This week: three regulators put real money and real megawatts behind demand flexibility — and each one tightened the test it has to pass on the way. On October 29 the CPUC authorized roughly $608 million of two-year bridge funding for the demand-response programs of PG&E, SCE and SDG&E ($110M/yr, $178M/yr and $15M/yr respectively), and in the same stroke exempted those portfolios from the Societal Cost Test and pushed the next DR application deadline out to November 1, 2028. The same day, PJM filed its amended reliability backstop procurement at FERC — a $555/MW-day maximum willingness to pay against a ~6,800 MW shortfall, auction opening March 1, 2027 — whose revisions narrow when demand response and DERs count as qualifying offsets, “rather than treating every behind-the-meter battery as automatically eligible.” And on October 23, FERC largely accepted MISO’s stiffer non-performance penalties for Load Modifying Resources, adding an auction-clearing-price-based financial penalty and splitting partial from complete failure, on the explicit finding that existing penalties deterred nothing because LMRs are called so rarely relative to the capacity revenue they collect.
Read together, the week’s instruments say the same thing from three directions. The demand side is being paid more and trusted less. A $608M multi-year commitment is a commission treating DR as a planned resource worth sustaining; a Societal-Cost-Test exemption is the same commission conceding it cannot yet say what the resource is worth. A $555/MW-day procurement is PJM paying nearly 1.7× its own auction cap for firm capacity; a narrowed offset definition is PJM declining to let distributed assets claim that value on assertion. A tougher penalty regime raises the accreditable value of a well-run program and raises the downside of a badly-run one by exactly as much.
That is the shape the credit fight took after last week’s diagnosis. Last week the argument moved from whether distributed flexibility is dispatchable — PJM answered that nine times in 2026 — to whether it gets counted. This week the counting machinery got built, and in every venue it was built with an eligibility screen attached. The corollary for anyone filing: the binding constraint is no longer the megawatt, it is the evidence that the megawatt is there when called, measured at the meter the regulator looks at. California made that literal — its companion R.21-06-017 decision orders the three IOUs to stand up open, interoperable communications for DER visibility, including to CAISO, and folds a Load Flexibility Scenario into distribution planning as a fourth required modeling case from the 2027–2028 cycle.
⚡ Virtual Power Plants & Demand Flexibility
FERC tightened the penalty regime on demand-side capacity resources — and said plainly why the old one failed. On October 23 FERC largely accepted MISO’s proposal to stiffen non-performance penalties for Load Modifying Resources (LMRs) — the demand-response and behind-the-meter-generation resources that collect capacity payments in exchange for curtailing when MISO calls an emergency. The framework expands participation restrictions, distinguishes partial from complete performance failures, and adds a new financial penalty keyed to auction clearing prices. The penalties take effect June 1, and FERC gave MISO 30 days to fix an inconsistency in how two of the penalty types are calculated.
FERC’s stated rationale is the part to internalize, because it is a structural critique of how DR capacity has been priced. The Commission found MISO’s prior penalties did little to deter fraud and manipulation because LMRs are called so rarely relative to the capacity revenue they earn — the asymmetry that makes overstating a resource rational. The enforcement record behind that finding is substantial: Voltus settled for $10.9M plus $7.1M in disgorged profits; Big River Steel for ~$22M; NIPSCO/Linde for $66.7M; Ketchup Caddy for ~$27M — all over overstated or fabricated DR resources in MISO’s capacity market. FERC rejected protests from Advanced Energy United, American Municipal Power and the Michigan Public Power Agency that the layered penalties were too severe, calling them “not overly severe or unduly punitive.”
The scale makes this a first-order item rather than a market-rule footnote: in MISO’s summer planning auction, demand-response LMRs cleared 9.1 GW and behind-the-meter-generation LMRs 4.3 GW — together 9.4% of all planning resources. A penalty redesign applied to a tenth of the planning base is a repricing of the resource class.
The takeaway is double-edged and both edges are load-bearing. Tighter penalties raise the credibility, and therefore the accreditable capacity value, of a well-run program — this series has argued for a quarter that verification is the gate, and FERC has now put a price on failing it. But Commissioners Chang and Rosner, and industry voices including Rakon Energy’s Rao Konidena, warn that excessive downside risk layered on slow aggregator node-designation timelines is “a recipe for losing demand response,” which could raise adequacy costs by driving good resources out. The practical response is conservative derate and contract-performance assumptions in any value stack, and a close read of who bears the penalty in an aggregation contract — because a penalty that lands on the aggregator and a penalty that lands on the enrolled customer produce very different enrollment curves. (Source: Utility Dive, October 28 — October 29 entry)
DTE launched Michigan’s first residential-battery VPP — and disclosed no capacity figure at all. DTE Energy and sonnen launched what they describe as Michigan’s first residential-battery virtual power plant pilot at Veridian at County Farm, a 170-home, mixed-income community on 14 acres in Ann Arbor where every home is equipped with a sonnen battery. The two-year, MPSC-approved pilot virtually connects nearly 100 of those homes into a pooled, grid-interactive resource responding to grid needs in real time. DTE frames it as a test of how residential VPPs can support high-demand periods, improve reliability and reduce system costs across its 2.3 million Southeast Michigan electric customers, and ties it explicitly to recent MPSC recommendations to maximize existing grid resources and expand distributed-energy participation.
Two things make this worth filing and one makes it uncitable as a capacity result. Worth filing: the site class — a new-construction community where the battery is standard equipment inverts the acquisition problem that defines residential VPP economics, because enrollment is a conversation with a developer at design stage rather than 100 separate retrofit sales. And the regulatory posture — an MPSC that has already recommended expanding distributed-energy participation is a commission receptive to counting aggregated behind-the-meter storage as a planning resource, which is the precondition for the pilot becoming a program. Not citable: the announcement provides no per-home or aggregate kW/MW, no enrollment or compensation terms, and no dispatch-performance commitments. That is the same disclosure failure this series flagged on the Buffalo multifamily battery last week, and the rule has not changed — require kW alongside kWh, and treat a VPP with no megawatt figure as a policy signal rather than a resource. (Source: GlobeNewswire, October 29 — October 30 entry)
Movement — California’s two new DER laws turn out to rest on device-level telemetry, and that is a DERMS requirement, not a policy detail. At an October 23 Advanced Energy United webinar, advocates detailed the implementation mechanics of SB 913 and SB 905, signed September 30 and reported in this series at the signing. The mechanics are where the substance is. SB 913 gives the CPUC a June 2028 deadline to let DER aggregations qualify as resource-adequacy capacity in utility resource plans and state procurements; bars duplicate compensation (NEM and net-billing customers cannot be paid twice for wholesale participation); and — the operative provision — lets a customer enroll multiple DERs behind a single meter with different aggregators managing different assets, enabled by device-level telemetry rather than whole-home metering through the smart meter.
Everyday Electric’s Erik Lyon calls device-level telemetry “the keystone that unlocks the rest of this bill’s opportunities,” and that is exactly right for the wrong-seeming reason. Financially settling two aggregators behind one meter is not a metering preference; it is an architectural requirement that most AMI-only designs cannot satisfy at all, because the settlement quantity is per-asset and the meter measures the net. This series has repeatedly specified multi-protocol, asset-level telemetry ingestion as the hard DERMS requirement rather than the glamorous one. California has now made it a statutory precondition for resource-adequacy qualification, which means the telemetry spec is on the critical path to the revenue, not downstream of it.
SB 905 complements it on the planning side: utilities must collect and publish grid-utilization data, the CPUC is directed toward performance-based metrics, and an investigation opens into “alternative financing” for utility capital — the transparency inputs that make a non-wires-alternative case auditable. The caveat travels with the citation: this is advocate framing of enabling laws, not an order. Real impact depends on pending CPUC and CAISO proceedings, and AEU’s estimate that CAISO’s proposed wholesale “accounting change” could add 2+ GW of DER participation is AEU’s projection, produced inside CAISO’s Demand and Distributed Energy Market Integration initiative. Cite the statutory deadlines; attribute the gigawatts. (Source: Utility Dive, October 28 — October 29 entry; movement on the September 30 signing reported in a prior digest — distinct source, URL and angle)
🔌 DERMS & Grid Integration Technology
California ordered its three large IOUs to build a common DER orchestration framework, open interoperable DER visibility including to CAISO, and a fourth distribution-planning scenario for load flexibility — the adopted version of a decision this series reported in draft. The companion October 29 CPUC decision in R.21-06-017 directs PG&E, SCE and SDG&E to each develop a utility-specific proposal defining grid services and use cases for DERs; to establish open and interoperable communications for DER visibility, including behind-the-meter grid services and visibility to CAISO; and — the integration mechanism — to retain a consultant to lead working-group meetings on a common DER orchestration framework across the three territories. On the planning side it directs the IOUs to hold a workshop presenting a new “Load Flexibility Scenario” and to include it as a fourth modeling scenario in the annual Distribution Planning and Execution Process beginning with the 2027–2028 cycle, with a joint Tier 3 advice letter to follow on the technical basis for the scenario’s inputs and implementation timing.
This series reported the bridge-funding and orchestration decisions at the draft stage from a September 30 Stoel Rives digest; what is new this week is that they were adopted, and what the adopted text actually requires. Three provisions were not visible in the draft reporting and change the specification:
First, “including to CAISO” is a two-sided visibility requirement. A DER-visibility mandate that stops at the distribution operator produces a platform that can see assets but cannot monetize them. Extending the requirement to the wholesale interface is what connects the orchestration layer to the resource-adequacy qualification SB 913 opens in June 2028 — the two California instruments are one mechanism viewed from the regulatory and the market side, and they now have the same architectural implication: asset-level telemetry that is legible to both the distribution utility and the ISO.
