Midnight MarketsV4.0
⚠ Not financial adviceNot financial advice. Independent research, not investment recommendations. The author and website make no representation as to accuracy or completeness. Do your own due diligence before acting on anything you read here. Read the full disclaimer →
Sector V · 10 chokepoints · 22 names

Space

The Orbital Stack — Space

Imagine the world is suddenly building 100 highways at once — but not on Earth. The highways are in orbit, on the Moon, and between the two. Most of the news, most of the money, and most of the stock chatter is about which rocket companies will be the biggest, which billionaires will plant their flags first, and which constellation will beam internet to the most people.

Meanwhile, in a quiet building in Manchester, New Hampshire, there's one company that makes the rad-hard chip every NASA mission relies on. There's one small factory in Jacksonville, Florida, that builds the roll-out solar panels powering the ISS and the Lunar Gateway. There's three Western suppliers of the engines that move satellites once they're up there. There's two companies with hourly-revisit cameras pointed at the Earth, and one of them just invented a way to sell whole satellites — plus their data feeds — to allied governments by the dozen.

SpaceX is the gravity. It dominates 90% of payload mass to orbit. It is also private — you cannot buy it. So every dollar that wants "space exposure" gets funneled into the public pure-plays. If you own a piece of the small suppliers, you don't care which rocket company wins. You're getting paid by all of them, by NASA, by the Space Force, and increasingly by Sweden and Japan and Germany. That's the narrow places. Same playbook as AI and Robotics. Different orbit.

Wildcard · Sector-Wide Catalyst

When SpaceX goes public.

SpaceX has not filed publicly as of May 21, 2026 — but tender-offer valuations have ranged from $350 billion to over $400 billion through 2025–2026, and both a Starlink-only spin and a whole-company IPO have been actively rumored. Whenever the filing lands — weeks, months, or quarters away — it reprices the entire space asset class in a single trading session. SpaceX flies roughly 90% of payload mass to orbit and is the gravitational center of every name on this page. Position sizing must account for this; the IPO is simultaneously the biggest sector catalyst and the biggest tail risk.

Choke 01

The Ride — Orbital Launch Vehicles

Electron · Neutron · Alpha · Eclipse · MLV-Class Rockets

Five companies in the world can put a payload in orbit on a Western rocket. Only two are publicly listed pure-plays.

Launch is the gate to space. SpaceX flies roughly nine out of every ten kilograms of payload that leave Earth — and it is private. ULA (Boeing/Lockheed) and Blue Origin are not separately listed. Among public pure-plays, Rocket Lab is the only company with a flight-proven, regularly cadenced orbital rocket (Electron, 21 launches in 2025 at 100% success), and it is building Neutron to compete in the SpaceX-vacated medium-lift gap. Firefly Aerospace IPO'd in August 2025 on the back of becoming the first commercial company in history to fully succeed at a Moon landing. There is essentially no Western "medium-lift" rocket between Falcon 9 and the heavy New Glenn / Vulcan class — that gap is the prize Rocket Lab and Firefly are racing to fill.

Why this is a chokepoint

If you can't get to orbit, nothing else on this page works. With SpaceX still private, every dollar of public "space exposure" has been funneled into Rocket Lab — pushing valuations to extremes but with real revenue underneath.

Price

Only public Western company that builds AND operates its own orbital rocket and its own satellites

Operates Electron (21 launches in 2025 at 100% success) and is in final integration of Neutron , an 8-ton-to-LEO medium-lift rocket targeting first flight Q4 2026. Q1 2026 revenue $200.3M (+63.5% YoY); backlog $2.2B; GAAP gross margin 38.2%. Signed largest-ever launch contract during Q1 (five Neutron flights for a confidential customer through 2029). Selected alongside Raytheon for Golden Dome's Space-Based Interceptor program; won a $190M MACH-TB 2.0 hypersonic block buy. Closed the Mynaric acquisition (laser inter-satellite links) Q1; signed definitive agreement for Motiv Space Systems (Mars-proven robotics). Liquidity exceeds $2B.

