bankerbros

@SuperBitcoinBro Great chart! What’s the typical duration of those past orange box? Thanks

Broadening wedges on software $IGV historically precede massive runs in $BTC. But what I want to draw your attention to is the period of chop in orange, before the run. Even after the bottom is in, your patience and conviction are still tested. You will wonder why you are holding bitcoin instead of stocks. Then when it exits the chop, you'll wonder why you bothered to buy anything besides bitcoin.

Vitol just overtook Shell in US physical gas trading. A commodity trading house is now handling more American gas than one of the world's largest energy companies.

When commodities are free - American gas edition: Natural gas at the Waha hub (West Texas) has traded at negative prices for >75 consecutive business days. Has any other commodity traded in any significant pricing hub **consistently** at sub-zero levels for such a long period?

Dependence on LNG and on gas for dispatchable power 2026 conditions have highlighted that countries that rely on LNG imports and have a large share of gas capacity will be particularly challenged. Get the full picture 👇

After listening to Illia’s interview on Medici Level Up, I think CT is framing $NEAR completely wrong. People keep trying to classify it as: - an AI infra play - a chain abstraction play - an intents/payment layer - an agent framework But Illia’s actual thesis is bigger: “AI in the front end, blockchain in the back end.” NEAR is building a vertically integrated stack for the agentic economy: - private/verifiable AI inference - IronClaw agents - AI-to-AI contracts + escrow - intents across 35+ chains - NEAR as the settlement + trust layer And the key line: “NEAR token is effectively looped into everything.” 1. Inference 2. Payments 3. Intents 4. Security 5. Storage Not every new token for every new project, just single $NEAR for all of this. Most L1s are optimizing one layer. NEAR is trying to own the whole AI → commerce → settlement loop. Still feels mispriced.

Confidential inference means you have end-to-end encryption with nobody else getting access to your prompts. But this only works on open weight models (for now). If you do still want ot use closed weight models - you can use it via @near_ai and it will remove all sensitive info All via single API to access anonymized versions of OpenAI/Anthropic/Google models with single payment in both firat and crypto.

STEPHON CASTLE POSTER ON HARTENSTEIN 🤯🤯🤯 DUNK OF THE PLAYOFFS. (via @NBA)

Ronald Acuña Jr. is back and having fun 😆

Sal Stewart MASHES a second-deck homer 😮 It's his 4th hit of the game!

🇺🇸US gas demand is about to grow faster than any other period in history. Almost all of it is LNG and power. 2025–2031 forecast: 27 BCF/d of new demand. The largest single period demand increase on record. The 2 drivers: → LNG feedgas: the dominant contributor. New export terminals coming online through 2030 need gas. → Power: AI data centers, electrification, are pulling more gas into the grid than at any point in the past decade. The next wave of US gas demand is export-driven and electricity-driven. Both are structural, not cyclical. Link in the comments, to see my stocks 👇 Source: RBN Energy, GasCon 2026, EIA / Oxford Institute for Energy Studies

BREAKING: US total crude inventories fell last week ~17.8 million barrels (that's commercial and SPR stocks combined). On that basis, it's the largest weekly fall since data is available starting in 1982.

JAMES WOOD INSIDE-THE-PARK GRAND SLAM!

Stuck in my head this morning from Jeff Currie: "Financial markets price expectations, commodity markets price the physical reality of today." Oil deficit now ➡️ oil shortages soon ➡️expectations shift ➡️rerating of back end of the oil curve ➡️significant room to run. Patience.

