The Physics of Rock Bottom

A $2.3B bet on orbit and a 25-year study of 2,420 lives agree on one rule: the lowest point of the path is where a push changes the most.

By Peter Plötner. Aerospace engineer and Wayfinder Life Coach. More about Peter →

NVIDIA just helped a 25-person company raise $250M to put data centers in orbit. Every satellite Starcloud flies will obey a rule of orbital mechanics that most of us get exactly backwards in the worst season of a life.

The news first, because it is worth a minute of wonder. Starcloud is 25 people in Woodinville, Washington. In November 2025 they flew an NVIDIA H100 to orbit on a satellite called Starcloud-1. It was the first data-center-grade GPU ever to reach space, and they used it to train an AI model and to run Google's Gemini up there. In March the company was valued at $1.1B. This week, after a $250M round led by Manhattan West with NVIDIA and Cisco joining, it is valued at $2.3B. The money pays for a 100,000 square foot factory, and the CEO says they intend to ramp toward 100 satellites a week.

Smart people are still arguing about whether data centers belong in space. The rule I want to borrow from Starcloud does not care who wins that argument. It is about the shape of every orbit they will ever fly.

Why you fire the engine at the lowest point

Most orbits are not circles. They are ellipses, stretched loops around the planet, with a highest point and a lowest point. Engineers call the high point apogee and the low point perigee. 2 facts about that loop decide how every mission is planned.

Fact 1: a spacecraft moves fastest at the bottom. At apogee, the high point, it is at its slowest. As it falls back toward the planet it trades height for speed, and at perigee, the low point, it is sprinting.

Fact 2: a burn at the bottom raises the top. If you want to lift your apogee, the highest point you will ever reach, you do not fire the engine up there. You fire it at perigee, and the far side of the loop rises.

Underneath those 2 sits the fact I love most. The same fuel buys more change the faster you are moving. Engineers call it the Oberth effect: an engine burn is worth the most at the point of highest speed. Which is the lowest point of the entire orbit. So mission planners do not treat the bottom as the sad part of the trip. They schedule their biggest burns for it. When Apollo left Earth for the Moon, the push came down low, deep in Earth's gravity well, where the ship was moving fastest.

The 25-year study that found the same shape

The same week as Starcloud's announcement, a study landed in Nature Mental Health that has nothing to do with rockets. Researchers followed 2,420 American adults for 25 years, part of a project called Americans' Changing Lives that began in 1986. They measured well-being the wide way: life satisfaction, self-acceptance, relationships, and the feeling of having a hand on your own controls.

4 paths showed up. 60% of people stayed roughly level. 8% started high and fell steeply. 2% swung up hard from very low and could not always hold the gains. And 30% started low and rose, gradually, year over year.

Here is the line worth reading twice. Among people who started at the bottom, with low well-being and real depressive symptoms, 57% rose over the 25 years. Starting low with symptoms did not predict staying down. It predicted the highest odds of marked improvement in the whole study. And starting high protected nobody: the group that fell steeply was the group that started high.

The researchers add a quiet, practical point: if you can tell which path someone is on, you can tell when help is most likely to land. Not just what kind of help. When.

What the bottom feels like from inside

From inside, a low point does not feel like a point on a path. It feels like a verdict. The quarter where the numbers stopped pretending. The month the co-founder left. The morning you sit in the car outside the company you built, not wanting to walk in. If you are a founder somewhere in that neighborhood, the sentence in your head is probably some version of: so this is what it is now.

I have had my own low seasons. What I remember most is how permanent each one felt while I was inside it. Every one turned out to be a position, not a verdict. I could not see that from the inside, and people telling me did not help much either. So I find it strangely comforting that the data now says my sample size of 1 was not luck. Paths bend. And they bend upward most often for the people starting lowest.

The orbit adds the part the study cannot measure. The bottom is not just survivable. It is the single point on the loop where a push changes the path the most.

Look at why the metaphor holds. At the bottom, you are moving fast whether you chose it or not. The old plan has already broken, so there is nothing left to protect. Decisions you postponed for years are suddenly live this week. The people around you already expect change, so change needs less explaining. Now try making a big move at the top instead, in the comfortable season: everything is slow, everything works well enough, and every change looks like pure loss. That is the expensive place to move. The cheap place is the bottom.