Second, the consultant-led working group across three territories is a de facto standards process, and it is the published statement of what a DERMS has to do. This series has argued since May that the central DERMS RFP deliverable is a normalization layer — the capability to present heterogeneous DER controls as one dispatchable resource. A common framework negotiated across PG&E, SCE and SDG&E will be the most detailed public articulation of that requirement in the U.S. Track its working-group materials as procurement specification, not as policy news, and read them beside the CEC interoperability workshop on November 4 (OpenADR 3.1, IEEE 2030.5, OCPI, OCPP), which is the same normalization problem addressed at the protocol layer.
Third, the Tier 3 advice letter is the contestable document. The Load Flexibility Scenario’s value depends entirely on its inputs — what flexibility is assumed available, at what notice, at what cost, with what derate. A scenario populated with conservative assumptions produces a deferral number too small to justify procurement, and the advice letter is where those assumptions get set. Tier 3 means Commission resolution rather than staff disposition, which means it is a venue a demand-side intervenor can actually enter.
The caveat is unchanged from the draft stage and should be stated plainly: this is a framework-and-process order, not a procurement. It commissions consultants, working groups, workshops and a filing; it commits no hardware dollars. Its near-term value is as a requirements template and a planning-integration pathway, and the 2027–2028 cycle is the first point at which load flexibility formally shapes distribution investment. (Source: Stoel Rives Energy Regulatory Update, October 28 — October 30 entry; adopted-decision movement on the draft reported from the September 30 Stoel Rives digest in a prior digest)
🔋 Storage, Distributed Solar & DER Procurement
SEIA and Current Energy Group published a procurement framework for buying DERs as capacity — the most direct statement of this series’ thesis to appear as a document. SEIA’s Distributed Power Project (which absorbed the Coalition for Community Solar Access) and Current Energy Group released the Distributed Capacity Procurement (DCP) Framework: three procurement pathways and 10 principles that regulators and grid operators can implement under existing authority to add capacity and serve load growth faster and cheaper by procuring distributed solar, storage and demand flexibility in place of higher-cost wholesale capacity and utility-owned generation. Two guiding principles are named: use the least restrictive pathway that meets the need, and make third-party- or customer-owned DERs the default, with utility ownership considered only in limited cases.
Its value is as a drafting instrument, and the timing is why it matters. The affordability mechanism is the same avoided-cost logic underneath the ~$66/kW-year DR capacity value and the ~$20.70/MWh efficiency-versus-gas comparison this series carries: competitive procurement of third-party DERs reduces a utility’s reliance on expensive wholesale capacity. It lands in the same week as Wisconsin certifying 1.5 GW of gas under a statute that bars the commission from weighing alternatives at all (below), Florida staff rejecting the state’s first large-load tariff, and PJM opening a $555/MW-day procurement with a narrowed DER offset definition. This is the counter-playbook a planner or intervenor cites to get distributed resources into a capacity decision that is currently structured to exclude them — and “implementable under existing regulatory authority” is the clause that makes it usable in a docket rather than a legislature.
The caveat is real and should travel: this is a framework, not a filing or an order. The coverage names no dollar savings figures and does not detail the three pathways, and adoption depends entirely on individual state commissions picking it up. Get the primary document rather than the trade summary, because the pathway detail is the part that becomes tariff language. (Source: Renewable Energy Magazine, October 26 — October 28 entry)
BloombergNEF raised its U.S. storage forecast ~30% in six months, and names data-center load-shifting as a driver. BNEF’s semiannual outlook now projects 213.9 GW of cumulative U.S. storage by 2030 and 434.4 GW by 2036 — up about 30% from six months earlier — with global capacity reaching ~3,098 GW / 11,106 GWh by 2036, roughly ten times today. Data-center onsite storage rises from 4.9 GW / 7.4 GWh in 2026 (up from 500 MW / 1.4 GWh in 2025) to 36 GW / 66 GWh by 2036, concentrated in ERCOT, PJM and MISO.
The finding that matters for a DERMS or VPP thesis is the use case, not the volume: BNEF expects PJM data centers to use storage mainly for load shifting rather than behind-the-meter gas, under tighter emissions rules. That is a large, fast-growing pool of controllable, dispatchable assets co-located with the loads that are driving the adequacy gap — which bears directly on hosting capacity, on large-load flexibility obligations, and on whether the orchestration layer California just ordered has anything worth orchestrating at the large-load end of the system. The policy hook: the One Big Beautiful Bill Act extended the storage ITC through 2035 while solar sunsets sooner, yielding a pipeline of 83.7 GW / 166.1 GWh of storage paired with 149.2 GW of solar — the colocated fleet most amenable to aggregation — and the solar-to-storage capacity ratio has fallen from 94:1 (2016) to 7:1 (2026), heading to 3:1. BNEF also flags the Google–Voltus PJM VPP, now >1 GW, as the clearest aggregation signal, which is this series’ own centerpiece from last week restated in a forecast. Caveat: a market forecast, not a procurement or a program result. (Source: Utility Dive, October 27 — October 28 entry; the Google/Voltus >1 GW figure is a carried benchmark from last week’s digest, not a new finding)
Missouri approved 545 MW of batteries and 400 MW of solar for Ameren, and the capacity-accreditation fight is the reusable part. The Missouri PSC on October 21 approved Ameren Missouri’s 545 MW of battery storage — Castle Bluff 95 MW (at the retired Meramec coal site), Huck Finn 200 MW, Millcreek 250 MW, all targeting 2028 — plus 400 MW of acquired solar (Ringer 225 MW due 2029, Tom Sawyer 175 MW due 2028), under Case EA-2026-0183, order effective November 20. The batteries are the flexible layer of Ameren’s 2025 preferred plan, which targets 1,000 MW of storage by 2030 alongside new gas (including the February 2026-approved 800-MW Big Hollow plant with a planned 400-MW battery), and the entire need case rests on 2.8 GW of signed large-load electric-service agreements.
The decisive detail is the Office of the Public Counsel’s estimate that the batteries carry only an ~8% default winter capacity accreditation, on which basis it calls the investment “alarmingly cost-prohibitive” and signals it may argue the large-load customers who drove the need should bear the full cost, and may seek prudence disallowances. That is last week’s winter exposure arriving as a number in a live docket. NERC’s Jim Robb argued on October 21 that batteries and solar “struggle to contribute reliably in some markets, especially in winter”; an ~8% winter accreditation on a four-hour battery is what that argument looks like when a state advocate applies it to 545 MW of approved capacity. Any value stack claiming winter capacity credit for four-hour storage now has to answer this figure specifically, and the answer has to be measured — morning-ramp delivery, multi-hour sustained output, state-of-charge at call time, cold-weather derate — because the modeled answer is the one being contested. Conditions attached: Tesla Megapack 3 safety testing, NFPA 855 compliance, per-site hazard-mitigation plans, and cost-variance notice if any project moves ±15%. Caveat: no dollar figures are public (the application redacts cost estimates) and all ratemaking is reserved for a future rate case, so net benefit and cost allocation remain undetermined. (Source: MGRID, October 24 — October 26 entry)
A fifth of Adirondack Park towns have paused battery storage, and New York has 576 MW built against a 6 GW goal. 21 of 101 Adirondack Park towns and villages have enacted 6-month-to-1-year battery-storage moratoriums since January 1, 2026, and three — Fort Ann, Gloversville, Forestport — have banned it outright (North Elba added a one-year moratorium October 13). The mechanism is jurisdictional: standalone batteries answer to local zoning, because New York’s Office of Renewable Energy Siting can override a municipality only for projects over 25 MW co-located with qualifying renewable generation — so merchant storage has no state backstop. S5506 / A8378 would extend state authority to standalone storage; both were still in committee as of May.
The statewide gap is the number any New York resource-adequacy or non-wires-alternative filing has to confront: ~576 MW of storage online as of August 31, 2026 against a 6 GW-by-2030 goal — under 10% — even though NYSERDA estimates the 6 GW target would avoid at least $2 billion in transmission infrastructure costs. That is a T&D-deferral benefit the state has already quantified and is failing to capture for reasons that have nothing to do with economics or interconnection. New York holds 65% of roughly 150 municipal moratoriums tracked across 17 states, and more than 1 GW of projects were stalled by 100+ municipalities as of January testimony. The operational lesson for DER programs: local permitting is now a critical-path item alongside interconnection, a January 1, 2026 fire code (on-site battery-fire expert within 4 hours, qualified personnel dispatchable within 15 minutes of an alarm, developer-funded peer review and first-responder training) adds cost and lead time, and developers are being pushed toward co-location with solar to gain state jurisdiction — which is a siting strategy with real consequences for where distributed capacity ends up on the network. (Source: MGRID, October 25 — October 27 entry)
SEIA credits solar and storage with $3.55 billion of Texas summer wholesale savings — while retail bills rose anyway. SEIA’s October 22 analysis claims solar and storage cut ERCOT summer wholesale energy costs by about $3.55 billion and CAISO’s by about $2.5 billion, comparing June–August 2026 to 2021 in inflation-adjusted dollars. ERCOT average day-ahead prices fell 42%, from $54.97 to $31.77/MWh, and the peak hourly price fell from $110.96 to $57.30/MWh, even as total summer load ran 30.1% above 2021 and ERCOT set a new summer peak record. The mechanism is the one avoided-cost models rely on — solar absorbs midday demand and batteries discharge into the late-afternoon ramp, pushing gas peakers out of the clearing price — backed by Texas battery capacity growing more than 26-fold to 33.1 GWh and Texas solar generation up 421% over five years; California batteries now supply 6–7% of daily energy at 62.0 GWh.