Meta is building a 7.5 GW AI campus in rural Louisiana. To power it, they're constructing 10 new gas plants. The Hyperion AI campus in Richland Parish will require up to 7.5 GW of power. 5 GW of that is pure compute load. To deliver it, Entergy Louisiana is building approximately 10 combined-cycle gas turbines in two waves: → Wave 1 (approved 2025): 3 CCGTs, 2.3 GW → Wave 2 (agreed March 2026): 7 more CCGTs, >5.2 GW CCS and hydrogen-ready That's the equivalent of multiple large gas fleets dropped into a single corner of the US South that was, until recently, farmland. Meta signs 15-year power contracts, underwriting the full revenue requirement of the plants. Entergy owns and operates the steel in the ground, earns a regulated return. The broader Gulf Coast gas grid Haynesville, Transco, NGPL sells the molecules. Meta takes demand risk. Entergy takes construction risk. Gas producers get a locked-in long-dated sink. Why gas and not nuclear or renewables? CCGTs can be built in 3-5 years. AI inference wants 24/7 firm power. Louisiana sits on top of one of the densest gas pipeline systems in the world. For this decade's AI wave, gas is the shortest path . Once fully operational, 10 CCGTs running at high load factors will lock in multiple Bcf/day of regional gas demand for 15+ years. That's functionally a long dated offtake contract for Gulf Coast gas production written by a hyperscaler, not an energy company. Virginia, Ohio, Texas are already running the same playbook. The Louisiana model hyperscaler guarantee, utility-owned gas and wires, existing pipeline network is becoming the template for the entire AI infrastructure buildout. Another data point for the trade I covered my latest article, the link for it is in the below comments👇

The current Jones Act waiver has exposed something that is typically overlooked: many types of oceangoing ships common elsewhere in the world do not exist in the US coastwise fleet. Let’s look at what some of these vessels are and why we don’t have them. 🧵

🇺🇸+🇨🇦 POWER REALITY CHECK 💥NATURAL GAS STILL RUNS NORTH AMERICA Forget the narratives. Look at the grid. In 2025, the biggest source of power across most US states and Canadian provinces is still fossil based. ⚡ Natural Gas dominates the US • Largest power source in 30+ US states • Backbone of the US emissions decline, coal → gas switch • Critical for reliability as power demand surges (AI, data centers, reshoring) Another data point for the trade I covered my latest article, the link for it is in the below comments👇

🔥 Hot of the Presses: We adjusted our global FID and LNG supply forecasts hours after Commonwealth LNG reached a final investment decision Subscriber link at is.gd/Y3SZqe #LNG #ONGT #NatGas #Shale #OOTT #Houston #SWLA #Louisiana

"THE OIL PRICE IS WRONG because it doesn't match my supply/demand modeling" seems to be a persistent view among oil specialists over the past year or so. Pre-Iran War the complaint was that oil prices were too high because everyone was modeling massive oversupply. Included in that view was that somehow China stockpiling didn't count as demand and Saudi/OPEC was flooding the market. Who knows whether Saudi-China anticipated SoH closure (probably not) but it sure felt at the time (as we wrote) that it was a purposeful decision to move below-ground inventory in Saudi to above-ground stocks in China--a great decision with hindsight for whatever the reasons were. Now here we are post SoH closure and many of the same oil glutters are discussing why oil is "only" $100/bbl +/-. Why did LNG spike so high in 2022? A rich region, Europe, was facing gas shortages post Russia cutoff and paid anything to outbid cargoes otherwise destined for Asia. Once that demand was satisfied, LNG prices understandably fell back. Why is oil not higher today? We don't have (and are unlikely to have) shortages in the U.S. Meanwhile Europeans, while concerned about jet and diesel, aren't yet facing shortages. Furthermore, developing Asia immediately went to COVID lockdown mode of work-from-home and cutting marginal travel. The oil price may well spike meaningfully if SoH stays closed! We remain deeply concerned with the current "messy peace" stalemate. We suspect it will be RICH world shortages that drive the big spike everyone thinks should already be here. Does back-end oil need to rally? I am not so sure. Maybe a bit, but not like what was needed over 2004-2014. The world isn't short known developable oil resource relative to expected demand growth in a 2.8% (pre War) trend global GDP world. That is a bid difference with China/BRICs oil super-cycle. The world is short (or tight) REFINING capacity and related MIDSTREAM infrastructure. Getting molecules to end markets is the issue at this time. Our core macro framework remains Geopolitical Super Vol. arjunmurti.substack.com/p/ep98-a-new-e…

🇺🇸President Trump is Unleashing American LNG: Today, @SecretaryWright attended the groundbreaking of the Commonwealth LNG export facility. This will bring Louisiana $13B in investments and is a direct result of President Trump’s LNG export agenda. kplctv.com/2026/05/15/wat…