Orbit Alpha Issue 25 · 4 Oct 2026

Starship Goes Orbital While the Smallsat Field Sells Off

Market Overview

🌎 Starship Goes Orbital While the Smallsat Field Sells Off

Space equities split into two distinct stories this week. A broad, sector-wide wobble hit the smaller names Monday through Thursday: Benzinga tied Monday's softness to a risk-off tape across aerospace and defense, with Boeing down 2.57% and the Industrials ETF (XLI) off 0.29% that morning, before most names found a bid into Friday, October 2. SpaceX was the exception throughout: it shrugged off a lockup-driven hangover from the prior week and closed the period up roughly 7% on the back of Starship's first successful orbital flight and a fresh round of bullish analyst notes on its AI-compute-leasing business.

SpaceX (SPCX)
SpaceX opened the week still digesting the September 24 lockup tranche and a $52 million insider stock sale by President/COO Gwynne Shotwell, which had pushed the stock down to roughly $148 by the close of September 25. The turning point came Monday, September 28, when Starship Flight 14 lifted off from Starbase, Texas and reached Earth orbit for the first time, deploying 26 next-generation Starlink V3 satellites before splashing down in the Pacific, while the Super Heavy booster completed a controlled Gulf of Mexico splashdown, per Space.com and CNBC. Analysts piled on from there: CLSA initiated coverage that same day at Outperform with a $250 target, TD Cowen's John Blackledge initiated/reiterated Buy with a $200 target on September 29, arguing SpaceX's AI-compute-leasing business (tied to existing Google and Anthropic deals) could become its largest revenue line by Q1 2027, and Deutsche Bank's Edison Yu reiterated Buy/$235. Momentum built Thursday, October 1, when SpaceX ran three orbital missions in roughly 13 hours: NASA's Crew-13 with a record-fast 7-hour-55-minute ISS docking, the Transporter-18 rideshare carrying about 130 payloads, and a classified NROL-97 mission on Falcon Heavy with both side boosters recovered. The same day brought news that UK supplier Filtronic landed its largest-ever single order (reported at $62–68 million, sources vary) for Starlink gallium-nitride components, and that SpaceX's Nasdaq-100 weighting had jumped to roughly 2.82% from about 1.28% in September's index rebalance. The week's biggest single-day move came Friday, October 2: SPCX closed at $158.96, up 7.35% on volume of roughly 119 million shares versus a ~93–103 million average, a 6.9% gain on the week from the prior Friday's $148.68 close. No lockup tranche actually unlocked inside this window (the next, roughly 7% of the staggered pool, falls around October 9); the one dated filing was a September 30 Form 4 amendment showing director Antonio Gracias distributing 42.8 million Class A shares in-kind among his own holding entities under a pre-set 10b5-1 plan: a structural transfer, not an open-market sale. A retrospective published Sunday, October 4, cited JPMorgan data showing retail investors sold a net $570 million of SPCX over three September weeks, the bank's largest recorded weekly retail outflow for the stock, framing the June-to-October slide from its $225.64 peak as speculative unwinding rather than a fundamentals problem. Short-interest data has not refreshed inside this window; the latest FINRA settlement print, as of September 15, showed 162.3 million shares short (8.97% of float), down from a July peak. We flag any fresher figure circulating this week as unconfirmed.

Rocket Lab (RKLB)
Rocket Lab's week was a sector-wide-softness story with a company-specific rally buried inside it. Shares opened down roughly 1% on September 28 in sympathy with the broader aerospace selloff, then jumped as much as 4.3% on Thursday, October 1, on a cluster of news: a record multi-year deal with Synspective for 20 more Electron launches (bringing contracted Synspective missions to 47 and Rocket Lab's total backlog above 100 missions for the first time), the company's 96th and 97th Electron missions deploying Synspective's 12th and 13th StriX radar satellites, and reports that ARK Investment Management bought roughly 360,000 RKLB shares during the rally. Net for the week, the stock finished almost exactly flat: $73.95 on September 25 versus $73.92 on October 2, masking that intraweek swing. New Street Research's Pierre Ferragu holds the Street-high $150 target, reiterated September 25 (just outside this window).