The caveat is the finding, and it is a methodological instruction for every avoided-cost filing. This is a two-summer comparison that does not isolate solar and storage from gas prices, weather or fleet turnover (Wood Mackenzie notes 2024–2026 summers were milder than 2022–2023), and the wholesale savings did not reach retail bills: Texas residential bills still rose 3% and California’s 22%, because distribution charges climbed 23% in Texas and 51% in California, transmission 17% and 12%, and wildfire-management charges rose. Generation’s share of the bill fell 4% in Texas and 10% in California. The lesson for a DSM or DR business case is that the strongest DER value often sits on the energy and capacity legs while the wires legs move the opposite way — so a credible avoided-cost case has to measure wires-side deferral with the same rigor SEIA applies to megawatt-hours, which is exactly what NYSERDA’s $2 billion transmission-avoidance estimate above is, and exactly what the CPUC’s Load Flexibility Scenario is designed to produce. SEIA is an advocacy group and the figure is a wholesale comparison, not a bill impact. (Source: MGRID, October 24 — October 26 entry)
🏗️ Data Centers & Large Load Growth
Arizona is the cleanest current picture of load forecasts being overrun, and of the cost fight that follows. SRP hit a 9,072 MW summer peak on July 24 — about 91% of its ~10,000 MW of installed supply and ~500 MW (≈2%) above the prior level — while APS saw data-center load climb from ~5% to an estimated ~7% of peak. Both utilities ran ~10% reserve margins at peak against 15–16% targets, and both peaks exceeded their 2026 forecasts.
The buildout response carries exactly the stranded-cost exposure that is the hedge-value argument for demand-side resources. APS has sought 8 GW of new capacity for 2031, with 4.5 GW already contracted to data centers and large industrials — atop a system that took ~140 years to reach a 9 GW peak — and is seeking a 14–16% residential rate increase. SRP is planning ~2.4 GW of transmission for a queue of 10 data-center customers totaling ~2 GW that may not all materialize and can take 5–10 years to reach full design capacity. Roughly 95 planned Arizona projects would about triple the state’s data-center energy use.
The cost-allocation fight is the regulatory core and the numbers are unusually specific. SRP recovers via minimum demand charges and requires data centers to pre-pay 60% of system-upgrade costs. APS’s regulated rate case has drawn a consumer-advocate finding of up to $600M in fuel costs recovered from customers as large-load demand grew 27%, a Western Resource Advocates estimate that residential ratepayers overpay ~$136M under APS’s proposed allocation, and an Attorney General (Mayes) call for an audit and possible refunds. The lesson mirrors the Pennsylvania and Missouri proceedings above: unconfirmed, lumpy large-load pipelines create overbuild exposure that firm, verified demand-side capacity is well positioned to hedge — and a ~10% reserve margin against a 15–16% target is a reliability gap that a dispatchable demand-side portfolio can close inside the 2027–2028 window, which no new supply-side resource in this article can. The article itself does not address DR or DERs directly; the relevance is the planning context. (Source: KJZZ, October 26 — October 29 entry)
Wisconsin certified ~1.5 GW of gas while stating that state law barred it from weighing need, cost or alternatives — the sharpest adverse worked example this series has recorded. On October 15 the Public Service Commission of Wisconsin unanimously granted conditional CPCNs for two Invenergy gas plants We Energies intends to acquire: Foundry Ridge (324 MW, Darien, Walworth County, 2028, three 108-MW turbines, Docket 9835-CE-100) and Red Oak Ridge (up to 1,186 MW, Paris, Kenosha County, 2030, five 242-MW turbines, Docket 9837-CE-100) — about 1.5 GW and ~$2.26 billion combined — while stating that state law barred the commission from weighing need, cost or alternative supply for a merchant developer’s project.
That is the structural point, and it is the one every non-wires-alternative business case should carry. Where the certificate stage cannot consider alternatives, demand-side resources cannot be offered as a substitute for the build, and cost is tested only later, in separate acquisition dockets. The need is explicitly data-center-driven — Wisconsin utilities expect a 40% electricity-demand increase by 2032 from loads including a Microsoft campus and a Vantage project leased to OpenAI and Oracle — making this the familiar load-growth-justifies-gas pattern with none of the cost-causation guardrails other states are building: by comparison, Michigan conditioned DTE’s data-center contracts on an 80% minimum monthly billing demand, a test Wisconsin’s framework lacks at the certificate stage. And Wisconsin does not require utilities to file IRPs, so each merchant application is reviewed piecemeal on its own record — PSC Chair Summer Strand called the result “ripe for legislative action.” That is the planning vacuum in which a demand-side portfolio’s avoided-cost case never gets heard, and it is the mirror image of the SEIA DCP Framework above: the framework describes how to let DERs compete; Wisconsin describes a venue where competing is not procedurally possible. Caveats: conditional CPCN with the Final Order and conditions still pending, both plants still need Wisconsin DNR air permits, and price and alternatives will only be litigated in the later acquisition proceedings. UW-Madison and Utility Dive estimate >100 premature deaths and >$1B in particulate damages over 30 years; Calumet County holds two more proposed Invenergy plants totaling 1,900 MW, and We Energies plans a separate 750-MW Wrightstown plant for late 2032. (Source: MGRID, October 25 — October 27 entry)
Entergy Texas brought a 1,215 MW combined-cycle plant online a season late and without a final cost — the comparator a demand-side portfolio is measured against, and it cannot be audited. OCAPS — the 1,215 MW Orange County Advanced Power Station — entered service October 14, past its summer 2026 target, in Entergy’s MISO footprint (not ERCOT), rated to serve more than 230,000 homes with two Mitsubishi M501JAC turbines. Planning context: Entergy Texas serves ~524,000 customers across 27 counties, projects ~13% demand growth from 2024 to 2028, and plans more than 2.4 GW of new capacity by 2028 (including the PUCT-approved Legend and Lone Star plants, >1,200 MW combined, due 2028).
The cost-disclosure gap is the item. Entergy claims more than $1 billion in fuel and operating savings and, on its project page, more than $1.5 billion in net customer benefits — but the only public capital figure is the $1.2 billion 2023 groundbreaking estimate, never updated, so the savings claims cannot be tested against the capital customers will repay. That is the verifiability problem this series invokes in reverse every week: DSM M&V rigor is demanded of a 70 MW DR portfolio and not of a 1,215 MW gas plant whose avoided-cost comparator — the demand response, storage and efficiency the utility could have procured instead — is not discussed at all. MGRID draws the contrast with Black Hills’ 564-MW Cheyenne plant for Google ($1.8B capital, 2027–2029 spend), which published current figures. A conventional generation-completion story with no demand-side content; its value here is as capacity-planning context and as a disclosure-asymmetry data point. (Source: MGRID, October 25 — October 27 entry)
Indiana signed an MOU for up to 6 GW of CANDU reactors with no site, no offtaker and no licensed design — the far end of the lead-time spectrum. On October 19 AtkinsRéalis, the State of Indiana and Hoosier Energy Rural Electric Cooperative signed an MOU to explore up to eight Enhanced CANDU 6 reactors and up to 6.0 GW of baseload. The agreement is discussion-only — no site, no power buyer, no budget — for a pressurized-heavy-water design not yet licensed in the United States (NRC Notice of Intent filed June 24, 2026; 1 of 7 planned topical reports submitted).
Two planning signals, both of which strengthen the near-term demand-side case rather than competing with it. First, the sequence is gas-first — simple- or combined-cycle plants before any nuclear, plus gas-to-nuclear conversion at retiring sites under evaluation — so even this nuclear ambition front-loads fossil capacity. Second, a generation-and-transmission cooperative reaching for unlicensed 2030s baseload is a measure of how thin the near-term firm-capacity bench is: that is the adequacy gap this series’ avoided-cost thesis leans on, and demand response, storage and VPP aggregation are the only resources that can deploy at scale before a reactor campus could exist. MGRID’s own framing is skeptical, contrasting the no-site/no-offtaker/no-design MOU with a same-week AEP / X-energy / Amazon 960 MW Sullivan County project that names site, technology and funder. No DER, storage or demand-side component whatsoever; carried strictly as load-growth-versus-lead-time context. (Source: MGRID, October 25 — October 27 entry)
📋 Regulatory & Policy
The CPUC authorized ~$608 million of two-year DR bridge funding and exempted it from the Societal Cost Test in the same decision. At its October 29 voting meeting the Commission authorized 2028–2029 “bridge” funding for the demand-response programs of the state’s three large investor-owned utilities — $110 million/yr for PG&E, $178 million/yr for SCE and $15 million/yr for SDG&E, roughly $304 million per year or about $608 million across the two bridge years (the agenda-stated total; the named components sum to $303M) — explicitly to maintain program continuity while the Commission considers broader DR reforms. The item carried on the Consent Agenda in R.25-09-004.
The procedural resets matter as much as the money, and two of them are new since the draft reporting:
- The next IOU DR application deadline moves from November 1, 2026 to November 1, 2028, covering program years 2030–2034.
- The Emergency Load Reduction Program pilot is trimmed: limits on pilot sub-groups, reduced minimum dispatch hours, and a cost-effectiveness finding required before continuation past the bridge years.
- The 2028–2029 bridge portfolios are exempted from the Societal Cost Test (per D.24-07-015), while the Commission directs that Societal-Cost-Test capability be built into the DR cost-effectiveness reporting tools before the bridge period ends.
That exemption is the most consequential sentence in the decision for anyone building a DR business case, and it cuts against the resource it funds. A commission that authorizes $608M and simultaneously suspends the broadest cost-effectiveness test for the portfolio it just funded is saying two things at once: DR is worth sustaining, and we do not yet have an agreed method for saying how much it is worth. The practical instruction follows directly and this series has given it before in weaker form: anchor avoided-cost claims in whichever test the commission in front of you actually applies. The ~$66/kW-year capacity benchmark and the ~$20.70/MWh efficiency-versus-gas comparison this series carries are defensible numbers, but they are produced by a societal framing that California has just declined to require of itself — so a filing that leans on societal benefits in a TRC or PAC jurisdiction is arguing a test the record does not contain.