The burn you have been saving

Most of us run it backwards. At the low point we hunker down and tell ourselves the big change can wait until things settle. I will deal with it when things calm down. In orbit terms: we coast up to apogee, the slowest point of the whole loop, and plan our burn for there. The fuel buys the least at exactly the spot where we plan to spend it.

1 honest caveat, in plain sight. 57% is an odds line across 25 years, not a promise for any single year, and 43% of that lowest group did not rise. A low that will not lift on its own deserves real support, the professional kind, not a metaphor. Take the orbit as a reason to reach for that support sooner, not later.

And remember what the burn at the bottom actually does. It does not just soften the bottom. It raises the far side of the loop, the highest point the whole path will ever reach. In orbital mechanics that is not a metaphor. It is just how ellipses work.

Knowing where you are on the path, it turns out, matters as much as how hard you push. That is its own story for another day.

So, 1 honest instrument reading, no judgment attached. Where on the loop are you right now: the comfortable top, where everything is slow, or somewhere near the bottom, moving faster than you would like? And if it is the bottom, what is the 1 burn you have been saving for calmer times? The conversation. The decision. The ask for help. Nobody in orbit waits for the top to fire. The cheapest fuel you will ever have is the fuel you burn down here.

Frequently asked questions

What did the new well-being study actually find?

Researchers followed 2,420 US adults for 25 years as part of the Americans' Changing Lives project, which began in 1986. Well-being followed 4 paths: stable (60%), gradually improving from a low start (30%), declining steeply from a high start (8%), and volatile (2%). Among people who began with low well-being and elevated depressive symptoms, 57% rose over the 25 years, the highest odds of marked improvement in the study. It was published in Nature Mental Health in August 2026.

Why do spacecraft fire their engines at the lowest point of an orbit?

2 reasons. A burn at the low point, called perigee, raises the opposite side of the orbit, so it is how you lift the highest point of your path. And because a spacecraft moves fastest at perigee, the same fuel buys more energy there, an advantage engineers call the Oberth effect. Mission planners schedule their biggest burns for the lowest, fastest part of the orbit.

Does hitting a low point guarantee things get better?

No. The 57% figure is about odds over 25 years, not a promise, and 43% of that lowest group did not rise. The essay's point is smaller and more useful: a low point is a position on a path, not a verdict, and it carries unusual leverage. If a low will not lift on its own, that is the moment for real support, including professional help, not for waiting.

What is Starcloud and why did it raise $250M?

Starcloud is a 25-person company in Woodinville, Washington that builds data centers in orbit, where solar power is continuous and heat can be radiated away to space. In November 2025 its Starcloud-1 satellite carried the first data-center-grade GPU, an NVIDIA H100, into orbit, trained an AI model, and ran Google's Gemini in space. This week it raised $250M at a $2.3B valuation, led by Manhattan West with NVIDIA and Cisco joining, to scale manufacturing toward a planned 100 satellites a week.

I am a founder in a bad quarter. What does firing at the bottom mean in practice?

Pick the 1 move you have been saving for calmer times: the honest conversation, the decision you keep circling, the ask for help. Make it while things are still in motion, because the motion is exactly what makes the move cheap. Waiting for everything to settle first is burning at the top, where the same effort changes the least. Start small, but start at the bottom.

One thing that closes soon. I just earned my certification as a Wayfinder coach, and to mark it I am giving away my full 3 month coaching package to 1 person, free. It is the same package paying clients get: a session every 2 weeks, support in between by message or voice, and the method itself, so you can run it again on your own later. No homework. I read every application myself and they close on September 20. If this essay found you somewhere near the bottom of an orbit, the application is here and takes about 5 minutes.

The companion pieces are Grounded Is Not Broken on what a forced pause really is, and Most Things Aren't Broken. They're Out of Fuel. on telling an empty tank from a broken engine. If you want a first instrument reading on your own orbit, the Essential Self Diagnostic is 15 questions and takes about a minute.

Get new essays by email

New essays, plus the occasional thing I only send by email.