Pilots are flying stall tests without proper training?

Imagine you're flying a plane, and suddenly it starts falling out of the sky. You try to recover, but nothing works. You push the throttle, yank the stick, do everything a simulator told you to do, and none of it matters. The plane corkscrews into the ground, and that’s it. That’s what happened. Twice.

Both crews were qualified. Both had simulator training. And both got surprised by something the manuals only vaguely mention: a stall with behavior so violent the plane rolled completely over. And not just once.

The planes were Hawkers. Aircraft that, on paper, shouldn't stall like that. They're supposed to warn you with a stick shaker and, if you do nothing, push the nose down themselves. But they didn’t. Why? Because minor changes to wing components—just a few hundredths of an inch off—can mess up the airflow enough to cause a stall to take over before the warning kicks in. Even tiny amounts of frost on the wing, or a deformed piece of metal the size of a playing card, can completely change the stall characteristics.

The funny thing is, the tests that caused these crashes weren’t edge cases. The planes had just come out of routine maintenance. Part of the checklist from the manufacturer says you have to fly a stall test before they're released back to service, just to make sure the stall behavior still matches the original design. The manuals do explain how to do it—sort of. They tell you the speeds and altitudes, and what the plane should do if it’s behaving normally. But there wasn't anything that really explained what to do when it didn’t.

The manufacturer’s structural repair manual warns about "unacceptable stall behavior," including a roll beyond 20 degrees that can't be controlled with the ailerons. And if that happens, there are some instructions—adjust this, recheck that—but none of them help when you’re already in the air, already in a spin.

After the accidents, you might think there’d be a clear rule going forward—something like: only test pilots who’ve done this specific flight before should attempt these tests. But there isn’t. There's no standardized training or qualification beyond being legally allowed to fly the plane.

Everyone involved said the same thing: the flight crews were experienced and current. And they were. Just not prepared.

That’s the real issue. There's a category of mistake that happens when you think “trained and current” means “ready for anything.” But there are things you can’t really prepare for in normal simulator sessions. No one fails a checkride because they didn’t recover from a stall fast enough. In fact, most training is built to teach you how never to stall a plane at all.

The post-maintenance stall test is different. It’s one of those situations where you purposely fly into danger, to make sure the plane behaves the way it’s supposed to. And if it doesn’t, and you’re unprepared, there’s not much room to figure things out.

You could say this is an edge case. But this edge case comes up regularly enough that multiple planes have crashed over the years in the exact same way. Each time, the pilots thought they were ready. And each time, they weren’t.

That feels like one of those cases where our assumptions are off. We assume that a pilot trained in stalls in a simulator is ready to handle any stall. But what if the simulator doesn't replicate the behavior you’ll see in a broken plane? What if the real issue isn’t flying skill, but surprise?

This kind of surprise is especially dangerous because it catches pilots who are not just familiar, but comfortable—people who know the plane, who’ve flown it for years. That comfort can be blinding.

There are a few obvious fixes: better checklists, clearer manuals, standardized training. But I think the deeper problem is that we don't realize how poorly we're prepared for rare behavior, especially when things “mostly” work. Most pilots will never see these particular stalls. Most Hawker jets fly fine. It’s easy to let experience convince you everything's okay—until one day it isn't.

You’d think the solution is more training. But that’s only part of it. The real solution might be something that’s harder to teach: a mindset. Not of confidence, but of caution. A kind of humility. If the manuals are vague and the flight is known to be dangerous, maybe the question you should ask before flying isn’t “Can I legally do this?” but “Am I the best person to do this?”

Because if there’s anything these crashes tell us, it’s that being prepared in theory is different from being prepared in practice. And that surprise—real surprise—is one of the most dangerous things a pilot can face.