Read the two California decisions as one instrument. The bridge decision buys the programs two years and defers the valuation question; the orchestration decision spends those two years building the visibility and planning machinery that could answer it. Whether the Societal-Cost-Test capability the Commission ordered into the reporting tools actually arrives before November 1, 2028 is now the single most consequential California tracking item in this series — because the 2030–2034 applications will be written against whatever test exists then. (Source: CPUC Public Agenda 3589, Item 22, R.25-09-004, October 29 — October 30 entry; adopted-decision movement with new dollar figures and the Societal-Cost-Test exemption, on the draft reported in a prior digest)
PJM refiled its reliability backstop on the October 29 deadline — at $555/MW-day, and with the DER offset definition narrowed rather than widened. PJM submitted its amended Reliability Backstop Generation Procurement Proposal to FERC on October 29, the date FERC’s suspension order required. The terms:
- Maximum willingness to pay: a MW-weighted average of $555/MW-day, for resources contracted up to 15 years — against the $325/MW-day cap that governs PJM’s own capacity auction.
- Procurement opens March 1, 2027; bid window closes March 28; final procurement target and supply amounts set April 28; results posted May 10.
- Target: the ~6,800 MW (6,831 MW) generating-capacity shortfall from the 2028/2029 Base Residual Auction, to be recalculated with updated information.
- Cost allocation — one of the three amended elements — assigns charges to load-serving entities based on Large Load growth between the actual 2026 summer peak and the 2028/2029 forecast, with electric distribution companies required to report firm Large Load growth, allocated by each zone’s share of that growth. The other two amended elements are LSE credit requirements and LSE exit responsibilities (an exiting LSE remains responsible for procurement charges for the rest of the commitment period).
- The proposal also supports private bilateral contracts between Large Loads and capacity suppliers as a parallel path.
The demand-side provision is the one this series has been chasing for two weeks, and it did not break the way the concessions discussed at PJM’s October 22 special Members Committee suggested it might. The revisions “clarify eligibility requirements for demand response and distributed energy resources as qualifying procurement offsets” — and the clarification is a narrowing: it specifies when DR and DERs can count “rather than treating every behind-the-meter battery as automatically eligible.” No public source states the criteria, whether dormant enrolled DR sites qualify, or whether incremental battery additions at existing sites qualify. Those are now the two highest-value unanswered questions in the filing, because this series flagged in October that a dormant-site inventory and planned battery additions are the cheapest megawatts a portfolio can offer into a procurement of this kind, and the amended text appears to be the instrument that decides whether they count.
The arbitrage is worth stating explicitly, because it is the strongest demand-side argument of the week. PJM is proposing to pay up to $555/MW-day — about $202,575/MW-year — for firm capacity on a 15-year term, which is roughly 1.7× its own auction cap and more than 3× the ~$66/kW-year avoided-capacity benchmark this series uses for DR. A resource that can be verified into that procurement is worth multiples of what the same resource earns in the program it is already enrolled in. That is the clearest price signal distributed flexibility has been shown in this series, and it is attached to an eligibility screen nobody outside the filing can yet read. Action: read the amended tariff sheets directly rather than the coverage, and inventory your dormant DR sites and planned battery additions against the actual offset language. (Source: PJM via PA Environment Digest and EnerKnol, October 29 — in-window targeted gap-fill on a standing open item; resolves the October 29 refiling deadline tracked since October 9)
Two state proceedings put demand-side obligations directly into large-load tariffs — one as the remedy sought, one as the gap alleged.
Pennsylvania: the Office of Consumer Advocate is asking for mandatory interruptible service, and that is a demand-response obligation written into a tariff. The OCA filed an amended complaint at the PUC arguing PPL Electric’s Rate LP-6 does not adequately shield other customers from data-center infrastructure costs, citing ~$178.4 million in proposed network upgrades for a single hyperscale campus of which only $21.5 million would be assigned to the data center and $156.9 million to other ratepayers — an allegation, not a commission finding. The remedy sought is the item: mandatory interruptible service requiring large customers to cut consumption under specified grid conditions, plus stronger minimum-billing, security and construction-contribution safeguards.
The forecast-risk framing is the reusable part. PPL’s advanced-stage pipeline is ~31.8 GW through 2034, yet its 2030 large-load forecast adjustment already fell from 13,343 MW to 12,764 MW between annual submissions, and the cited analyst (Neil Osnato) warns that signed agreements prove commitment but not on-schedule arrival. That is the stranded-cost exposure that makes verified demand-side capacity the lower-risk hedge against overbuilding firm capacity — and the complaint connects cost-causation, project-readiness verification and an explicit demand-side remedy in a single filing, which is an unusually clean template. The caveat the article itself preserves is the one that matters: interruptible service’s real value “depends on contract terms and the customer’s ability to deliver the reduction” — the same M&V rigor this series applies to any DR resource, now applied to a 100-MW-class industrial load. Harvard’s Ari Peskoe notes gaps remain and that interstate wholesale-market costs also push Pennsylvania prices up. (Source: TechTarget, October 27 — October 28 entry)
Missouri: the tariff terms are unusually explicit, and the consumer advocate wants them harder. The Consumers Council of Missouri argues the PSC has “err[ed] on the side of promoting data centers,” as Ameren seeks a $343 million (~10%) base-rate increase for 2027 — about $13/month on the average residential bill — which it attributes to grid upgrades and the Council ties to data-center need. Ameren’s large-load tariff (approved November 2025, effective December 4) applies to 75 MW+ customers and requires a 12-year minimum contract (plus a 5-year option), upfront payment of full hookup and infrastructure costs, a monthly demand charge at 80% of contracted maximum demand, collateral equal to two years of minimum bills, and 36 months’ exit notice with termination fees. The Council wants 15–20-year terms, a 95% demand charge and 10 years of collateral, explicitly to manage stranded-cost risk if an “AI bubble” leaves plants built for loads that never arrive, and opposes Ameren discretion to aggregate loads or grant variances without PSC approval.
The venue is the point. Senate Bill 4 (signed April 2025) overhauled Missouri’s integrated resource planning and mandated the large-load tariffs, and the PSC is now writing contested IRP rules that could force utilities to correct insufficient plans — which is precisely the proceeding in which a demand-side portfolio’s avoided-cost case should be heard, and which the Council argues it still is not. Pipeline context: Google’s New Florence campus at 1.2 GW / $15B (Project Spade) and an AWS ~$10B campus, against a Pettis County data-center moratorium passed in October. Regulators and Ameren maintain existing protections are sufficient; no order issued from this dispute. (Source: ABC17 News, October 26 — October 28 entry; distinct from the October 26 entry’s Missouri PSC approval of Ameren batteries and solar — different proceeding, article and URL)
Florida PSC staff told regulators to deny the first large-load tariff filed under SB 484, and the commission votes November 3. Duke Energy Florida petitioned April 22 for a Large Load Customer Policy covering firm loads ≥50 MW: a 20-year minimum term, minimum demand of 75–85% of contract capacity, a 60% minimum load factor, a $150,000 non-refundable study fee, and termination and security terms (a fee of 36 months of minimum bills plus net book value, with security floored at 10% of that fee when it exceeds $100M) — but set no rate, leaving large loads on existing GSD-1/GSDT-1 schedules until a new class takes effect January 1, 2028. On October 22 PSC staff recommended denying the tariff with leave to refile, on two grounds: the framework “only addresses half of the statute” because whether each large customer bears its full cost of service “remains an open question” under SB 484 (effective July 1, 2026), and the tariff improperly covers only firm-service customers, leaving non-firm loads above 50 MW outside it.
The non-firm exclusion is the demand-side finding. A tariff that reaches only firm service leaves the interruptible class — the loads whose flexibility is the cheapest reliability instrument on the system — outside the cost-allocation framework entirely, which is the inverse of Pennsylvania, where the consumer advocate is trying to pull interruptibility in. Staff also flagged the capacity arithmetic that makes the demand-side case: Duke’s smallest study request is 160 MW against the 55 MW it says it can serve before 2028 (lines absorb up to 200 MW before upgrades, not CIAC-covered), so any new large load likely forces new generation — precisely the capacity need a DR or VPP portfolio could defer. Thirteen study requests exceed 50 MW; the GSD-1/GSDT-1 parity index is 0.91 (~91% cost recovery); 4,400+ public comments were filed. Caveats: staff’s denial rests on absence of evidence of harm rather than a finding of harm (base rates stay frozen through 2027), the commission votes November 3 in Docket 20260064-EI with a 10-business-day refile window, and Tampa Electric filed its own large-load tariff in late September — so this decision sets the state template. (Source: MGRID, October 24 — October 26 entry)
Walmart asked North Carolina to reopen comments on a 150-MW data-center tariff settlement negotiated without it. Duke Energy Carolinas and Progress filed a settlement October 6 in the large-load tariff portion of their two North Carolina rate cases (NCUC E-7 Sub 1329 and E-2 Sub 1380), negotiated with Amazon Data Services, Google, Microsoft, Andale, industrial customer groups, the U.S. Department of Defense and the Public Staff, repurposing the existing Schedule HLF to replace confidential one-off service agreements. Terms define a large-load customer as ≥50 MW at an 80%+ rolling-12-month load factor, or any customer ≥150 MW, with a 75% minimum billing demand and 10-year (<100 MW) / 15-year (≥100 MW) minimum terms carrying early-exit liquidated damages including 25% of remaining minimum-demand charges. On October 12 Walmart — not in the talks, and saying it learned of them only after filing — asked the NCUC for another comment round, not rejection (proposing an October 30 supplemental deadline), objecting to the 150-MW threshold applied regardless of load factor, the rolling-12-month load-factor measure, and a signatory-only commitment to draft a standard service agreement within 30 days of a final order. Supported by CUCA, NCSEA, the NC Justice Center coalition and CCEBA.
