Summary
- The two Batch Zero classes are not a firmness ladder: the one named for further study is the one that competes for remaining transmission capacity, and it carries about two thirds of the gigawatts.
- The split was issued under a Commission-excused deadline, later than the substitute date the operator had given, on numbers its own 5 November update still treats as unfinished.
Two gates into the same study
The clearest description of what the two labels do comes from a company with projects inside the process, in a release republished on 8 September 2026 by Nasdaq: “ERCOT's Batch Zero process groups qualifying projects into a single, system-wide study that allocates available transmission capacity. Base Load projects are modeled in the study and are not subject to further capacity allocation, while Studied Load projects are subject to capacity allocation as part of the study” (Galaxy, republished by Nasdaq).
Read that twice, because the plain English points the wrong way. The label that sounds like a floor — base — is the one that has already been settled inside the study. The label that sounds like an unfinished homework exercise — studied — is the one still exposed to an allocation of whatever transmission capacity is left after everybody above it has been modelled. One gate stands open onto the yard. The other still has somebody counting what fits behind it.
That is not a distinction between serious and unserious projects. It is a distinction between two positions in a queue.
The numbers put the bigger projects behind the harder gate
Gibson Dunn, in a client note updated on this dateline, states what the two classes actually hold: “ERCOT issued conditional classifications in early September, placing 204 projects (66.4 GW) in base load and 158 projects (127.9 GW) in studied load, subject to the outcome of the verification process. Most developers were expected to receive a second, short round of project-specific questions with a five-business-day response window” (Gibson Dunn).
The firmer's other points — Batch Zero as a one-time transitional study for Large Loads of 75 MW and above under PGRR145, a report targeted for 10 December 2026, suspended approvals to energise new Large Load data centres and crypto facilities, a 21 September directive to the Texas Commission on Environmental Quality on data-centre permits — are the firm's summary of the procedure, and I label them as the firm's account rather than as the operator's words.
The two passages together give an arithmetic nobody has printed as a relationship. Mine, and labelled as mine: 66.4 GW divided by 204 is about 0.33 GW per project in the class that skips allocation; 127.9 GW divided by 158 is about 0.81 GW per project in the class that competes for it. The average studied-load project is roughly two and a half times the average base-load project. The two classes together are 194.3 GW across 362 projects, and the class that competes for capacity holds 65.8 per cent of those gigawatts on 43.6 per cent of the projects.
So the heavier study burden sits on the heavier projects. Anything a reader infers about which class is more likely to be built has to survive that fact.
The clock moved twice before anybody was classified
Batch Zero had a date. ERCOT's own presentation to the Commission on 20 August 2026 states the request in the operator's words: “To excuse ERCOT from the August 7, 2026 classification deadline in Planning Guide § 9.3.1(2)(a) and instead allow ERCOT to conditionally classify large loads by August 31, 2026” (ERCOT presentation, 20 August 2026). The same deck gives the reason — time “to conduct the eligibility verification before final classification of large loads included in the Batch Zero interconnection process” — and names the deliverable, a Batch Zero Eligibility Verification Report.
The Commission granted the relief. ERCOT's market notice of 21 August says so in one sentence: “The PUCT issued an order granting ERCOT's requests on August 20, 2026” (M-A080326-02). Utility Dive reported the same day that “The PUCT approved good cause exceptions for ERCOT to an Aug. 7, 2026, Batch Zero classification deadline that Abbott's moratorium interrupted” (Utility Dive), and the Commission's 20 August agenda lists the item: “Project No. 59142 – Review of ERCOT's Interconnection Processes for Large Loads. (Discussion and possible action, including discussion and possible action on ERCOT request for a good cause exception)” (PUCT agenda). An agenda schedules; the operator's notice reports the disposition. I keep them apart.
Then the substitute date was missed too. On 31 August 2026 ERCOT published a notice recalling its own plan: it “indicated that it planned to notify Interconnecting Distribution Service Providers and Interconnecting Transmission Service Providers of each Large Load's conditional classification by August 31, 2026, as authorized by the Public Utility Commission of Texas's (PUCT) August 20, 2026 order granting ERCOT's good cause exception” (M-A080326-03). The notice is headed delayed issuance. The classifications themselves are described as issued in early September.