AST SpaceMobile (ASTS)
AST SpaceMobile's week paired hardware progress with an analyst downgrade. On September 30, the company shipped three more commercial BlueBird satellites (units 14–16) from its Texas facility to Cape Canaveral, and CEO Abel Avellan met with FCC Chairman Brendan Carr to discuss spectrum access; coverage the next day noted the company's production sequence had reached "BlueBird 50," with the operational constellation at 13 satellites in orbit against a reported $1.3 billion contracted revenue backlog. Shares rose roughly 3% on the shipment news. Then on October 2, B. Riley's Mike Crawford downgraded ASTS to Neutral from Buy and cut his target to $65 from $85, citing delayed launches, rising deployment costs, and intensifying competition from rivals including Viasat and Space42's Equatys joint venture, though the stock was little-changed on the call itself. Net for the week, shares were roughly flat to modestly lower.

Spire Global (SPIR)
Spire's clearest event this week was operational: on October 2, the company announced it had launched the first three satellites built at its new U.S. manufacturing facility, flying aboard a SpaceX Transporter-18 rideshare, with management pitching the combined U.S.-and-European manufacturing footprint (300–400 satellites/year of capacity) to "national security and defense partners" wanting domestic-soil production. Shares rose roughly 4% that day but that wasn't enough to offset a rough week: SPIR fell from $12.08 on September 25 to a Tuesday low near $11.15 before closing Friday at $11.11, down roughly 8% on the week. The only SEC filing in the window was a routine Form 144 notice from CEO Peter Platzer on October 1 for a proposed (not executed) sale.

Redwire (RDW)
Redwire generated the most newsflow of the smaller names this week, all partnership announcements. On September 30, Redwire and Sophia Space signed an MOU to develop in-orbit AI/data-center computing infrastructure, explicitly pitched at DARPA, the Air Force Research Laboratory, the Space Development Agency, and the NRO. Shares rose 2.7% that day. Two days later, on October 1, Redwire and Honda announced a joint robotic lab-automation system for future commercial space stations, pairing Honda's multi-fingered hand with Redwire's STAARK robotic arm; shares rose another 2.9%. Redwire was also presenting at the International Astronautical Congress in Antalya, Türkiye, starting September 30. Despite the two positive news days, broader sector pressure won out: RDW fell from $11.62 on September 25 to $10.88 on October 2, down roughly 6.4% on the week.

Planet Labs (PL)
Planet Labs shares were volatile but finished the week essentially flat ($17.40 on September 25 versus $17.51 on October 2, +0.6%) after sliding to roughly $16.14 by Wednesday and then rallying more than 8% into Friday. The rebound coincided with Planet's October 1 launch of 20 spacecraft on SpaceX's Transporter-18 rideshare, including Google's Project Suncatcher AI-chip-in-orbit demonstrator, the Tanager-2 hyperspectral satellite, and 18 SuperDoves, with initial contact confirmed on several. A routine, pre-scheduled Rule 10b5-1 insider sale of 11,500 shares (roughly $188,000) by a Planet filer was reported via Form 144/Form 4 around October 1–2.

Globalstar (GSAT)
Globalstar stayed quiet and essentially flat, closing at $83.22 on October 2 versus $82.88 on September 25 (+0.4%), trading near its 52-week high. Globalstar's XCOM RAN 5G platform was named "Warehouse Technology Innovation of the Year" in an industry awards program on September 29, with negligible stock impact. The dominant context for Globalstar's low volatility is background, not this week's news: Amazon's pending ~$11.6 billion, $90.00-per-share acquisition of Globalstar (announced in April) already has written shareholder consent from holders of 57.6% of shares, with closing still expected in 2027; GSAT's roughly $7 discount to the deal price reflects ordinary time-value/closing-risk, not any new setback.

Firefly Aerospace (FLY)
Firefly shares fell on the week (from $24.36 on September 25 to $23.32 on October 2, down roughly 4.3%) despite a positive mid-week catalyst: on September 30, Firefly announced a partnership with Starcloud to test AI computing hardware aboard Firefly's Elytra orbital transfer vehicle on a mission planned no earlier than 2028, which sent shares up roughly 6% intraday before the gain faded. A Form 4 filed October 2 showed an entity tied to major holder AeroEquity GP distributing 42,215 shares internally to a limited partner "for no consideration," a technical transfer rather than an open-market sale, though worth watching given AeroEquity's roughly 30% stake.