The load-factor objection is substantive and it is a demand-side argument in disguise. A ≥150 MW regardless of load factor threshold treats a flat 150-MW data center and a peaky 150-MW industrial load identically, which erases the distinction that demand flexibility monetizes; a rolling 12-month measure smooths exactly the volatility a flexibility obligation would be written against. On cost allocation: dedicated facilities are paid upfront, but other network-upgrade costs are initially spread across all customers and recovered over the contract — SELC’s Nick Jimenez: if a center is never built, “the rest of us are stuck holding the bag.” And the number that frames the whole proceeding: Duke uses an 8-GW Carolinas large-load forecast by 2035 to justify 9.7 GW of new gas — the capacity build a demand-response or VPP alternative would compete against. Caveats: procedural motion, no stated dollar impact; the 75% minimum bill is the main protection in force until a large-load cost-of-service study arrives in Duke’s next rate case (~two years out); and 75+ such tariffs have now been proposed or approved across ~35 states. (Source: MGRID, October 24 — October 26 entry)
Movement — NERC’s computational-load standards cleared the Board on October 26, and the flexibility they do not credit is the gap. NERC’s board adopted the foundational computational-load reliability standards on October 26 — covering modeling, operational information, communications and disturbance monitoring, with the operational standard requiring response to grid instructions subject to safety limits. U.S. approval and phased compliance remain pending, so they are not yet enforceable. This closes the Board-action milestone this series tracked to the date; the FERC filing is expected in early November, ahead of the December 31 deadline previously carried.
The analysis carrying this week’s milestone is itself the better item. A Forbes piece argues that data centers willing to accept binding, verifiable demand ceilings during scarcity should receive faster grid connections, and that earlier interconnection is often worth more to a developer than a cheaper bill — converting curtailable large load into a planning resource the way the curtailment-credit and flexible-interconnection tariffs this series inventories do. It is unusually rigorous on why flexibility claims need scrutiny, and the three objections are directly procurement-relevant:
- Measurement point. Meta reported 30–40% power swings within its Prometheus AI training cluster but provided no trace showing the same swing at the grid interface. A cluster-level swing is not a grid-visible reduction.
- Duration. A 20 MW cut for ten seconds and the same cut for four hours are different products; the latter implies ~80 MWh of usable storage before losses.
- Rebound and repeatability. NERC documented a 2024 event where data centers switched to backup power after a fault and abruptly dropped a large block of load — flexibility that destabilizes on exit.
The contract language that follows is the deliverable: define when the ceiling applies, how response is measured at the grid connection (including repeated calls and post-event rebound), and what fallback protects equipment during an event — because penalties after a failure do not keep the lights on during one. The existence proof is already on the record: Portland General Electric’s 2026 proxy credits GridCARE-identified planning and flexible resources, including batteries, with freeing 80 MW of data-center headroom in 2026, and a 100 MW campus could connect sooner by holding grid imports ≤80 MW under specified conditions. (Source: Forbes, October 28 — October 29 entry; Board-action movement on the NERC CLO standards tracked since September, and the PGE/GridCARE 80-MW figure is a carried benchmark reported in a prior digest)
🏢 Benchmark Hygiene & Carried Context
Capacity-cost pressure: the figures in circulation this week are carried benchmarks, not new findings — and one of them is now contradicted by PJM’s own filing. A ChargedUp! analysis dated October 28 restates the PJM 2028/2029 Base Residual Auction clearing at $325/MW-day — the rule cap, about $118,625/MW-year — producing a $16.4 billion capacity cost on 138,317.8 MW cleared against an uncapped simulation implying roughly $29.7 billion, with procurement 6,831 MW short of the reliability requirement at a 14.4% reserve margin against a 20% target. It adds EIA data showing average commercial electricity revenue of 14.47 ¢/kWh in August 2026, up 4.3% year over year, with forecast growth easing from 4.4% (2026) to 1.5% (2027).
Treat the auction figures as the carried benchmarks they are — this series has reported the $325/MW-day clear, the 6,831 MW shortfall and the ~$29.4–29.7B capacity cost in prior digests, and the Dominion IRP restated the shortfall inside a utility filing last week. The piece also recaps policy this series reported in full: FERC’s October 9 rejection of PJM’s curtail-first Interim Resource Adequacy Service (keeping the 50 MW large-load definition and a registry, effective October 12, compliance ~November 8), Missouri’s 75 MW Ameren large-load tariff, and Colorado’s pending Xcel tariff (Proceeding 26AL-0137E). Its NOI and property-value calculations are the newsletter’s own estimates aimed at a commercial-real-estate audience, not primary utility-sector data.
The one genuinely new thing to do with the $325/MW-day figure is set it beside PJM’s October 29 filing, because the same operator is now proposing to pay $555/MW-day for the shortfall its own capped auction left behind. That spread — $230/MW-day, roughly $84,000/MW-year — is the measured cost of the capacity the market did not clear, and it is the number a demand-side portfolio should be priced against in PJM for the 2028/2029 delivery year. The retail-side figure is the complement: a 4.3% commercial price increase is the rate pressure that makes a commercial customer willing to enroll, and it is where capacity-focused DR rather than kWh-only efficiency accrues value. (Source: ChargedUp!, October 28 — October 30 entry; carried benchmarks and a policy recap, read against the in-window PJM filing)
🔬 EPRI Research Spotlight
No EPRI publication fell inside this week’s window — the sixteenth consecutive week — and a targeted direct check again returned only prior-vintage material. A search for EPRI research dated October 26–30 surfaced no new release. The available material remains earlier work: the 2024 technical update on Grid Services Provided by Distributed Energy Resources, the DERMS Control Architecture for Grid Services in the Distribution and Bulk Power Systems product, the fast-and-flexible demand-response work, the DER-in-resource-planning product, and EPRI’s standing NERC Large Load Actions material, alongside its VIEWS from TOP large-load and resource-adequacy briefing. Report this as cadence, not omission: EPRI’s release rhythm is slower than a weekly digest, and the institute continues to reach the trade and regulatory record through citation and personnel rather than publication.
This week sharpens last week’s observation into something closer to a finding, because the instrument EPRI’s work sits beneath has now been adopted. The NERC computational-load standards cleared the Board on October 26 and go to FERC in early November. EPRI’s large-load material is the substrate beneath those standards — and the standards mandate observability for large loads while crediting none of their flexibility. NERC’s own remedy list, delivered at FERC on October 21, omits demand response and VPPs entirely and treats DERs only as an integration risk. The institution whose research underpins the standards is not the institution deciding what the standards are for, and the adopted text is now the proof rather than the prediction.
The methodological consequence has become measurable this week, and it runs in EPRI’s favor. Three of this week’s items are regulators asking for exactly the thing a research institution produces and the trade record does not: the CPUC ordering Societal-Cost-Test capability built into DR cost-effectiveness reporting tools; the CPUC’s Tier 3 advice letter on the technical basis for Load Flexibility Scenario inputs; and PJM narrowing DER offset eligibility without any public criteria. Each of those is a method question with money attached — what test, what inputs, what counts — and none of them is answered by the news cycle. The Public Counsel’s ~8% winter capacity accreditation for Ameren’s 545 MW of batteries is the same gap stated as a number, and it is the figure that most needs an independent methodological answer before it becomes the default assumption in the next ten dockets.
The measured record meanwhile keeps accumulating outside the research channel. The only large-scale measured large-load event remains the 3,800-MW July 22 trip in which data centers disconnected too early, and the 96-MW Aurora reference installation in Manassas, Virginia still targets late 2026 with no acceleration or slip detected — against which Forbes this week set Meta’s unverified 30–40% Prometheus cluster swing and NERC’s 2024 fault-triggered abrupt load withdrawal. The regulatory calendar has now closed on the research calendar: the standards were adopted October 26 and will be filed in early November, before the measured operational data exists.
Carry forward EPRI’s standing contribution, because it is directly on point for this week’s DTE pilot and Ameren approval: EPRI’s Baum — microgrids and distributed assets must move from emergency-only to everyday dispatch, because “the majority of the time, the power is on.” A two-year, nearly-100-home pilot with no disclosed megawatts and a 545-MW battery fleet credited at ~8% in winter are two versions of the same unmeasured resource.
🚩 Utility-Sector Relevance Flags
⚑ PJM Will Pay $555/MW-Day — and Just Narrowed Which DERs Can Claim It. Read the Tariff Sheets.
Topic: Backstop Procurement Eligibility / Offset Qualification
Relevance: PJM’s October 29 amended filing proposes a MW-weighted-average maximum willingness to pay of $555/MW-day for up to 15-year contracts — about $202,575/MW-year, roughly 1.7× its own $325/MW-day auction cap and more than 3× the ~$66/kW-year avoided-capacity benchmark — against the ~6,800 MW 2028/2029 shortfall, with bidding March 1–28, 2027, targets set April 28 and results May 10. The revisions “clarify eligibility requirements for demand response and distributed energy resources as qualifying procurement offsets,” specifying when they count “rather than treating every behind-the-meter battery as automatically eligible.” No public source states the criteria, nor whether dormant enrolled DR sites or incremental battery additions qualify. Cost allocation now runs to LSEs by Large Load growth between the actual 2026 summer peak and the 2028/2029 forecast, with EDCs reporting firm growth; exiting LSEs stay liable for the commitment period.