Price

Only commercial company in history to fully soft-land on the Moon — plus a small rocket and a medium-lift in joint development

Operates the Alpha small launcher (return-to-flight cleared late 2025), the Blue Ghost lunar lander (Mission 1 completed all 17 objectives on the Moon, March 2, 2025 — first fully-successful US lunar landing since Apollo 17), and the Elytra orbital transfer vehicle. Developing the medium-lift Eclipse rocket with Northrop Grumman ($50M strategic investment). FY25 revenue $159.9M (+163% YoY); FY26 guidance $420–450M. Blue Ghost Mission 4 awarded by NASA July 2025 at $176.7M; backlog above $1.3B. Selected for Golden Dome SBI; closed the SciTec acquisition for missile-warning data processing.

Choke 02

The Brain — Radiation-Hardened Electronics

RAD750 · RAD5545 · RTG4 FPGAs · PolarFire SoC

The chips that survive cosmic radiation are made by three Western suppliers. One pure-play. Two are buried inside conglomerates.

Ordinary semiconductors die in orbit — cosmic rays would flip the bits of a normal smartphone chip within days. Rad-hard chips are specifically engineered, tested, and qualified — qualification cycles run 5 to 10 years. The dominant US single-board computer is BAE Systems' RAD750 (powers James Webb, Curiosity, Perseverance, GPS III; over 150 units in service). Microchip Technology supplies the rad-hard FPGAs and microcontrollers used on most US national-security smallsats. Frontgrade (formerly Cobham AES, Veritas Capital portfolio) is the third leg and is private. ITAR export controls plus the qualification moat make these three a rotating monopoly.

Why this is a chokepoint

Every satellite needs rad-hard parts. Substitution is impossible within five years — this is the most durable moat in the entire space stack.

Price

Maker of the RAD750 and RAD5545 — de facto US standard rad-hard space computer

BAE's Electronic Systems sector houses the world's longest-running rad-hard chip line . RAD750 is still flying on James Webb, Curiosity, Perseverance, multiple GPS III birds, and LRO. RAD5545 (45nm) powers software-defined radio payloads for Lockheed Martin. RAD510 is the next-gen. In 2021 DoD selected BAE to build the RH12 storefront — 12nm rad-hard ASIC design services for the US defense industrial base. BAE rad-hard systems have logged "over 11,500 years of successful operations in space" with no known failure.

Price

Dominant supplier of rad-hard FPGAs and microcontrollers for US national-security smallsats

Microchip's aerospace & defense business — built on the Microsemi/Actel acquisition — supplies rad-hard FPGAs (SmartFusion, RTG4, PolarFire SoC), microcontrollers, memory, and analog/mixed-signal parts . PWSA Tranche 1 and Tranche 2 satellites are filled with Microchip parts. Pure space revenue is a small fraction of Microchip's ~$5B top line.

Choke 03

The Wings — Satellite Structures, Solar & Deployables

ROSA Solar Arrays · Booms · Reaction Wheels · Star Trackers · Lunar Power

Every satellite needs a body, wings, and the moving parts to point itself. One US company owns the dominant flexible solar array.

Beyond launch and brains, you need the body: structure (the satellite bus), power (solar arrays), pointing (reaction wheels, star trackers), and deployables (booms and antennas that fold for launch and unfurl in orbit). Redwire owns roll-out solar arrays (ROSA) — flexible blankets that unfurl in space, which power the ISS's recent upgrade and which NASA selected as the baseline power source for the Lunar Gateway. Lockheed closed its Terran Orbital acquisition late 2024 , taking the leading public smallsat-bus integrator off the board. Maxar was taken private. Public pure-plays shrink every year — Redwire is increasingly the only one standing.

Why this is a chokepoint

Contested chokepoint with a thinner moat than rad-hard — but Redwire's ROSA franchise plus the Andromeda IDIQ position it for the next decade of GEO and lunar power systems. Scarcity value compounds as public peers get acquired.

Price

US prime for roll-out solar arrays, VLEO platforms, lunar power infrastructure — plus a tactical-drone arm

Q1 2026 revenue $96.97M (+57.9% YoY); record backlog $498.1M (+71% YoY) ; book-to-bill 1.92. Selected as one of 14 vendors on the US Space Systems Command Andromeda IDIQ — originally $1.8B over 10 years for geosynchronous space-domain-awareness spacecraft, with target shared ceiling raised to more than $6 billion . Gross margin jumped to 26.6% from 14.7%. FY26 revenue guidance $450–500M (~42% growth midpoint). Other lines: SabreSat VLEO platform ($44M DARPA Otter Phase 2), Eclipse Prime lunar surface infrastructure, IBDM on The Exploration Company's Nyx, 11 active payload facilities on the ISS . Edge Autonomy delivered 100+ Stalker/Penguin drones across seven countries since June 2025, including US Army LRR. Truist initiated with a Buy in May 2026.