BlackSky Technology (BKSY)
BlackSky shares fell from $23.18 on September 25 to $21.92 on October 2, down roughly 5.4%, despite a commercial-distribution announcement: on September 30, SkyFi launched a "Satellite Intelligence" feature inside Robinhood's Agent App offering retail investors satellite monitoring of sites tied to heavily traded stocks, with BlackSky named the app's first imagery supplier. The market read it as a low-margin licensing deal rather than a meaningful revenue driver, and the stock actually dipped roughly 2.7% that day. Four routine Form 4 filings dated October 1 showed BlackSky directors receiving quarterly compensation in shares (1,055 shares each), not open-market buying.

Takeaway
The week's real story wasn't any single headline: it was the gap that opened up between SpaceX and nearly everyone else it shares a sector with. Starship's first orbital flight and a wave of bullish AI-compute-leasing analyst notes let SPCX shrug off its own lockup-driven retail outflows and post the week's best move by a wide margin, while a broad, largely catalyst-less small-cap wobble knocked down MDA, BlackSky, Firefly, Spire, and Redwire by mid-single-digit percentages even on weeks where several of them had genuinely good news. Watch for SpaceX's next lockup tranche around October 9 to test whether this week's rally can hold through fresh insider-selling pressure, and watch whether the smaller names' late-week rebounds (Spire, Redwire, Viasat, Planet all firmed into Friday) continue or prove to be dead-cat bounces inside a still-soft tape.

Consensus Prices

Launch Watch

• Oct 5 · SpaceX · Falcon 9 · SDA Tranche 1 Transport Layer A · Vandenberg SFB, California
• Oct 7 · KARI (South Korea) · KSLV-2 (Nuri) · NeonSat-2 through NeonSat-6 and others · Naro Space Center, South Korea
• Oct 9 · CASC (China) · Long March 12 · Payload undisclosed · Wenchang Space Launch Site, China
• Oct 10 · SpaceX · Falcon 9 · Starlink Group 15-25 · Vandenberg SFB, California
• Oct 13 · SpaceX · Falcon 9 · Dragon CRS SpX-35 (ISS cargo) · Cape Canaveral SFS, Florida
• Oct 13 (NET, unconfirmed) · ISRO · GSLV Mk II · NVS-03 / IRNSS-1L navigation satellite · Satish Dhawan Space Centre, India
• Oct 19 · JAXA (built by Mitsubishi Heavy Industries) · H3 (Flight 10) · Martian Moons eXploration (MMX), carrying the Idefix rover · Tanegashima Space Center, Japan

Orbit Focus

🎯BETWEEN THE UNLOCKS: WHAT A QUIET WEEK ON SPACEX'S LOCKUP CALENDAR IS TELLING INVESTORS

SpaceX closed this week at $158.96, up roughly 7%: a genuinely strong week by any measure. But the more interesting story isn't the rally itself, it's what happened around it: a stock that just absorbed one of the largest retail-selling episodes on record, sitting in a brief calm spot on its own lockup calendar, that still managed to rip higher on operational news. Understanding why requires walking through the mechanics of SpaceX's unusual post-IPO share-release schedule and the data that's accumulated around it since the company went public in June.

The calendar. SpaceX IPO'd on Nasdaq on June 12, 2026 at $135 a share, opening its first trading day near $150 and briefly implying a market capitalization north of $2.9 trillion before settling down. Unlike a conventional single-date lockup expiration, SpaceX structured its release schedule as a series of staggered tranches: roughly 7% chunks of the restricted share pool unlocking every several weeks rather than one cliff-edge event. The most recent tranche hit September 24, freeing approximately 328.4 million shares (worth roughly $48.7 billion at that day's price) into a market that, by definition, hadn't had to absorb that kind of float before. The next tranche is expected around October 9 (just days after this issue publishes), with a larger roughly-28% release penciled in after Q3 earnings in early November, and full expiry of the remaining restricted pool on December 8, six months after the IPO. Notably, this staggered structure does not apply to Elon Musk's own roughly 6.4 billion Class B shares, which carry 94% of SpaceX's voting power and remain locked until June 12, 2027 regardless of what happens to the smaller, more liquid tranches: meaning every one of these lockup events is a story about early employees and outside investors cashing out, not about Musk's control position changing.