Action Signal: Engage — Get the amended tariff sheets, not the coverage, and test your dormant-site inventory and planned battery additions against the actual offset language. This is the highest price signal distributed flexibility has been shown in this series, and it is gated by a screen written in the last ten days. Two things to file on if the criteria are adverse: (a) a behind-the-meter battery that is metered, telemetered and contractually dispatchable is not the same object as one that is merely installed — propose the distinction rather than letting PJM draw it; (b) the cost-allocation basis is large-load growth, so a large load that reduces its own 2028/2029 forecast through verified flexibility reduces its zone’s charge — make that arithmetic explicit to your large customers.
⚑ California Funded DR for Two Years and Suspended the Test That Values It — Anchor Your Avoided Cost to the Test in Front of You
Topic: Cost-Effectiveness Methodology / DR Program Continuity
Relevance: The CPUC’s October 29 decision in R.25-09-004 authorized 2028–2029 bridge funding — PG&E $110M/yr, SCE $178M/yr, SDG&E $15M/yr; ~$304M/yr, ~$608M total — moved the next IOU DR application deadline from November 1, 2026 to November 1, 2028 (program years 2030–2034), trimmed the Emergency Load Reduction Program pilot (fewer sub-groups, reduced minimum dispatch hours, a cost-effectiveness finding required before continuation), and exempted the bridge portfolios from the Societal Cost Test (per D.24-07-015) while directing that Societal-Cost-Test capability be built into the DR cost-effectiveness reporting tools before the bridge ends.
Action Signal: Implement — A $608M authorization paired with a Societal-Cost-Test exemption says the resource is worth sustaining and the method for valuing it is unsettled. Re-check every avoided-cost claim in your portfolio against the test your commission actually applies — TRC, PAC, RIM or PCT, not a societal framing California just declined to require of itself. The ~$66/kW-year and ~$20.70/MWh benchmarks remain defensible, but cite the test that produced them. And treat the two-year deferral as planning runway, not relief: the 2030–2034 applications will be written against whatever test exists in late 2028, so the reporting-tool buildout is the proceeding to participate in now, while the specification is open.
⚑ FERC Raised the Price of DR Non-Performance Across 9.4% of MISO’s Planning Base — Re-Run Your Derates
Topic: DR Performance Penalties / Accreditation Credibility
Relevance: FERC’s October 23 order largely accepted MISO’s stiffer Load Modifying Resource non-performance penalties — an auction-clearing-price-based financial penalty, expanded participation restrictions, and partial-versus-complete failure distinctions — effective June 1, with 30 days given to fix a calculation inconsistency. FERC’s finding: existing penalties deterred little because LMRs are called so rarely relative to the capacity payments they receive. Scale: DR LMRs cleared 9.1 GW and BTM-generation LMRs 4.3 GW — 9.4% of all planning resources. Enforcement backdrop: Voltus $10.9M + $7.1M disgorged; Big River Steel ~$22M; NIPSCO/Linde $66.7M; Ketchup Caddy ~$27M. Commissioners Chang and Rosner warn excessive risk could push good resources out and raise adequacy costs.
Action Signal: Implement — Verification just acquired a price, and it cuts both ways. Re-run program economics with conservative derate and contract-performance assumptions, and read who bears the penalty in every aggregation contract you hold — a penalty landing on the aggregator and one landing on the enrolled customer produce different enrollment curves and different attrition. Then use the order affirmatively: a program that can document delivered-versus-committed MW per event now has a regulator-endorsed reason to claim higher accreditation than one that cannot. The enforcement record is the argument for your M&V budget.
⚑ An ~8% Winter Capacity Accreditation Is Now in a Live Docket Against 545 MW of Approved Batteries
Topic: Seasonal Capacity Credit / Storage Accreditation
Relevance: The Missouri PSC approved Ameren’s 545 MW of storage (Castle Bluff 95 MW, Huck Finn 200 MW, Millcreek 250 MW, all due 2028) and 400 MW of solar under EA-2026-0183 on October 21, effective November 20 — with the need case resting on 2.8 GW of signed large-load agreements and a 1,000-MW-by-2030 storage target. The Office of the Public Counsel estimates an ~8% default winter capacity accreditation, calls the investment “alarmingly cost-prohibitive,” and signals it may argue large-load customers should bear full cost and may seek prudence disallowances. No dollar figures are public; all ratemaking is reserved for a future rate case.
Action Signal: Implement — This is last week’s winter exposure arriving as a number, and it will be cited against four-hour storage in other states. Produce the measured counter-record before it becomes the default assumption: morning-ramp delivery, multi-hour sustained output, state-of-charge at call time, and cold-weather derate, per event, per site. A modeled winter accreditation will not survive contact with an advocate quoting 8%. Note the conditions as a procurement template too — Megapack 3 safety testing, NFPA 855 compliance, per-site hazard-mitigation plans and ±15% cost-variance notice are becoming standard storage approval terms worth pre-building into a filing.
⚑ Device-Level Telemetry Is Now a Statutory Precondition for Resource-Adequacy Qualification in California
Topic: DERMS Architecture / Settlement Granularity
Relevance: SB 913 gives the CPUC a June 2028 deadline to let DER aggregations qualify as resource-adequacy capacity, bars duplicate compensation for NEM/net-billing customers, and permits multiple DERs behind a single meter with different aggregators managing different assets, enabled by device-level telemetry rather than whole-home smart-meter metering. The companion October 29 R.21-06-017 decision orders PG&E, SCE and SDG&E to define DER grid services and use cases, stand up open interoperable communications for DER visibility including to CAISO, and retain a consultant to lead working groups on a common DER orchestration framework across all three territories. SB 905 requires published grid-utilization data and directs the CPUC toward performance-based metrics. AEU estimates CAISO’s proposed wholesale accounting change could add 2+ GW of DER participation (AEU’s projection).
Action Signal: Implement — Settling two aggregators behind one meter is an architectural requirement, not a metering preference: the settlement quantity is per-asset and the meter measures the net. Most AMI-only designs cannot satisfy it at all. Put asset-level telemetry, multi-protocol ingestion and per-asset settlement on the critical path in your DERMS specification, and require the vendor to demonstrate two aggregators settling independently behind one service point rather than assert it. Then track the three-IOU working-group materials as procurement specification — it will be the most detailed public articulation of the normalization layer in the U.S. — and read it beside the November 4 CEC interoperability workshop (OpenADR 3.1, IEEE 2030.5, OCPI, OCPP).
⚑ The Load Flexibility Scenario Is Adopted; the Tier 3 Advice Letter Decides What It Is Worth
Topic: Distribution Planning Integration / Deferral Value
Relevance: The October 29 R.21-06-017 decision folds a new “Load Flexibility Scenario” into the annual Distribution Planning and Execution Process as a fourth modeling scenario beginning with the 2027–2028 cycle, preceded by an IOU workshop and followed by a joint Tier 3 advice letter on the technical basis for the scenario’s inputs and implementation timing.
Action Signal: Engage — The scenario was the win; the inputs are the value, and the advice letter is where they get set. A scenario populated with conservative assumptions — low available flexibility, long notice requirements, high cost, heavy derate — produces a deferral number too small to justify procurement, which converts a structural victory into a rounding error. Tier 3 means Commission resolution rather than staff disposition, so it is a venue an intervenor can actually enter. File on the inputs specifically: assumed enrolled capability, notification window, dispatch duration, seasonal derate and the cost basis. And propose the same mechanism in your own jurisdiction — a required load-flexibility scenario in distribution planning is a smaller, more winnable ask than a procurement target, and it generates the deferral numbers that justify procurement later.
⚑ Three Large-Load Tariff Proceedings Diverged on Whether Flexibility Is Inside the Tariff at All — Screen Yours
Topic: Large-Load Tariff Design / Flexibility Obligations
Relevance: Pennsylvania — the OCA’s amended complaint seeks mandatory interruptible service in PPL’s Rate LP-6, alleging $178.4M of network upgrades for one hyperscale campus split $21.5M to the data center and $156.9M to other ratepayers, against a ~31.8 GW advanced-stage pipeline whose 2030 forecast adjustment already fell 13,343 → 12,764 MW. Florida — PSC staff on October 22 recommended denying Duke Energy Florida’s ≥50 MW firm-load tariff with leave to refile, partly because it improperly excludes non-firm loads above 50 MW; vote November 3, Docket 20260064-EI; Duke’s smallest study request is 160 MW against 55 MW it can serve before 2028. North Carolina — Walmart objects to a ≥150 MW regardless of load factor threshold and a rolling-12-month load-factor measure in Duke’s October 6 settlement (E-7 Sub 1329 / E-2 Sub 1380, 75% minimum billing demand, 10/15-year terms), filed without it. 75+ large-load tariffs now proposed or approved across ~35 states.
Action Signal: Engage — Screen every large-load tariff in your territory on one question: does flexibility appear as an obligation, an option, or not at all? Pennsylvania is trying to write it in; Florida’s draft left the interruptible class out of the framework entirely; North Carolina’s thresholds erase the load-shape distinction that flexibility monetizes. A load-factor-blind threshold and a rolling-12-month measure are the two design choices that most reliably exclude demand flexibility from a large-load tariff — flag both wherever they appear. And carry the Florida capacity arithmetic as the standing argument: when the smallest study request is three times what the utility can serve before 2028, any new large load forces new generation — which is the deferral opportunity, stated by staff.
⚑ Prove Flexibility at the Grid Interface, for a Stated Duration, With a Defined Rebound — or It Is Not a Resource
Topic: Large-Load Flexibility M&V / Contract Terms
Relevance: NERC’s board adopted the foundational computational-load reliability standards on October 26 (modeling, operational information, communications, disturbance monitoring; the operational standard requires response to grid instructions subject to safety limits) — U.S. approval and phased compliance still pending. Against that, the verification objections: Meta reported 30–40% power swings inside its Prometheus cluster with no trace showing the same swing at the grid interface; a four-hour 20 MW reduction implies ~80 MWh of usable storage before losses; and NERC documented a 2024 fault-triggered event where data centers abruptly dropped a large load block onto backup power. Existence proof: Portland General Electric’s 2026 proxy credits GridCARE-identified planning and flexible resources with freeing 80 MW of data-center headroom, and a 100 MW campus could connect sooner by holding imports ≤80 MW (carried benchmark).