Choke 04

The Push — In-Space Propulsion

Hall Thrusters · Ion Engines · Chemical Bipropellant · Solid Rocket Motors

Every satellite that lives more than 18 months needs to maneuver. Pure-play public exposure is unusually thin.

Once in orbit, atmospheric drag and orbital mechanics push your satellite off station. Stationkeeping requires propulsion. Two flavors: chemical thrusters (instant, high-thrust) and electric/ion/Hall-effect thrusters (slow, fuel-efficient, ideal for stationkeeping). Aerojet Rocketdyne was the dominant US chemical-propulsion house until L3Harris acquired it in 2023 for $4.7B . Moog is the next-largest public supplier. Hall-thruster specialists Busek and Phase Four remain private. Rocket Lab introduced its Gauss electric thruster in Q1 2026. Cleanest public exposure today is via L3Harris and Moog — both diversified conglomerates.

Why this is a chokepoint

As constellations proliferate, every satellite needs propulsion. Volume is exploding; pure-play public access is rare. Solid rocket motors (Golden Dome interceptors) are in critical shortage — only Aerojet and Northrop's ATK qualify domestically.

Price

Owns Aerojet Rocketdyne — dominant US chemical & electric in-space propulsion, plus SRMs

L3Harris absorbed the dominant US chemical propulsion supplier in 2023 — engines on the Centaur upper stage, in-space attitude-control thrusters on most NASA missions, the RS-25 SSME for SLS, and solid rocket motors for missile-defense interceptors . Aerojet's electric propulsion (XR-5 Hall thrusters) is a primary candidate for proliferated-LEO. L3Harris is also a prime on multiple PWSA tranches.

Price

Motion-control components for the entire satellite-prime universe

Supplies motion-control components — propulsion thrusters, valves, slip rings, fluid control — to satellite primes worldwide. Repeat business with essentially every public and private satellite builder.

Choke 05

The Doorway — Ground Stations & Antennas

Phased Arrays · VSAT Terminals · OpenSpace Orchestration · Multi-Orbit Roaming

A satellite is useless if you can't talk to it. The ground segment is in the middle of a once-a-generation upgrade.

Every photo, text, and command between Earth and a spacecraft passes through a ground station. Big steerable parabolic dishes are being supplemented by electronically-steered phased arrays that track multiple LEO birds simultaneously. Multi-orbit support — roaming between Starlink, OneWeb, Iridium, Kuiper, and military networks on the same terminal — is now table stakes. Kratos supplies the largest US commercial satellite ground network and is the primary vendor for the SDA's ground operations. Viasat owns gateway and aviation broadband. Gilat is the leading commercial VSAT vendor for LEO.

Why this is a chokepoint

Every new constellation needs hundreds of new antennas plus orchestration software. Kratos's OpenSpace virtualized ground software is already the operational backbone for parts of SDA Tranche 0/1.

Price

Dominant US commercial satellite ground network + hypersonics + Valkyrie drones + jet engines + SRMs

Q1 2026 revenue $371M (+22.6% YoY); organic growth 15.8%; book-to-bill 1.6×; record backlog $2.0B . FY26 revenue guidance raised to $1.700–1.760B. Hypersonics business on track for $400M in 2026 and $700M in 2027 ; "$1 billion-plus sole-source hypersonic program expansion verbal award" expected to be formalized shortly. Valkyrie XQ-58A drone scaling to ~40 units/year by early 2028. Jet-engine LRIP for cruise missiles begins later in 2026 (target: several thousand engines annually in 2027). Kratos OpenSpace virtualized ground software is the operational ground segment for parts of SDA Tranche 0/1. Per Q1 2026 8-K: "fiscal 2027 national security spend currently projected to be $1.5 trillion, an approximate $411 billion increase above Fiscal Year 2026."

