The OODA loop began with a fighter pilot who wanted to teach what he knew. In the 1950s, John Boyd had a reputation as the best dogfighter at the US Air Force's Fighter Weapons School, with a standing bet that he could start in a losing position and beat any pilot within forty seconds. He never lost. What set him apart is that he did not just win, he worked out why, and wrote the first manual of jet air-combat tactics.
Over the next twenty years he kept pushing the idea further. What he had worked out in a cockpit turned out to describe how people decide under pressure of any kind, when the stakes are real and the answer is not laid out in front of them. He called it the OODA loop: observe (take in what is happening), orient (make sense of it), decide (choose what to do), and act (do it, and see what changes). How fast you run it turns out to matter as much as the four steps themselves.
The example he came back to was the Korean War. American F-86 pilots kept beating the MiG-15 even though the MiG could climb faster and fly higher, and the reason was not on the spec sheet. The F-86 had a bubble canopy that gave its pilot a clear view all around, where the MiG pilot's framed canopy left blind spots, and the F-86's powered controls let the pilot switch from one move to the next faster. So the American could see the fight, make sense of it, and act a beat ahead, again and again, until the other pilot was always a step behind. The engineering of the jet mattered, but only because it let the pilot run the loop quicker.
So the loop is not really about flying. It works on any hard problem where the situation keeps changing while you deal with it: a stuck negotiation, a business under pressure, a fault you cannot find in a system. You run through the steps, and because acting changes the situation, you are back to observing. But you return with new information that thinking alone would not have given you, and yourself a little more experienced for having acted.
Worth holding as you read on: whether your read of the situation is honest, whether you are moving at the right speed for the problem (which the problem sets, not you), and what to do when what you are up against is another person reading you back.
The loop is not a circle
People usually picture the loop as four boxes in a circle, an arrow from each one to the next, and you walk around the circle. That picture is wrong, and Boyd's own diagram is the fix.
The diagram shows a few things the neat circle leaves out.
- Orient is the centre, not one stop of four. How you orient changes what you notice, what choices you can even see, and how you act. The other three steps bend around orient.
- You can act without deciding. When your read is clear enough, the action just follows, with no separate moment of deciding. Boyd called this implicit guidance and control. You know the feeling: you catch a falling glass before you have decided to reach for it, or you hit the brake before you know why. Nothing got skipped out of carelessness.
- A surprise throws you back. One thing you did not expect can drop you into making sense of the situation again, mid-action, with no tidy return to the start.
- One read can set off several actions at once, not a single next move.
So it is not observe, then orient, then decide, then act, then round again. You move between the four in whatever order the situation forces on you, and a lot of it is knowing when to go straight from a clear read to acting, instead of working all the way around the loop.
Orient, the heart
Orient is the hard part. The other three steps are quick; orienting is slow, because it is where you build your read of the situation, and everything you do next runs on that read. So orienting well is most of the skill, and there is no formula for it. A read can go wrong in more ways than anyone could list: your own bias, an oversimplification you never noticed, someone feeding you a line, an AI telling you what you want to hear.
Whatever threw your read off, the move is the same: watch for the mismatch. You are always carrying a picture of how things are, and when something you see does not fit that picture, that mismatch is worth more than almost anything else you have, because it is the sign that your read and the world have come apart somewhere. You do not need to know which cause is to blame. The hard part is that the mismatch is usually the thing you least want to see, so noticing it at all is a skill: catch the thing that does not fit, and act in a way that tests it instead of explaining it away.
One situation is worth pulling out on its own, because it defeats that tool: the thing you are reading is not the weather or a machine but another person, and they are reading you back. When someone is actively shaping what you see, noticing your own mismatches is not enough. So you run their loop inside your own: how do they see this, what are they expecting, where is their picture wrong. Boyd called this strategic empathy. Get it right and you can spot a trap before you step in it, because you can see what it is built to make you do, and you can move in a way they did not plan for. It works both ways: against an opponent it finds the assumption their plan rests on, so you can take it away; in a negotiation it finds the deal, because you can offer something they want before they have the words for it.
The loop sharpens you over time
None of this happens in one pass, and it does not happen on its own. Run the loop again and again, let each result land honestly, and your read keeps getting better: a picture that was too simple picks up the nuance it was missing, a belief you held more out of loyalty than evidence gets tested and dropped. Run the same loop while rationalizing every result that does not suit you and it does the opposite: it just makes you surer of a wrong read. The loop sharpens you only as much as you let reality correct you. Done that way, over many cycles your read stops being a fixed opinion and becomes a working model that closes the gap with how things actually are. That is the difference between someone who has an opinion and someone who has been around the loop a hundred times and listened each time.