The selling. The September 24 tranche didn't sit idle. Gwynne Shotwell, SpaceX's president and COO, filed to sell 342,170 Class A shares under a Rule 10b5-1 plan she'd adopted back in June, realizing roughly $52 million across several tranches at prices between $151 and $155: the first major post-IPO insider sale from SpaceX's most senior non-Musk executive, and exactly the kind of signal retail investors watch closely even when it's entirely pre-scheduled and non-discretionary. The stock fell about 4% the day that filing hit the tape. By the time this week opened, JPMorgan's trading-flow desk was reporting something more structural: a net $570 million in retail selling of SPCX shares spread across three weeks in September, described by the bank as the largest weekly retail outflow it has on record for the stock, which a Sunday, October 4 retrospective explicitly tied back to the lockup unlock rather than to any fundamental deterioration in the business. Combined with the broader slide from SpaceX's $225.64 all-time high (hit June 16, four days after the IPO) down to roughly $158 today, the picture through late September was of a stock giving back a large chunk of its post-IPO "euphoria" premium as insiders took profits and retail investors who'd chased the open followed them out the door.

The counterweight. What makes this week worth a full feature isn't the selling: it's that the selling didn't win. Starship Flight 14's successful orbital insertion on September 28 (the vehicle's first time reaching orbit, despite one of six Raptor engines shutting down during ascent) gave the bulls a concrete technical milestone to point to, and the analyst community piled in behind it: CLSA initiated at Outperform/$250, TD Cowen initiated/reiterated Buy/$200 with a thesis built specifically around SpaceX's AI-compute-leasing business potentially becoming its largest revenue line within two quarters, and Deutsche Bank held its Buy/$235 call. A Thursday stretch of three successful missions in 13 hours (a NASA crew flight, a 130-payload rideshare, and a classified national-security launch) reinforced the operational-execution narrative just as index mechanics added a technical tailwind: SpaceX's Nasdaq-100 weighting jumped to roughly 2.82% from 1.28% in the September rebalance, meaning passive funds tracking that index had to buy meaningfully more SPCX regardless of their own view on the stock. The net effect was a 7.35% single-day jump on October 2 that erased essentially all of the prior week's lockup-driven weakness and then some, closing the week up roughly 6.9%.

What this tells you. The lesson isn't that lockups don't matter for SpaceX: the $570 million in retail outflows and Shotwell's $52 million sale are real, measurable effects of real selling pressure. It's that lockup-driven selling pressure and operational-catalyst-driven buying pressure are running on different, overlapping clocks, and in any given week one can simply overwhelm the other. The September 24 tranche produced several days of weakness; Starship's orbital success produced one day of outsized strength that was larger in magnitude. That asymmetry is worth watching closely as the calendar moves forward: the next tranche (roughly October 9) is similar in size to the one that just caused the dip, but it lands in a week with a thinner known catalyst calendar than the one SpaceX just had. If insider selling resumes without a Starship-sized offset to absorb it, the more instructive test of this dynamic may be just days away. The November tranche is larger still (roughly 28% of the restricted pool, timed right after Q3 earnings), which makes earnings-day execution unusually important for SpaceX this cycle: a strong print could give the market another catalyst big enough to absorb a much bigger unlock, while a disappointing one would remove the cushion at the exact moment the largest wave of new sellable shares hits the market. One data gap worth flagging for investors tracking this closely: official FINRA short-interest settlement data for SpaceX is only current through September 15 as of this writing, so any claims about SPCX short interest circulating this week should be treated with real skepticism until the next settlement print is published.

Takeaway. SpaceX's lockup calendar isn't a single cliff-edge risk: it's a recurring, roughly monthly pressure test, and this week showed that a strong enough operational catalyst can outrun it. The company just proved Starship can reach orbit and that three-launches-in-a-day operational tempo is achievable; what it hasn't yet proven is whether that kind of catalyst will be available every time a tranche unlocks. With the next release due around October 9 and a much larger one following Q3 earnings in early November, the next six weeks are a genuine test of whether SpaceX's stock price is being driven by its launch cadence or by its float mechanics, and right now, it's not obvious which one is actually in charge.

This newsletter is for informational purposes only and is not investment, legal, or financial advice. We believe our sources are reliable but can't guarantee accuracy or completeness, and all figures should be independently verified before you rely on them.

Previous: Issue 24: A Sector Rally Meets a Wall of Unlocking Shares · Next: Issue 26: SpaceX Turns Carrier, and the Direct-to-Device Trade Pays the Price