Price

Global satellite broadband + gateway/IFC + Inmarsat L-band spectrum

Operates ViaSat-3 broadband constellation (Flight 1 in service; Flight 2 entering service May 2026; Flight 3 launch late summer 2026), the Inmarsat L-band fleet , and the world's largest aviation IFC install base. Q3 FY26 revenue $1.16B (+3% YoY); adjusted EBITDA $387M (33% margin); backlog ~$4B. Management has guided to positive FCF in FY26.

Price

Modems + gateway equipment + managed services for global constellation operators

Sells satellite modems, gateway equipment, managed network services . Acquisitions of Stellar Blu (electronically-steered antennas for IFC) and Wavestream (Ka-band HPAs for US government) shifted the mix toward higher-growth segments.

Choke 06

The Pipes — Satellite Communications Services

L-Band MSS · Ka-Band Broadband · IoT · Spectrum Re-Rating

Five publicly listed operators run the meaningful Western satellite networks. Amazon's $11.6B Globalstar acquisition just reset the spectrum floor.

The big public Western pure-plays are Iridium (L-band IoT, voice, government), Viasat (Ka-band broadband, IFC — covered in Choke 05), Globalstar (L/S-band IoT, Apple emergency messaging — being acquired by Amazon), AST SpaceMobile (S/L-band direct-to-cellphone — covered in Choke 10), and the private Eutelsat-OneWeb combine. Spectrum rights — granted by national regulators and the ITU — are the moat.

Why this is a chokepoint

Amazon's $11.6 billion acquisition of Globalstar (announced April 14, 2026) re-rated spectrum-rich operators across the board. The Street has not fully digested this yet — Iridium's average analyst PT is still $32.40 vs current $45.74.

Price

Only operator with truly global, pole-to-pole L-band coverage — the US government IoT network

Operates 66 cross-linked LEO satellites in L-band — voice, low-bandwidth data, IoT, aviation/maritime safety. Q1 2026 revenue $219M. Announced May 13, 2026: Iridium is purchasing the remaining 61% of Aireon (aviation surveillance JV) for $366.7M. The DoD EMSS Gateway contract — the dedicated US government Iridium gateway — remains in place.

Price

Apple's emergency-messaging satellite partner — being acquired by Amazon for $11.6 billion

Operates a LEO constellation that powers Apple's Emergency SOS via Satellite (66% of Q1 2026 revenue of $70.1M came from Apple wholesale capacity). On April 14, 2026, Amazon announced a definitive agreement to acquire Globalstar at $90/share cash or 0.3210 AMZN shares , valuing equity at ~$11.6B. Apple publicly endorsed the transaction. 58% of GSAT shareholders had already approved as of announcement.

Choke 07

The Eyes — Earth Observation & Geospatial Intelligence

Daily Revisit · Hourly Revisit · Sovereign Sat-as-a-Service · RF Geolocation

Three public companies image the entire planet daily or near-daily. Sovereign sat-as-a-service is reshaping the business model.

Planet Labs has the largest daily-revisit constellation. BlackSky offers hourly revisit with 35cm Gen-3 satellites. Satellogic is the low-cost player. Maxar is private. Airbus owns the European very-high-resolution end. The biggest 2024–2026 shift is sovereign satellite-as-a-service — selling allied governments their own physical spacecraft plus a multi-year data contract. Sweden, Germany, Japan have signed .

Why this is a chokepoint

Ukraine, Gaza, Taiwan have proven that commercial EO is now a national-security utility. Every allied government is buying. Planet's three sovereign-sat deals in 12 months total over half a billion dollars in contracted recurring revenue.

Price

Largest daily-revisit Earth-observation constellation — pioneer of sovereign sat-as-a-service

Operates ~200 satellites — the SuperDove fleet (3-meter resolution, full Earth daily) and the new Pelican high-resolution constellation (sub-meter, with on-orbit AI processing). First Pelicans launched May 3, 2026 from Vandenberg, including the first sovereign Swedish Armed Forces satellite . FY26 revenue $307.7M (+25.9% YoY); Q4 FY26 revenue $86.8M (+41% YoY); RPOs $852M (+106% YoY); backlog over $900M (+79% YoY). Three sovereign-satellite-services contracts in 12 months totaling over half a billion dollars (Japan/JSAT, Germany, Sweden). R&D partnership with Google to explore data centers in space.

