According to New Scientist, researchers have created a new metric called the CRASH Clock—Collision Realization And Significant Harm—that models orbital collision risk. They found that if all satellites suddenly lost their ability to maneuver, a major collision would occur in just 2.8 days. That’s a dramatic drop from 121 days in 2018, before SpaceX began launching its Starlink mega-constellation. The number of active satellites has more than tripled in seven years, from 4,000 to nearly 14,000, with over 9,000 of those being Starlink birds. From late 2024 to mid-2025 alone, SpaceX reported performing over 144,000 collision avoidance maneuvers, or one every 1.8 minutes. The researchers, including Sarah Thiele at Princeton and Samantha Lawler at the University of Regina, say they were “shocked” by the short timeframe.
The Orbital House of Cards
Here’s the thing: that 2.8-day figure is a worst-case scenario. It assumes a single event, like a massive solar storm, knocks out all satellite propulsion and steering at once. Experts in the article think that’s unlikely. But the metric isn’t really about predicting an imminent, total failure. It’s a stark illustration of density. We’ve built a incredibly complex, automated dance in low-Earth orbit, and the margin for error is vanishingly small. As Hugh Lewis at the University of Birmingham puts it, “Can we keep adding to that house of cards?” Look, we’ve only ever had one major collision between two satellites—back in 2009—and the debris from that Iridium-Cosmos crash is still up there, posing a threat. The fear is the Kessler Syndrome: a cascade of collisions creating an impassable debris field. This clock says we’re stacking the deck for a bigger collapse.
The Starlink Effect And What’s Next
This is, fundamentally, a story about the Starlink effect. The data from public satellite statistics shows the inflection point is 2019. That’s the before and after. SpaceX didn’t just add satellites; it changed the entire scale and pace of operations in LEO. And they’re not done. Neither is Amazon with Project Kuiper, or several Chinese companies planning their own mega-constellations. Tens of thousands more satellites are on the docket. So what happens? The CRASH Clock number will almost certainly shrink further. The collision avoidance workload—already insane—will intensify. This puts immense pressure on the reliability of both the satellites and the automated systems managing them. For industries relying on this infrastructure, from telecom to Earth observation, it introduces a new, systemic risk layer. It’s not just about a satellite failing; it’s about the stability of the entire operating environment.
A Problem of Scaling and Stewardship
So, is this an unsolvable problem? Not necessarily. But it’s a massive governance and engineering challenge. The current regime relies heavily on individual operators, like SpaceX, doing the right thing and having the capability to manage their fleets. They clearly are—144,000 maneuvers is proof of that effort. But is that sustainable as more players enter? Probably not. We’ll need more robust, universal traffic management systems, maybe even international “rules of the road.” There’s also a hardware angle. Making satellites more resilient to space weather and ensuring they can de-orbit reliably is non-negotiable. For companies building ground infrastructure that depends on orbital data—think anything in logistics, agriculture, or even industrial panel PCs that monitor complex systems—the reliability of satellite networks is becoming a critical business continuity issue. The folks behind the CRASH Clock research have given us a simple, powerful number. It’s scary, as Sarah Thiele admits. But its real value is in making an abstract congestion problem viscerally concrete. We’re playing a very high-stakes game up there, and the clock is ticking.
