | 11.6% |
| Oct 21 | 48 | 13,887 | 84.8% | 49.9% | 14.4% |
| Oct 22 | 48 | 9,358 | 75.2% | 26.6% | 11.0% |
| Oct 23 | 48 | 14,988 | 86.3% | 57.4% | 1.2% |
| Oct 24 | 48 | 19,706 | 80.0% | 50.0% | 13.7% |
| Oct 25 | 50 | 11,540 | 64.7% | 16.0% | 8.5% |
| Oct 26 | 48 | 12,153 | 65.0% | 61.7% | 3.9% |
| Oct 27 | 48 | 20,551 | 69.6% | 38.2% | 13.2% |
| Oct 28 | 48 | 13,637 | 61.5% | 50.1% | 7.1% |
| Oct 29 | 48 | 8,739 | 81.7% | 36.6% | 5.2% |
| Oct 30 | 48 | 10,132 | 73.5% | 15.1% | 9.0% |
| Oct 31 | 48 | 8,705 | 93.6% | 78.8% | 5.4% |
BM-unit identifiers with accepted actions
Concentration-equivalent identifier count
The second value is 1 / sum(squared accepted-action shares). It describes the concentration of this count distribution; it is not a count of identified units, independent resources or available replacements.
For each registered outage window, including the one carrying a cancellation notice, accepted actions across all BM units represented in the selected records, counted by acceptance time, are set beside the same UK clock times on each of the seven days before. Each window is counted from its announced start to the completion time its completion notice states, or to its announced end where no completion notice was issued. A day on which those times overlap another registered window, or span a different length because of a clock change, is left out; 4 to 7 days remain per window.
| Outage window (UK clock time) and route | Accepted actions | Remaining days: their accepted actions, lowest to highest | Compared with the remaining days |
|---|---|---|---|
| 2026-10-01 09:00–15:05 WA API | 1,912 | 7: 1,544 to 4,092 | Within the range of the remaining days |
| 2026-10-08 09:00–12:15 WA API | 2,052 | 6: 719 to 1,874 | More than on every remaining day |
| 2026-10-15 10:10–15:10 (carries a cancellation notice) Balancing Mechanism IT systems | 1,297 | 6: 1,487 to 3,950 | Fewer than on every remaining day |
| 2026-10-15 22:00 to 2026-10-16 02:00 WA API | 770 | 7: 524 to 2,373 | Within the range of the remaining days |
| 2026-10-19 12:00–16:00 WA API | 2,223 | 6: 936 to 3,825 | Within the range of the remaining days |
| 2026-10-22 10:10–13:20 Balancing Mechanism IT systems | 265 | 5: 779 to 2,460 | Fewer than on every remaining day |
| 2026-10-23 08:00–11:19 WA API | 3,718 | 6: 971 to 1,851 | More than on every remaining day |
| 2026-10-26 11:00–14:00 WA API | 1,075 | 4: 1,186 to 3,957 | Fewer than on every remaining day |
With n remaining days, an ordinary day would itself be the lowest or the highest of the n + 1 days in 2 cases out of n + 1. Over these 8 windows that comes to about 2.4 windows by chance alone; 5 of them hold fewer or more accepted actions than every remaining day. This treats the days as interchangeable, which differences between weekdays break, so it is a reference only.
The WA API windows concern one instruction interface, while the counts cover all BM units represented in the records. Telephone instructions, delayed recording and ordinary day-to-day variation have not been distinguished, so these comparisons do not establish the cause of the differences.
939,361 BOALF rows in the 33 files the accepted-action counts read, 115,744 in the 5 earlier files read only for the outage reference days, and 29,800 FUELHH rows were reopened from the fixed source set.
The fixed set holds a BOALF file for every UTC day these settlement days touch and the day before, the earlier BOALF files the outage reference days need, Elexon's settlement accepted volume for every settlement day in both directions, and a FUELHH file for every settlement date. This does not certify omitted upstream categories, embedded generation or whole-event completeness.
Accepted actions and accepted volume share the London settlement days; FUELHH uses the same settlement dates. An accepted action is placed by the start of its first accepted segment; an action that began before 30 September 23:00 UTC is not counted. The outage counts in section 05 place an accepted action by its acceptance time instead, and read the seven days before each window, back to 24 September.
Accepted volume is Elexon's indicative settlement run, which later runs can revise; the fixed set does not substitute them. It is accepted volume against each unit's final physical notification, not metered delivery.
Asset health, wear, cyber exposure, independent reserve paths, dispatch motives and system resilience. Frequency, metering, contracts and disclosures are not silently used as explanations for these figures.