Price

Highest-revisit-rate EO operator — hourly cadence on key regions, 35cm Gen-3 in operation

4 Gen-3 birds operational, with 35cm imaging plus AI analytics through the Spectra platform . Q1 2026 revenue $20.8M. Up to $160M in new contracts year-to-date 2026 : $99M sole-source IDIQ from AFRL, $25M multi-year subscription with an international MoD, an international defense customer scaling to ~$30M ARR. FY26 revenue guidance raised to $130–150M. Space-based-intelligence services expected to grow over 50% in 2026.

Price

Low-cost constellation-as-a-service for emerging-market sovereigns

Q1 2026 revenue $6.1M (+80% YoY) with first-ever positive net cash from operating activities . Cantor PT raised to $10 from $7 (May 13, 2026); Northland to $9.

Price

Radio occultation weather data + RF geolocation + ADS-B

Q1 2026 revenue $15.8M (-34% YoY but +13% ex-maritime), gross margin 40%. NOAA radio-occultation contract expanded . RFGL pipeline strengthened with US/European customers (including NRO). Recent $65.5M private placement strengthened cash to ~$115M pro forma. A Canadian wildfire-monitoring contract worth up to C$71.8M was terminated post-quarter.

Choke 08

The Highway — In-Space Services & Cislunar

CLPS Lunar Landers · Near Space Network · Commercial Space Stations · Debris Removal

NASA's CLPS program turned one private company into a publicly-listed lunar prime. Voyager IPO'd in 2025 to fund the ISS replacement.

Above LEO, the action is on the Moon, around the Moon ( cislunar ), and keeping LEO clean. NASA's Commercial Lunar Payload Services (CLPS) program contracts private companies to land payloads on the Moon — Intuitive Machines is the prime that's done it twice (IM-1 February 2024, IM-2 March 2025). Artemis II's crewed lunar flyby is targeted for late 2026 / early 2027 and is the catalyst calendar's biggest event. Voyager Technologies (NYSE: VOYG, IPO'd June 2025) is the prime for Starlab. Astroscale (Tokyo 186A.T) is the global leader in active debris removal.

Why this is a chokepoint

Artemis is the multi-decade demand catalyst. LROC, NSN, and Human Landing System contracts are multi-billion-dollar pipelines. The ISS decommissions by 2030 — the Phase 2 CLD award is the prize.

Price

The only publicly listed lunar lander prime that has actually soft-landed on the Moon

Q1 FY26 revenue $186.7M and first-ever positive adjusted EBITDA of approximately +$2.7M . Contract pipeline includes the Near Space Network (NSN) services contract from NASA (multi-billion-dollar cislunar comms relay), LROC prime designation, multiple CLPS task orders (IM-3, IM-4), and the recent Lanteris acquisition (propulsion and integration). Cantor Fitzgerald's Andres Sheppard raised PT to $43 from $26 in mid-May 2026 and called LUNR the top public-market Artemis II beneficiary . Other raises: Roth $50, B. Riley $45, Clear Street $44.

Price

Prime for Starlab — one of three NASA-selected commercial ISS replacements + defense systems

IPO'd on NYSE June 11, 2025 at $31. Three segments: (1) Defense & National Security ; (2) Space Solutions ; (3) Starlab Space Stations . Starlab is a JV (Voyager 67% / Airbus 30.5% / Mitsubishi / MDA Space / Palantir) developing the commercial ISS replacement. Voyager has received $217.5M in NASA Space Act Agreement funding . Total liquidity at end of Q2 2025 was $669M, debt-free.

Price

Global leader in active debris removal, end-of-life deorbit, GEO life extension

Operates ADRAS-J (first close-approach inspection of a derelict rocket body in LEO) and is building EOL-J for JAXA debris removal. Multiple ESA, UKSA, and Space Force/SDA contracts via UK and US subsidiaries. Targeting FY26 revenue of ¥36B (~$240M) for operating breakeven.

Choke 09

The Shield — Defense Space & Golden Dome

PWSA Tranches · MILNET · Space-Based Interceptors · Hypersonic Defense

Golden Dome — $185B official, $1.2T CBO-modeled — is the largest defense procurement in a generation.