The calendar is not decoration here. Seven August to thirty-one August is 24 days. The order to the substitute date is 11. The substitute date to the scheduled report is 101. From the operator's 5 November update to that report is 35. Every one of those intervals is my subtraction, not a published measure, and none of them is a deadline anybody has published a consequence for.
What is still conditional, and who says so
The numbers above are conditional classifications. ERCOT's update of 5 November 2026 says plainly that the review was not finished: “ERCOT must complete its review this month to file a report with the PUCT on December 10, 2026” (ERCOT RFI update). The same update reports first-round responses from 257 developers, about 5,140 questions, with a second round going out.
Utility Dive's August reporting had already framed the filing as an intention rather than an event, quoting ERCOT General Counsel Chad Seely: “Our goal is to head toward a December 10 filing,” including a comprehensive report on the verification and audit process delivered to the Commission a week before its open meeting that month.
As of this dateline, 10 December 2026 has not arrived. Nothing in the record I read shows the report filed, and nothing shows which projects, if any, will leave either class when it is.
Four documents, four different hands
The reason to care about grades of authority is that four instruments are in play and only one of them is a Commission rule.
The adopted rule is 16 TAC §25.194. Per a legal commentary published on 18 September 2026: “On September 18, the Public Utility Commission of Texas (PUCT) adopted the new 16 Texas Administrative Code (TAC) §25.194. The rule puts into effect Public Utility Regulatory Act (PURA) §37.0561” (JD Supra). Note the order of events: the classifications described above were issued in early September, the exception was granted in August, and the rule was adopted on 18 September. The rule does not produce the split described in this article. It arrives after it.
The transitional study itself is a planning guide revision request, PGRR145 — named that way in the Gibson Dunn note and named in the title of ERCOT's own slide deck. The clock that governed the classification is planning guide text, §9.3.1(2)(a). And the instrument that moved the clock was a Commission order in Project No. 59142.
That order is the one document here I could not read. Its address is in the record (Order Granting Good Cause Exceptions, Project No. 59142), and retrieval on this dateline returned HTTP 402 Payment Required. So the order is described in this article and never quoted: its date, its existence and its subject rest on the operator's notices, the Commission's agenda and independent reporting. I do not know its operative conditions, its expiry, or what it requires anybody to file. What can be said is structural — a planning-guide deadline was excused by a docketed order, and neither the guide nor the request is a rulemaking.
What this record does not carry
The operator's planning guide text was not read for this article. So nothing here states §9.3.1's criteria, its definitions, or whether it restates §25.194. The PGRR145 text was not read either, which is why the mechanics above are attributed to Galaxy and Gibson Dunn rather than described as the operator's wording.
No security, deposit, interconnection-agreement or phased-energisation milestone wording from any document was read, so this article does not answer which instrument sets those obligations. No document I read imposes a consumption-reduction duty on either class; that is a statement about the documents read and not a finding about the rule or the guide.
The 75 MW threshold appears here only as the Gibson Dunn note's summarising prose, not as a quoted operator or Commission sentence. The capitalisation of the two labels is not harmonised between the two quoted passages — lower case in one, capitalised in the other — and I have printed each as its publisher prints it rather than pick a winner. No roster of individual project classifications was located, so the class-level aggregates are the only per-class quantities in this article.
What a reader building in Texas can actually use
Two things, and neither is a definition.
First, the label a project receives is not a statement about how real the project is. It is a statement about where the project sits in a single system-wide study and whether it still has to compete for capacity inside it. On the printed figures, the class that competes holds more megawatts and bigger average projects.
Second, the sorting that produced those labels was itself issued late, under an exception whose text is not in public reach, on numbers its author still calls conditional. If a financing model treats a classification as settled, it is modelling a distribution that the operator has not published.
ERCOT will file, or will not file, a report on 10 December. The Commission meets on 17 December. Somewhere between those two dates, the words base and studied either acquire criteria the public can check, or they keep the shape they have had since early September: two gates, one borrow of time, and 194.3 GW standing behind them.
Member Briefing
Read the full trend briefing
Members read the full briefing with the source notes behind every finding.
Strategic Circle
A daily briefing in your language, built from changes we have verified.
Join for USD 20/monthCancel anytime.
Members also get the full research library, a daily briefing on the networks they follow and the private forum.