Action Signal: Implement — Write the three missing terms into every large-load flexibility agreement: when the ceiling applies, how response is measured at the grid connection (including repeated calls and post-event rebound), and what fallback protects equipment during an event. Penalties after a failure do not keep the lights on during one. The trade worth offering is the one the analysis identifies: binding, verifiable demand ceilings in exchange for faster interconnection — earlier access is often worth more to a developer than a cheaper bill, which is the same instrument as SPP CHILL, ERCOT WLPUN/PCLR and the rejected PJM IRAS under a bilateral name. Cluster-level telemetry is not grid-interface telemetry; require the latter.
⚑ SEIA Published the Counter-Playbook the Same Week Wisconsin Showed Why It Is Needed
Topic: DER-as-Capacity Procurement / Venue Selection
Relevance: SEIA’s Distributed Power Project and Current Energy Group released the Distributed Capacity Procurement (DCP) Framework — three procurement pathways and 10 principles implementable under existing regulatory authority, guided by “use the least restrictive pathway that meets the need” and third-party- or customer-owned DERs as the default with utility ownership the limited exception. In the same window, the PSCW granted conditional CPCNs for 1,186 MW and 324 MW of Invenergy gas (~1.5 GW, ~$2.26B) on October 15 while stating state law barred it from weighing need, cost or alternative supply, in a state that does not require IRPs and against a projected +40% demand by 2032; Chair Summer Strand called it “ripe for legislative action.” No dollar savings figures or pathway detail appear in the DCP coverage.
Action Signal: Engage — Get the primary DCP document; the pathway detail is what becomes tariff language, and “implementable under existing authority” is the clause that makes it usable in a docket rather than a legislature. Then do the venue analysis first, because Wisconsin is the proof that the best framework loses to the wrong procedural posture: identify, per jurisdiction, whether the proceeding that approves capacity is permitted to consider alternatives at all. Where it is not, the demand-side case belongs in the acquisition or rate docket, not the certificate docket — and where there is no IRP requirement, the structural ask is the IRP, not the megawatt.
⚑ New York Has Quantified a $2 Billion Transmission Avoidance and Is Capturing Under 10% of It — for Zoning Reasons
Topic: Local Permitting as Critical Path / T&D Deferral
Relevance: 21 of 101 Adirondack Park towns and villages have enacted 6-month-to-1-year battery-storage moratoriums since January 1, 2026, three have banned it outright, and North Elba added a one-year moratorium October 13 — because standalone storage answers to local zoning: ORES can override a municipality only for projects over 25 MW co-located with qualifying renewables. Result: ~576 MW online as of August 31, 2026 against a 6 GW-by-2030 goal (<10%), while NYSERDA estimates the 6 GW target would avoid at least $2 billion in transmission costs. New York holds 65% of ~150 municipal moratoriums tracked across 17 states; >1 GW stalled by 100+ municipalities. S5506 / A8378 would extend state authority to standalone storage (in committee as of May). A January 1, 2026 fire code requires an on-site battery-fire expert within 4 hours and qualified personnel within 15 minutes of an alarm.
Action Signal: Watch — Move local permitting onto the critical path in every distributed-storage deployment schedule, alongside interconnection — and note the strategic distortion the jurisdiction rule creates: developers are being pushed toward co-location with solar purely to gain state jurisdiction, which changes where distributed capacity lands on the network for reasons unrelated to grid value. Two uses for the NYSERDA figure: it is a state agency’s own quantification of storage’s T&D-deferral benefit, which is a citable precedent for the wires-side leg of a value stack; and the 576 MW versus 6 GW gap is the strongest available argument that town-by-town approval cannot deliver a state target — the case for the pending bills, and for the fire-code and first-responder-training funding that buys local consent.
⚑ Wholesale DER Savings Do Not Reach Bills — Measure the Wires Leg or Lose the Argument
Topic: Value-Stack Completeness / Avoided T&D
Relevance: SEIA’s October 22 analysis claims solar and storage cut ERCOT summer wholesale costs ~$3.55B and CAISO’s ~$2.5B (summer 2026 vs. 2021, inflation-adjusted): ERCOT average day-ahead prices −42% ($54.97 → $31.77/MWh), peak hourly $110.96 → $57.30/MWh, with load +30.1%, Texas battery capacity up >26× to 33.1 GWh and Texas solar generation +421% over five years. Yet residential bills rose 3% in Texas and 22% in California, because distribution charges climbed 23% and 51%, transmission 17% and 12%, and wildfire charges rose; generation’s bill share fell 4% and 10%. Wood Mackenzie notes 2024–2026 summers were milder than 2022–2023, and the study does not isolate solar and storage from gas prices, weather or fleet turnover.
Action Signal: Implement — The strongest DER value sits on the energy and capacity legs while the wires legs move the opposite way, so an avoided-cost case that measures only megawatt-hours is arguing the leg that is already winning and ignoring the one that is losing. Measure wires-side deferral — the specific distribution and transmission investment a DR or DER portfolio defers, by circuit and by year — with the rigor SEIA applies to energy. NYSERDA’s $2 billion transmission-avoidance estimate and the CPUC’s Load Flexibility Scenario are both instruments for producing exactly that number; use them. And when citing SEIA: advocacy source, two-summer comparison, wholesale not retail — state all three.
⚑ Three Load Forecasts Were Overrun in One State and the Refund Fight Followed — This Is the Hedge Argument
Topic: Forecast Risk / Stranded-Cost Exposure
Relevance: SRP peaked at 9,072 MW on July 24 — ~91% of ~10,000 MW of supply, ~500 MW (≈2%) above the prior level — and APS’s data-center load rose from ~5% to ~7% of peak; both ran ~10% reserve margins against 15–16% targets and both exceeded their 2026 forecasts. APS seeks 8 GW of new capacity by 2031 (4.5 GW already contracted to large loads) on a system that took ~140 years to reach a 9 GW peak, plus a 14–16% residential rate increase; SRP is planning ~2.4 GW of transmission for a ~2 GW, 10-customer queue that may not fully materialize and can take 5–10 years to ramp. Cost fight: a consumer-advocate study alleges up to $600M of fuel costs recovered from customers as large-load demand grew 27%; WRA estimates ~$136M of residential overpayment under APS’s proposed allocation; AG Mayes seeks an audit and possible refunds. SRP requires data centers to pre-pay 60% of system-upgrade costs. ~95 planned Arizona projects would roughly triple state data-center energy use.
Action Signal: Watch — A ~10% reserve margin against a 15–16% target is a near-term adequacy gap that a dispatchable demand-side portfolio can close inside 2027–2028, and no supply-side resource in this record can. Lead with that timing. Then carry Arizona as the clearest current instance of the standing argument: unconfirmed, lumpy large-load pipelines create overbuild exposure, and firm, verified demand-side capacity is the cheapest hedge against it — a transmission plan sized for 2 GW of queue that may not arrive is the risk, and SRP’s 60% pre-payment requirement is the cost-causation instrument other utilities should be asked about by name.
⚑ Storage Supply Is Being Revised Up 30% and Data Centers Are Buying It to Shift Load — Not to Island
Topic: Asset Availability / Orchestration Opportunity
Relevance: BNEF raised its U.S. storage forecast ~30% in six months: 213.9 GW cumulative by 2030, 434.4 GW by 2036 (global ~3,098 GW / 11,106 GWh by 2036). Data-center onsite storage: 4.9 GW / 7.4 GWh in 2026, up from 500 MW / 1.4 GWh in 2025, to 36 GW / 66 GWh by 2036, concentrated in ERCOT, PJM and MISO — and BNEF expects PJM data centers to use it mainly for load shifting rather than behind-the-meter gas, under tighter emissions rules. OBBBA extended the storage ITC through 2035 (solar sunsets sooner), yielding a pipeline of 83.7 GW / 166.1 GWh of storage paired with 149.2 GW of solar; the solar-to-storage ratio fell 94:1 (2016) → 7:1 (2026), heading to 3:1. Google–Voltus PJM VPP now >1 GW (carried benchmark).
Action Signal: Watch — The orchestration opportunity is moving to the large-load side of the meter. If PJM data centers deploy 36 GW of storage for load shifting rather than islanding, those assets are price-responsive by design and co-located with the loads driving the adequacy gap — which is a far better aggregation target than a retrofit residential fleet, and it is the resource the CPUC’s orchestration framework and PJM’s offset definition will both end up governing. Track it as hosting-capacity and flexibility-obligation input, and ask in large-load dockets who dispatches the on-site storage and whether its flexibility is credited — the question this series has had open on ~2 GWh in the NiSource and Entergy packages since last week, now with a 36-GW denominator. Forecast, not procurement.