Golden Dome is a layered missile-defense system. Three pillars: (1) SDA's Proliferated Warfighter Space Architecture for tracking and command/control (Tranche 1 ~154 satellites in 2026; Tranche 2 ~270 satellites starting late 2026; Tranche 3 funding restored); (2) Space Data Network / MILNET backbone (largely SpaceX Starshield, contractor-operated for Space Force and NRO); (3) Space-Based Interceptors (SBI) — kill vehicles in orbit. The FY26 Defense Appropriations Act allocated ~$13B for Golden Dome / SDA , including $7.2B for military space-based sensors, $2.2B for hypersonic defense, $2B for AMTI satellites, $1.9B for ground-based radars. Largest US primes — Lockheed, Northrop, L3Harris, RTX, Booz Allen — are the integrators. Pure-play winners include Kratos (Choke 05), Mercury Systems , Rocket Lab (Raytheon-partnered SBI), Voyager, Redwire (Andromeda), Spire (RFGL), BlackSky (AFRL EO).

Why this is a chokepoint

Dominant defense-budget catalyst through the late 2020s. Each contract win is a multi-year ARR step. Fiscal 2027 national security spend projected at $1.5 trillion — a ~$411B increase over FY26.

Price

Rugged signal-processing, RF, and storage subsystems — including data recorders on SDA Tranche 3

Q3 FY26 (May 5, 2026) was a record: bookings $348.3M (book-to-bill 1.48); backlog ~$1.6B (+18% YoY) ; revenue $235.8M (organic growth 11.5%); adjusted EBITDA $36.1M (+46% YoY, 15.3% margin). Headline win: L3Harris selected Mercury to provide Solid-State Data Recorders for SDA Tranche 3 Tracking Layer satellites (April 2, 2026). Closed the SolderMask acquisition (March 2026).

Price

Diversified defense conglomerates with Golden Dome exposure inside broader businesses

Lockheed Martin (LMT) — prime on Next-Gen OPIR; acquired Terran Orbital late 2024. Northrop Grumman (NOC) — Tranche 1/2/3 Transport Layer prime; SRM franchise. L3Harris (LHX) — Tranche prime; owns Aerojet Rocketdyne. RTX — Raytheon partnered with Rocket Lab on SBI. Booz Allen (BAH) — top systems integrator on Golden Dome.

Choke 10

The Phone — Space-Based Cellular / Direct-to-Device

D2C/D2D · BlueBird · Starlink+T-Mobile · Apple/Globalstar→Amazon Leo

Three architectures compete to put a cell tower in space. Two are publicly investable; one is dominant in commercial service today.

"Direct-to-cell" lets a normal smartphone — no special hardware — connect to a satellite when terrestrial coverage fails. Largest TAM expansion in mobile telecom in a generation. Three credible architectures: 1. AST SpaceMobile (ASTS): Huge phased arrays in LEO using MNO-licensed spectrum. Highest peak speeds (98.9 Mbps measured). 45 BlueBirds targeted by end-2026. Nearly 60 MNO partners covering 3 billion subscribers. 2. SpaceX Starlink D2C + T-Mobile: Already in US commercial service. Thousands of cellular-payload Starlinks deployed. Not directly investable except via speculation on a SpaceX IPO. 3. Apple Emergency SOS / Globalstar → Amazon Leo: Lower bandwidth today, but installed in 1B+ iPhones globally. Amazon's $11.6B acquisition (Choke 06) reshapes this lane.

Why this is a chokepoint

Whichever architecture dominates will price like a global telecom utility. ASTS reflects "growth/risk"; GSAT reflects "M&A arb." SpaceX is the unmodelable wild card.

Price

Only public pure-play in the race for true 4G/5G broadband direct to unmodified smartphones from space

Deploying the BlueBird constellation — massive phased-array antennas. Q1 2026 revenue $14.7M versus a $37.5M sell-side consensus. FCC granted full commercial Supplemental Coverage from Space authorization during Q1 2026. Nearly 60 global MNO partners covering 3 billion subscribers. Network deployment target: 45 BlueBirds in orbit by end of 2026 (BlueBird 8/9/10 launching mid-June 2026 on Falcon 9). ~$3.5B in cash. FY26 revenue guidance reaffirmed at $150–200M. New record 98.9 Mbps peak data speed to an unmodified smartphone.