$395.44/MWh separated the top and bottom of the ERCOT nodal stack in the 11:50 UTC interval, 05:50 CST on a Wednesday. JUNCTIONRN printed $280.73/MWh while GALORN1 and MAVCRKWALL sat tied at $-114.71/MWh. HB_NORTH cleared at $71.28/MWh ten minutes later, so this is a wires story, not a system one.
The shape of the stack says congestion rather than scarcity. System load was 45,876 MW at 11:55 UTC against net load of 41,833 MW, with natural gas the main generation source; nothing in that pairing justifies a $280 node. The high side is deep: CFLATSUNIT at $245.48, RATRN1 at $230.16, LGDLANGFORD at $213.04, FERMIALL at $173.22. The low side is just as deep: RNRTS1 at $-80.26, RADNSLRALL at $-43.20, PRIDRN at $-26.99. The live feed lists SARFRAN1, BLESSING1382 and 6635_G as the top constraints at 12:01 UTC but reports no shadow price for any of them. The angle's West Texas export read fits the node mix, but the source data does not place GALORN1 or MAVCRKWALL geographically, does not give current wind or solar output behind the negatives, and does not show whether the spread survived past 11:50 UTC. Treat the export thesis as the working hypothesis, not the confirmed cause.
Supply-side noise is background, not driver. ERCOT logged at least 2,149 unplanned outage events over the seven days to today. The sample from 2026-10-31 shows HKSNSLR fully out on Maintenance Level 2 for 158 MW, TOPAZ forced out for 45 MW with the record appearing twice, and wind units PCNORTH and AQUILLA down 54 MW and 14 MW. None of that moves a 45,876 MW system; it does thin the margin on whichever side of the constraint those units sit, and the feed does not say which.
Three checks, all against the 15:00 UTC SCED interval, which is 09:00 CST. First, if GALORN1 still prints at or below $-114.71/MWh at 15:00 UTC, the negative pin has outlasted the morning ramp and the constraint is structural for the day. Second, if JUNCTIONRN still clears above $280.73/MWh while HBNORTH holds at or below $71.28/MWh, the load pocket is being priced by the wire, not by the hub, and the export read gains weight. Third, if the 15:00 UTC snapshot shows a non-null shadow price on BLESSING1382 or SARFRAN1, the mechanism is identified; if all three constraints still read n/a with the max spread above $395.44/MWh, the data feed, not the grid, is the thing to question.
> A $395 spread with a $71 hub is a wire pricing itself, and until a shadow price shows up the only evidence for the West Texas story is the node list.
Not investment advice. For informational purposes only.
Yesterday's tape: PJM WESTERN HUB lmppeak >= 90.04 — verified (observed 371.48); PJM WESTERN HUB lmppeak >= 72.14 — verified (observed 371.48); PJM PJM-RTO lmp_peak <= 64.95 — not triggered (observed 264.67).
Grid Alpha — daily briefs · live dashboards, all nine markets · public call record