Sangha's 'Base Load' Data Center Win: Why the Classification Matters — and Why There's No Stock to Chase

A developer announcing it booked power for an AI data center is usually background noise these days. Every week brings a press release claiming another campus is "powered up" or "shovel-ready." So it's worth pausing on Sangha Renewables' announcement this week, because the status it secured — a "Base Load" classification in ERCOT's new "Batch Zero" interconnection process — is the one a data center developer actually wants, and the one most people misread when they see the headline.

Sangha is a small, private West Texas developer that began life as a Bitcoin miner. In December it switched on a 19.9 MW behind-the-meter data center sitting on a 150 MW solar farm in Ector County, drawing power directly from the solar source rather than the grid. That site, now branded "Project Genesis," is the campus ERCOT just classified: a total of 110.4 MW, co-located with the utility-scale solar, aimed at AI, cloud, inference, and high-performance computing workloads.

Why "Base Load" is the prize

To see what the classification means, you have to understand the grid context. Texas is drowning in interconnection requests — more than 438,000 MW of large-load applications, roughly 90% of it data centers, in a queue that dwarfs the transmission space available to serve it. In a single quarter this year, 198 GW of large load applied. Grid access, not land and not compute, has become the binding constraint on the entire AI buildout.

In June the Public Utility Commission approved a new batched process for loads of 75 MW or more, and "Batch Zero" is its first cohort. ERCOT sorts those projects into tiers. The critical one is Base Load: subject to further reliability evaluation or MW allocation. That matters enormously. "Studied/allocated" loads, by contrast, face reliability assessment, possible curtailment obligations, and transmission-upgrade costs — and ERCOT this month reclassified a number of original base-load submissions down into that more onerous studied tier.

The practical consequence: a base-load classification keeps the project in the cohort that is modeled in the study without further reliability evaluation or MW allocation, rather than the studied/allocated tier that faces additional grid studies and possible transmission-upgrade costs. It is an interconnection and eligibility status, not a guarantee of firm power delivery. In a market with no capacity market and emergency curtailment rules, that classification still doesn't exempt a load from curtailment obligations — energization remains contingent on ERCOT's verification being completed. Sangha's designation, if it holds, is the good version of this story—so far as it goes.

What "conditional" really means

Here is where the headline overpromises. That classification is conditional, and the timing math tells you how early this is.

First, ERCOT's entire classification process is paused and under audit. In early August, Governor Abbott ordered comprehensive verification of every data center advancing through interconnection, and ERCOT halted Batch Zero classification activity in response, saying it would not authorize large loads to energize until the verification is done. The screening ERCOT then ran sorted roughly 250 to 300 projects — about 200 GW of prospective demand — into "conditional inclusion," "studied-load," and "excluded." Conditional inclusion is not final interconnection approval, and it establishes no energization date. Projects that fail verification, or don't fix deficiencies in their dynamic-model data, can be dropped from Batch Zero entirely. Sangha itself says it expects to navigate the ongoing audit "expeditiously" — which is exactly what a company that needs the result to close a deal would say.

Second, the classification covers a campus with very little live power in it. Today the project has access to roughly — the amount already energized. The remaining 90.5 MW of the 110.4 MW total is an additional allocation that isn't expected until . In data-center terms that is not a campus building out now; it is a right to build years from now, pending both the audit and the transmission buildout.

The real point of the announcement

That's the frame that puts the whole press release in its proper place. Compare Sangha's 110.4 MW with the roughly 200 GW, or 200,000 MW, sitting in this first Batch Zero screen — it is about one-twentieth of one percent of the queue. This is not a hyperscale campus in the making. It is a small private developer holding a scarce, coveted interconnection right and putting it on the market: Marathon Capital has been hired as exclusive financial advisor for a "strategic process" on the opportunity. In plain terms, the company is shopping the powered land and the base-load entitlement to someone bigger who actually needs firm power and has the balance sheet to wait for 2028.

None of that is a criticism of the project. If anything, it's the market working — the value created by holding a scarce interconnection right is real, and the people closest to the backlog are the ones monetizing it. But it does change what a retail investor should take from the headline.

What the investor should actually take away

First, Sangha itself is not an investable stock — it's private, and this announcement is a project-financing and disposition update, not an earnings event. So there is no position here to chase, and no ticker to buy on the news.

What the episode does offer is a cleaner way to read every "data center got power in Texas" headline you'll see for the next year. Ask three things. Was the status Base Load or the weaker studied/allocated tier? Is the classification final or conditional on a verification audit that has no completion date yet? And how much is live power now versus a promised allocation years down the road? Sangha's release scores exactly one out of three today — real base-load tier, but conditional, with 90.5 of the campus's 110.4 MW not expected until 2028.

The fundamental point holds: firm interconnection is the scarce asset at the center of the AI buildout, and its value is real and rising. But announcements like this confirm the scarcity — they don't confirm that a campus is operational, and they don't put a return in a retail investor's pocket. The disciplined read is to treat the classification as a useful signal about how valuable grid access has become, and to weigh any actual investment against live, firm megawatts and a balance sheet that can survive the wait — not against the press release.