How fast should the loop run?
Here is the part the four letters hide. A loop has a speed, and the right speed is not as fast as possible.
Speed follows from your read. When your read is clear you can act almost at once, the way you brake without deliberating. When it is not, you need the slower, fuller pass, and forcing a speed you have not earned is just reckless.
Tempo means how quickly you cycle compared to whatever you are adapting to. It is relative, not absolute. The familiar lesson is that the faster side wins. If you cycle faster than the other side can re-orient, each move you make lands before they have finished making sense of the last one, and they end up responding to a situation that no longer exists. Boyd had a name for doing this on purpose: getting inside the other side's loop. That is what happened in the dogfight at the top of this page: the F-86 pilot got a beat ahead and stayed there.
That much is true, and it is also where most people stop, which is the mistake. Speed has a price. A loop you run faster is a loop you thought less inside. Every step you take quickly is analysis you chose not to do. So there is a real trade-off here, and the honest version of OODA is about getting it right, not about always being the fastest.
Run too slow for the situation and the world moves on without you. Your careful conclusion ends up describing a situation that has already changed, and a large plan can be out of date before it is finished. Run too fast for the situation and you act on a shallow read, swing between half-formed moves, and step past the point where reality could have corrected you. Going faster than you need to does not make you sharper. It just makes mistakes you did not have to make.
What sets the right speed is the situation, not your preference. The question to ask is how fast the thing in front of you is actually changing, and whether anyone is adapting against you. When they are, you do not set the clock alone: the other side is pushing the tempo too, and sometimes a deadline or an outside force sets it for both of you, so the rate the situation is really asking for is not always obvious. A decision with a five-year horizon and nobody working to outmanoeuvre you should be taken slowly and carefully. Nothing is moving the board while you think, and the expensive error there is a shallow read, not a slow one. A negotiation where the other side is moving, or a problem where an adversary is shaping what you see, sets a faster pace, and there a months-long analysis is just a careful answer to a question that has already changed. The skill is reading that clock and running at exactly the tempo it asks for.
This is also a big part of why large, intricate plans tend to fail, and it is not that they are slow. A big plan commits a large, hard-to-reverse action on a single read that nothing has tested yet, and every assumption inside it keeps decaying while it runs. The way to buy speed and analysis at the same time is to make your actions small and reversible. Each one is cheap, each one lets reality correct your read before you have bet much on it, and ten small actions test ten assumptions in the time one big plan tests none.
A fluent AI makes the opposite easy: it is very good at helping a plan feel more solid, adding detail and answering objections, and that is not the same as the plan being more tested. The polish is not contact with reality.
Why it has to be trained
None of this is complicated to read. It is hard to do, and the two are not the same thing. Think about when you actually reach for something like this. Maybe you are stressed, with too much riding on the outcome to think calmly, or so frustrated you cannot think straight. Maybe you are raw and angry after being treated unfairly, or staring at a problem you do not know the first thing about. Those are the moments that need the loop, and they are the same moments that make it hardest to run, because a steady read and the right tempo are the first things to go when you are stressed, furious, or out of your depth.
That is why being told the steps does not transfer. A calm room can teach you the four words, but it has taken away the very pressure the habit exists to handle, so the habit never takes hold. The skill has to be built while you are in one of those states, on a real problem with real stakes, or it will not be there when you need it.
From one loop to missions and campaigns
A single loop is the smallest unit. Real work is made of loops put together.
They nest: a fast loop handling the immediate piece of a problem runs inside a slower loop that holds the larger aim, so you can react to what is in front of you without losing the thread of where you are going. And they sequence: what you learn in one loop sets up the next. A mission is a structure of loops built on purpose. A campaign is a line of missions pointed at something larger than any one of them.
The tempo trade-off comes back here as structure. A campaign runs loops at different speeds at the same time: a slow outer loop turning over the long aim, faster loops spinning inside it for the moves in front of you, each kept at the speed its own clock asks for. Part of the craft is not letting the fast loops rattle the slow one into thrashing, and not letting the slow one starve the fast ones of the reactivity they need.
This is the layered version of the same skill, and it is not something you pick up in an afternoon.
Missionloops is where you practise this on your own real problems. You start with one loop on one problem you actually have, and the harder compositions come as you are ready for them. Over time you need the software less, not more: the mindset and the hundreds of small skills underneath it get drilled in until they are yours, and the tool fades into the background. The member page explains how it works, and you can reach us at scott (at) missionloops (dot) ca.