Steidle's Law

Every time I go to the Udvar-Hazy Center, the first thing I go see is the X-35B. It’s the Lockheed demonstrator that won the Joint Strike Fighter competition, sitting there with its lift fan, looking a lot less important than it was. And every time, I end up explaining the whole story to whoever I’m with: the fly-off, the down-select, why a plane that never went to war matters so much. For me the story isn’t really about the airplane. It’s about a way of running a program that I learned from the man who ran that competition.

Way back in the day I worked for Rear Admiral Craig Steidle. Before NASA he ran the Joint Strike Fighter program, and when I knew him he was running Exploration Systems at NASA, from early 2004 to mid 2005. I learned more from him than from almost anyone. We talked a lot about systems of systems and spiral development.

But the thing he said that stuck with me most had nothing to do with engineering. It was about politics, and it went something like this:

Every two years, hand the people who funded you a win they can take credit for.

I call it Steidle’s law.

It sounds simple, and the arithmetic is. Every two years the whole House and a third of the Senate face voters. Every four years there’s a new president, or the same one who needs something to show. Every six years a senator has to explain what they did with their term.

Space is exactly the kind of spending that gets scrutinized. It’s expensive, it’s far away, and the payoff is years out. Nobody in Washington has to defend a highway. A moon program, they do.

We still had the buy-in. Congress and the White House both backed us. The vision was beautiful, but what kept them on board was that we kept giving them wins.

As Steidle explained it, every one of those people who voted for our budget was making a political bet. They were putting their name next to a program that could easily become a punchline, in front of constituents who’d rather see the money somewhere closer to home.

When you deliver on a milestone, you’re paying back that bet. You’re giving them something to stand in front of and say “I made this happen.” Do that on the cycle, and the next vote is easy. Miss it, and you’ve made their bet look bad, and nobody keeps backing a bet that makes them look bad.

That’s why Steidle cared so much about spiral development, and he didn’t invent it at NASA. He brought it from the Joint Strike Fighter. JSF didn’t pick a winner off a stack of paper. Several companies competed, the field was cut to two, and Boeing and Lockheed each had to build and fly a real demonstrator before anyone was chosen. The airplane itself was eventually fielded in blocks, each adding capability to the last.

He set Exploration Systems up the same way. Several contractors studied the architecture, and the plan was to down-select to two crew vehicle teams that would each have to fly a demonstrator in 2008 before a winner was picked. The whole program was laid out in spirals: get a crew vehicle flying, then go back to the moon, then go further. That structure was his engineering answer to the political problem. Every spiral ends with a real thing to show. And a competition that has to fly is a win you can stand in front of all by itself.

Two-year wins can turn into stunts: a demo built for the photo, a milestone redefined until it’s met. That isn’t what he meant. The win had to be a real step toward the goal, and people can usually tell the difference. I’ll admit that’s where the idea is weakest. A program under enough pressure can talk itself into calling a stunt a step, and I’m not sure the cadence alone protects you from that.

---

We delivered. Then the leadership changed, the program was restarted, and the cadence broke. Deadlines started slipping. Suddenly there was nothing new to stand in front of, and the program we had was gone.

The replacement didn’t fare better. Constellation was proposed for cancellation in 2010 and shelved by Congress that fall. Pieces of it survived, the capsule most of all. Orion was picked in 2006 and didn’t carry a crew until Artemis II flew this April. Twenty years.

I don’t think the people who restarted it were fools. I think they underestimated the clock. A better architecture that can’t show a win inside an election cycle loses to a good architecture that can. I can’t prove our plan would have held up either. I was inside it, and I’d like to think it would have, but nobody got to run that experiment.

Strip away the rockets and the Senate and Steidle’s law comes down to this: hit your milestones or it’s over. I learned it inside politics, where the consequences are loud, but it works the same anywhere someone funds you on a cycle, whether that’s a board, investors, or a CEO deciding where next year’s money goes. They’re all making a bet on you, and they all need something to point to.

So I still ask the question Steidle taught me: in two years, what will the people who backed this be able to stand in front of? If I don’t have an answer, I don’t have a plan yet.