I-SEM Daily Briefing — 30 May 2026

Price Profile #

Std dev €17.5/MWh · Median €125.14/MWh · Periods above €150: 7 of 48 (15%)

A boundary-distorted Saturday. The headline peak (€172.98 at 23:00) is Friday’s late-evening scarcity tail spilling into Saturday’s file — the same SEM day boundary problem that has appeared on May 12, May 16, and other high-volatility transitions.

The “real” Saturday begins at 00:00 with €161 and falls cleanly: €146 at 00:30, €140 at 01:00, into a smooth overnight floor of €121–140 through 06:30. Wind held 24–44% across overnight — modest. The morning ramp barely existed: €124 at 07:00 → €126 at 07:30 → fell back to €123 by 08:00. No morning peak — Saturday demand softened the ramp out of existence.

Midday belly held €105–117 from 10:30 to 16:00 — six and a half hours of shallow trough, with the floor at €104.61 at 14:30. Evening built a contained climb — €106 at 15:00 → €152 at 21:30. The actual Saturday evening peak topped at €152.58, not the €173 the headline shows. Late evening faded to €141.

Strip the boundary periods and Saturday’s real range is €121 to €153 — barely €32 of intra-day variation. The flattest underlying day of the working week with no morning peak, a contained evening peak, and a long shallow trough.

7 of 48 periods cleared above €150 (4 of those are the Friday spillover); 0 above €200.

Price vs Wind #

Mean wind: 37.2% · Range: 23.9%–46.5%

Steady, mid-range wind across the entire day. Wind held 24–47% with no major shortfalls. Overnight (after the period 3 wind glitch) ran 24–34%, morning rose to 42–44%, midday held 39–47%, and evening drifted down to 29–34%. The narrow band (23-point range) is the third-narrowest wind range of the 27-day run.

With wind reliably present and demand at a weekend-low 3,567 MW, the merit order found gas-marginal at low load across virtually all 48 half-hours. No periods needed peakers; no periods saw surplus; clearing prices stayed in the €105–155 band that gas at moderate load produces.

This is the genuinely-calm Saturday archetype — what May 23 might have looked like with slightly less wind. May 23 broke through to €78 at the trough because 31% wind into 3,511 MW demand produced surplus at midday. May 30’s 39% wind into 3,567 MW demand stayed gas-marginal — the surplus condition never quite materialised. The 6-percentage-point wind increase from May 23 to May 30, combined with steadier wind timing, lifted the trough by €26 (€78 → €105). A small change in wind shape produces large changes in trough depth at the surplus threshold.

Week in Context #

Three consecutive days of wind-shape variation: May 28 V-out-of-evening, May 29 V-into-late-evening, May 30 wind-flat-all-day. The dispatch outcomes track the shape: €77, €155, €66.

Price Duration Curve #

Periods above €150: 7 (15% of day) · Above €200: 0 (0% of day)

Today’s PDC has the cleanest “low-and-narrow” shape of the run — even more contained than May 13’s calm Wednesday. Forty-one of 48 periods cleared below €150; the bottom 12 averaged €110; the top 7 are heavily boundary-distorted.

Strip the boundary periods and the underlying Saturday PDC is essentially flat: 46 periods within €120–155. This is what the wind-rich weekend with no morning peak looks like in its purest form. A consumer’s dream, a flexibility operator’s nightmare.

Price Spread #

Captured spread: €57/MWh · Charge: €105/MWh (13:00) · Discharge: €163/MWh (23:00)

BESS Dispatch Signal #

Simulated 1MW/2MWh battery, one optimal DAM cycle. Gross before network charges and capacity costs.

€66 gross. The optimiser picked the Friday spillover as the discharge window — fourth time this has happened (May 12, May 16, plus implicit on other days).

The mechanics: the four consecutive highest-priced half-hours within Saturday’s SEM-DA file are periods 1–4 (23:00–00:30, averaging €162.50). These are Friday’s evening peak crossing the SEM day boundary. The real Saturday-internal best discharge would have been periods 44–47 (20:30–22:00 at €149.6 avg) — €13 lower, which would have produced gross around €40.

The €66 is mechanically correct and physically misallocated. A rolling 24h optimisation would have captured those four periods as part of Friday’s discharge window. Saturday’s “real” day, stripped of the spillover, would have been the run’s third-weakest BESS day at around €40 gross — only May 21 (€38) and possibly May 9 in earlier weeks would have been worse.

The 27-day cumulative now stands at €2,757, mean €102/day. Saturday added €66 but roughly half of that is misallocated revenue that physically belongs to Friday. The cumulative across the series is right; the day-by-day attribution across high-volatility transitions needs a caveat.

Commentary #

A boundary-distorted Saturday on a genuinely calm underlying day. The headline peak (€172.98 at 23:00) is Friday’s late-evening scarcity spilling into Saturday’s file — the SEM day boundary problem returning. Strip those periods and Saturday’s real intra-day range is just €121–153, the flattest underlying day of the working week. Wind held 24–47% with no shortfall, demand was at a weekend-low 3,567 MW, and the merit order ran gas-marginal at moderate load across virtually all 48 half-hours.

The boundary effect is the second major distortion on this Saturday. The off-peak window (22:00–07:00) caught Friday’s late-evening scarcity tail at the start and Saturday’s own elevated overnight at the end — producing a technically elevated overnight average on a day where the real cheap hours were midday. On a genuinely calm Saturday, the cheap hours are in the middle of the day, not overnight; a rolling 24h window would read this correctly where the SEM-DA window cannot.

For storage, €66 gross — but the optimiser’s discharge window is the Friday spillover (periods 1–4, averaging €163), not the Saturday evening peak (€150). The 27-day cumulative now stands at €2,757, mean €102/day. The day-by-day attribution problem is now an established feature of the data: roughly 5–10% of the running BESS total sits on the wrong day because of how the SEM-DA day is defined. The cumulative is approximately right; the daily numbers across high-volatility transitions need a caveat. Worth fixing in a future iteration with a 24h rolling-window optimisation.