Texas has 474GW of data centre requests and no idea how many are real

Texas is reported to have 474 GW of data-center connection requests queued to join the state’s power grid, versus 48 GW in 2023 and more than 5x the state’s record peak demand. The article argues this queue does not translate to hundreds of gigawatts of data centers actually being built, implying a significant power-availability bottleneck. That dynamic increases near-term execution and cost risk for data-center expansion plans in Texas.

Analysis

The key market read is not the headline volume; it is that Texas power access is becoming the binding constraint on AI capex. That shifts value toward assets with existing electrons or the ability to monetize scarcity: merchant generation, peakers, storage, and grid equipment. In that setup, names like NRG, CEG, VST, GEV, ETN, and PWR have a cleaner second-order benefit than the obvious hyperscale adjacency trade, because the bottleneck is interconnection and dispatch, not compute demand.

The queue number itself is low-quality signal until deposits, site control, and financing are visible. Over the next 1-3 months, the real catalyst is regulatory triage in ERCOT/PUCT: tighter screening could purge a large share of speculative requests, which would be bearish for the AI-power narrative but bullish for companies already producing or delivering power. Over 6-18 months, if even a modest share converts, Texas power prices and capacity payments should remain firmer than the market expects, supporting merchant FCF and equipment backlogs.

Contrarian view: the market is still underestimating cancellation risk and self-generation substitution. Hyperscalers are likely to bypass the grid where possible via behind-the-meter gas, PPAs, or non-Texas siting, which means the demand is real but the public-equity beneficiaries are narrower than the hype implies. That argues for trading the infrastructure bottleneck, not the data-center fantasy land bank.

Key Decisions for Investors

• Initiate a small long NRG / long VST basket on any post-news pullback; 3-6 month horizon, as merchant ERCOT exposure is better positioned than broad utility beta if real load eventually converts.
• Buy GEV or ETN on weakness as a 6-12 month infrastructure bottleneck trade; the risk/reward is favorable if equipment backlogs keep extending while actual grid interconnects lag.
• Pair trade: long GEV / short DLR or EQIX for a 3-6 month relative-value expression; thesis is that power scarcity monetizes grid hardware faster than it creates incremental Texas colocation supply.
• Set a watch item, not a trade, for ERCOT/PUCT rule changes and interconnection deposit requirements; if the speculative queue is cut sharply, fade any near-term bullish reaction in power-linked names.
• No direct STT trade here; absent a clearer channel to private-market infrastructure exposure, keep this as an energy/grid watchlist item rather than forcing a position.