📌 Sources
October 26, 2026 Entry
– MGRID — Missouri Approves Ameren’s 545 MW of Batteries and 400 MW of Solar, But Who Pays Waits for a Rate Case (October 24, 2026) (first appearance; the week’s storage-accreditation anchor)
– MGRID — SEIA Credits Solar and Storage With $3.55 Billion of Texas Summer Savings, While Retail Bills Rose Anyway (October 24, 2026) (first appearance)
– MGRID — Florida PSC Staff Tells Regulators to Deny Duke Energy’s 50 MW Large Load Tariff, the First Filed Under SB 484 (October 24, 2026) (first appearance)
– MGRID — Walmart Asks North Carolina to Reopen Comments on Duke’s 150 MW Data Center Tariff Deal, Negotiated Without It (October 24, 2026) (first appearance)
October 27, 2026 Entry
– MGRID — A Fifth of Adirondack Park Towns Have Paused Battery Storage, and New York Has 576 MW Built Against a 6 GW Goal (October 25, 2026) (first appearance)
– MGRID — Wisconsin Approves 1,186 MW and 324 MW Invenergy Gas Plants for We Energies, With Need and Cost Off the Table by Law (October 25, 2026) (first appearance)
– MGRID — Entergy Texas Brings Its 1,215 MW Orange County Gas Plant Online, Past the Summer 2026 Target and Without a Final Cost (October 25, 2026) (first appearance; capacity-planning and disclosure-asymmetry context, no DER content)
– MGRID — Indiana Signs an MOU for Up to Eight CANDU Reactors and 6 GW, With No Site, No Offtaker and No NRC-Licensed Design (October 25, 2026) (first appearance; lead-time context only, no DER content)
October 28, 2026 Entry
– Renewable Energy Magazine — New Whitepaper From SEIA Shows How Grid Operators and Energy Regulators Can Use Solar and Storage to Reduce Energy Costs (October 26, 2026) (first appearance; the SEIA/Current Energy Group Distributed Capacity Procurement Framework — distinct SEIA document from the October 26 entry’s Texas summer-savings analysis)
– TechTarget — Pennsylvania Complaint Challenges PPL’s Ratepayer Protections (October 27, 2026) (first appearance; distinct from the PA PUC’s October 1 tentative order on large-computational-load cost allocation, Docket M-2026-3064961, carried in prior digests)
– ABC17 News (KMIZ) — Consumer Advocates Warn Missouri Lacks Data Center Cost Safeguards, But Regulators Disagree (October 26, 2026) (first appearance; distinct proceeding from the October 26 entry’s Missouri PSC approval of Ameren batteries and solar)
– Utility Dive — BNEF Ups US Energy Storage Deployment Forecast by 30% (October 27, 2026) (first appearance; the Google–Voltus >1 GW PJM VPP figure inside it is a carried benchmark from last week’s digest)
October 29, 2026 Entry
– Utility Dive — FERC Approves Tougher MISO Penalties for Nonperforming Demand-Side Resources (October 28, 2026) (first appearance; the week’s DR-regulatory anchor)
– Utility Dive — California Utilities, Aggregators and Customers to Benefit From New DER Laws: Advocates (October 28, 2026) (movement — the September 30 signing of SB 905 and SB 913 was reported in a prior digest; this is the Advanced Energy United implementation analysis, a different source, URL and angle)
– Forbes (Güney Yıldız) — Data Center Grid Flexibility Must Be Proven, Not Promised (October 28, 2026) (first appearance for the analysis; carries the NERC October 26 Board adoption of the computational-load standards, which is movement on an item tracked since September, and the PGE/GridCARE 80-MW figure, a carried benchmark)
– KJZZ (Taylor Griffith) — Arizona Utilities Scramble to Keep Up as Data Center Boom Drives Rapid Expansion of the Power Grid (October 26, 2026) (first appearance)
October 30, 2026 Entry
– California Public Utilities Commission — Public Agenda 3589, Item 22, R.25-09-004: Two-Year Bridge Funding for Demand Response Programs and Pilots (Voting Meeting, October 29, 2026) (adopted-decision movement — the bridge funding and the November 1, 2028 application deadline were reported at the draft stage from a September 30 Stoel Rives digest in a prior digest; new here are the per-utility dollar figures, the ELRP pilot trims and the Societal-Cost-Test exemption)
– Stoel Rives LLP — Energy Regulatory Update, October 28, 2026 (CPUC R.21-06-017, DER Orchestration and Incorporation of Load Flexibility Scenarios Into Distribution Planning) (adopted-decision movement — the orchestration framework and the Load Flexibility Scenario were reported at the draft stage in a prior digest; new here are the CAISO visibility requirement, the three-territory consultant-led working group and the joint Tier 3 advice letter)
– GlobeNewswire — DTE Energy and sonnen Launch Michigan’s First Residential Battery Virtual Power Plant Pilot to Explore New Ways to Lower Energy Costs (October 29, 2026) (first appearance)
– ChargedUp! — The True Cost of Power: A $325 Capacity Price and a 4.3% Commercial Price Increase Put New Pressure on NOI (October 28, 2026) (carried benchmarks and policy recap only — the $325/MW-day clear, the 6,831-MW shortfall, the ~$29.7B uncapped simulation and FERC’s October 9 IRAS rejection were all reported in prior digests; the new EIA retail figure and the $555-versus-$325 spread are the usable parts, and the NOI and property-value calculations are the newsletter’s own estimates)
In-Window Targeted Gap-Fills
– PA Environment Digest — PJM Interconnection Files Revised Reliability Backstop Generation Procurement Proposal With FERC; Looks to Hold Next Auction March 2027 With Price Cap of $555/MW-Day (October 29, 2026) (first appearance; in-window gap-fill resolving the October 29 refiling deadline this series has tracked since October 9 — the price cap, auction schedule, cost-allocation basis and LSE credit/exit revisions)
– EnerKnol — PJM Files Revised Backstop Procurement Plan to Address Capacity Shortfall (October 29, 2026) (first appearance; in-window gap-fill supplying the demand-side provision — the revisions “clarify eligibility requirements for demand response and distributed energy resources as qualifying procurement offsets”)
– Microgrid Media — PJM Backstop Procurement: March 2027 Bidding Proposed (October 2026) (first appearance; corroborates the October 29 filing date and the $555/MW-day cap, and states the offset clarification specifies when DR and DERs count “rather than treating every behind-the-meter battery as automatically eligible”)
Method: All five weekday entries — Monday October 26 through Friday October 30 — are present in the research log; no days are missing. The window is strict Monday–Friday and no item from a prior week is reported as new. Items are reported on the day they entered the log, with their original publication dates shown; several carry pre-week publication dates inside the daily entries’ trailing-48-hour windows (the four MGRID pieces of October 24 and October 25, the KJZZ Arizona report and the ABC17 Missouri report of October 26, the Renewable Energy Magazine SEIA item of October 26, the TechTarget PPL item and the Utility Dive BNEF item of October 27, and the Utility Dive, Forbes, Stoel Rives and ChargedUp! items of October 28). Cross-week deduplication, which is where this week’s hard cases were: both CPUC decisions are adopted-decision MOVEMENT, not new findings — R.25-09-004 bridge funding and R.21-06-017 DER orchestration with its Load Flexibility Scenario were both reported in the October 2 digest at the draft stage from a September 30 Stoel Rives digest, so they appear here for what the adopted text added (the per-utility dollar figures totaling ~$608M, the ELRP pilot trims, the Societal-Cost-Test exemption, the CAISO visibility requirement, the three-territory consultant-led working group and the joint Tier 3 advice letter); the Advanced Energy United analysis of SB 905 and SB 913 is movement on the September 30 signing reported in the October 9 digest, carried for the implementation mechanics rather than the signing; NERC’s October 26 Board adoption of the computational-load standards is movement on a track reported through its October 15 final ballot last week; the PJM 2028/2029 $325/MW-day clear, the 6,831-MW shortfall, the ~$29.7B uncapped simulation, FERC’s October 9 IRAS and BYOC rejection, the Missouri 75 MW Ameren large-load tariff, Colorado’s pending Xcel tariff in Proceeding 26AL-0137E, the Google–Voltus >1 GW PJM VPP and the Portland General Electric / GridCARE 80-MW headroom figure are all carried benchmarks cited in one line each rather than re-reported; and the Missouri items are two distinct proceedings — the October 26 entry’s PSC approval of Ameren’s batteries and solar in EA-2026-0183 and the October 28 entry’s Consumers Council challenge to Ameren’s large-load tariff and SB 4 IRP framework — which share a utility and a state but not a docket, an article or a URL. Zero source URLs overlap last week’s Sources section. Within-log deduplication: no development appeared on more than one day this week, so no items required consolidation; the DERMS and ADMS searches in all five daily entries returned only DistribuTECH and IEEE PES T&D 2026 session listings and 2025–2036 market-sizing reports, none newly published, so there is no standalone DERMS vendor or deployment item this week and the DERMS section is built on the CPUC orchestration mandate and the California telemetry requirement instead. Targeted search was used only to verify or fill clear gaps. It resolved the single highest-value open item on the board: PJM filed its amended Reliability Backstop Generation Procurement Proposal at FERC on October 29, the deadline this series has tracked since October 9, at a MW-weighted-average maximum willingness to pay of $555/MW-day for up to fifteen-year contracts, with bidding March 1–28, 2027, and with revisions that clarify — and narrow — when demand response and distributed energy resources count as qualifying procurement offsets. What it did not resolve, and what now carries forward as the sharpest unanswered question in the series: the actual offset eligibility criteria, and specifically whether dormant enrolled DR sites and incremental battery additions at existing sites qualify, neither of which any public source states. A search for NERC’s energy assurance standard — last week’s highest-value new tracking item — returned only the 2026–2028 Reliability Standards Development Plan and no dedicated project, docket or development timeline, so whether demand-side resources are in scope remains unestablished after one run. The Pennsylvania Bulletin publication date that starts the comment clock on the PA PUC’s October 1 tentative order in Docket M-2026-3064961 was not pursued this week and remains unverified after three prior runs; pacodeandbulletin.gov refuses programmatic retrieval and no secondary source states it, so it needs a manual check or a call to the PUC Secretary’s Bureau. PJM’s deployed demand-response megawatts for the 2026 shoulder-season events remain unpublished. EPRI direct check: no publication dated inside the window — the sixteenth consecutive week — with EPRI appearing instead as the substrate beneath the computational-load standards the NERC board adopted on October 26.