The DECIDE Model, the Six-Letter Loop the FAA Built to Slow You Down When a Flight Starts Going Sideways

The DECIDE model is the FAA’s six-step decision-making loop for pilots: Detect, Estimate, Choose, Identify, Do, Evaluate. It exists to interrupt your mental autopilot when a flight starts drifting away from the plan, forcing you to notice a change, judge whether it matters, and act before conditions decide for you. Its real purpose isn’t passing a checkride - it’s breaking the chain of small, reasonable decisions that leads most trained, current, legal pilots into accidents.

What problem does the DECIDE model solve?

Most accidents that kill trained, current, legal pilots are not stick-and-rudder accidents. The pilot didn’t forget how to fly. The pilot made a chain of small decisions, each one reasonable on its own, and the chain led somewhere terrible.

The FAA has spent decades studying these accidents, and the pattern is almost always the same: nobody chose the crash. They chose to keep going - over and over, they chose to keep going.

So the FAA published a set of tools for aeronautical decision making (ADM), and DECIDE is one of them. It’s a loop of six steps, and the whole point is to force you to actually look at what’s in front of you instead of running on the mental autopilot that’s fast, mostly right, and terrible at noticing when the situation has changed underneath you.

What do the six letters in DECIDE stand for?

• D - Detect: You detect that something has changed.
• E - Estimate: You estimate whether that change needs a response.
• C - Choose: You choose an outcome - what you actually want to happen.
• I - Identify: You identify the actions available to get there.
• D - Do: You act.
• E - Evaluate: You evaluate the effect of what you did - and then the loop starts over.

Detect, Estimate, Choose, Identify, Do, Evaluate - and then around again. That final step is what makes DECIDE a loop rather than a checklist.

How does the DECIDE model work in a real flight?

Picture a Saturday. Not a dramatic day - the days that kill people almost never look dramatic at the start.

You’re a fresh private pilot, maybe 80 hours total. You’ve planned a cross-country in a Cessna 172 to a little airport about 90 miles away for a good breakfast. High pressure, light winds. Your preflight weather showed a chance of afternoon buildups back home, but you plan to be on the ground eating pancakes before any of that.

You launch. Everything’s perfect for the first 45 minutes. Then you notice the visibility ahead getting soft. The horizon’s gone a bit milky. Not clouds exactly - just hazy, less sharp than it was. Here’s how each step plays out from that moment.

Detect: name the change out loud

The hardest part of Detect is not analyzing the change - it’s admitting the change is happening at all, because you have a plan and your brain loves the plan.

The force fighting you here is plan continuation bias, sometimes called get-there-itis: the pull to keep executing the plan you already made even as the world quietly stops matching it.

The first skill, before any acronym, is to notice. Say it out loud in the cockpit even though nobody’s listening: “Visibility is dropping.” Naming the change pulls it out of your blind spot and puts it on the table where you can work on it.

Estimate: gather information, then judge the trend

Does this change need a response? Look at your instruments, look outside, dial up the nearest automated weather. This is where you gather information instead of feeling your way forward.

Say you learn the visibility at your destination has dropped to about four miles in haze and is trending down. Ceilings are still fine, but the trend line is the story - it’s not where you are, it’s where things are going. Haze thickening on a day with afternoon moisture in the forecast is a something, not a nothing.

Choose: name the outcome you actually want

This is the most underrated step, because most pilots skip it entirely and jump straight from “uh oh” to “what buttons do I push.” Choose comes first.

The outcome you want is simple: on the ground, at an airport, with the airplane and everybody in it undamaged, before the weather forces your hand. Not pancakes. The pancakes were never the outcome - somewhere in the last 45 minutes they quietly promoted themselves to the mission, and part of choosing is demoting them back to optional.

Your goal in flight is almost never the thing on the schedule. Your goal is a good outcome. The schedule is a suggestion that the weather, the airplane, and your own fatigue all get a vote on.

Identify: count all the doors, not just two

Now you get tactical. What can you actually do?

• Press on to the destination and hope the haze holds at four miles.
• Turn around and go back the way you came, where the weather was good 45 minutes ago.
• Divert to the airport about 15 miles off to your right that you passed, which was reporting clear.
• Get help: climb, descend, call flight service, or pick up flight following.

When plan continuation bias has you in its grip, it whispers that there are only two choices - keep going or crash - and keeping going always sounds better than crashing. That’s a lie. There are almost always more than two doors. Identify forces you to walk down the hallway and count them.

Do: commit decisively while you still have room

Pick one and act while you still have the room to do it. Say you choose the airport off to your right that’s reporting clear. You don’t dither - you make the radio call, point the nose, and start setting up.

Doing is a real skill. Some pilots are great at the first four steps and then just sit there admiring their analysis while the window closes. A decent decision executed now beats a perfect decision executed too late - every single time.

Evaluate: check whether it’s working, then loop again

You turned toward that airport. Is it working? Is the visibility actually better in that direction, or did you trade one bad situation for another? Is that field really still clear, or was the report 20 minutes old?

This is the step that makes DECIDE a loop instead of a line. When you Evaluate, you might Detect a brand-new change, and around you go. Maybe the haze is worse than advertised - fine. Detect. Estimate. Choose. Identify. Do. Evaluate. The loop doesn’t care how many times you run it. The pilots who get into trouble are usually the ones who ran the loop zero times and just flew the original plan into the ground.

Will the DECIDE model be on my checkride?

Yes. The Airman Certification Standards (ACS) weaves aeronautical decision making and risk management all the way through - it is not a bonus round.

When the examiner hands you a scenario - “You’re halfway to your destination and the ceilings are coming down, what are you going to do?” - they are not testing whether you can recite the letters. They’re testing whether your brain actually works this way under a little pressure.

The examiner is looking for a pilot who detects the change out loud, gathers information instead of guessing, names a safe outcome, lays out real options, commits to one, and keeps re-checking. Narrate that - let them hear you think - and you’ll look like exactly the pilot they want to sign off. It’s not a trick you perform for them; it’s the habit that keeps you alive after they’ve gone home.

How do I actually train the DECIDE model?

In the real cockpit, you will not think in six neat letters. It won’t feel like a flowchart. It feels like your stomach tightening and a little voice saying “something’s not right here.”

The DECIDE model is not meant to be recited in the moment. It’s meant to be trained in the calm, so that when your stomach tightens, your hands already know the shape of the response. You practice it in the chair, on the drive, in the debrief, so that in the airplane it runs mostly on its own. These acronyms aren’t scripts for the emergency - they’re rehearsals for it.

The single most useful habit is to set your bail-out points on the ground, before you start the engine. Say them out loud: “If visibility drops below this, I divert here. If I’m not on the ground by this time, I’m not going.” You make the hard decision once, in the calm, when it’s easy - so the airplane doesn’t get to make it for you later, when it’s hard.

Stick and rudder gets you your certificate. Judgment is what lets you use it for the next 40 years - and decision-making is genuinely the hardest part of flying, harder than crosswind landings, because you can’t see it and nobody can grade it from the ground.

Key Takeaways

• DECIDE stands for Detect, Estimate, Choose, Identify, Do, Evaluate - a continuous FAA loop for aeronautical decision making, not a one-time checklist.
• Detect is the hardest step because plan continuation bias (get-there-itis) makes you want to ignore the change; naming it out loud defeats the blind spot.
• Choose your outcome before your actions - a safe landing, not the mission on the schedule - because most pilots skip straight to tactics.
• There are almost always more than two options. Identify forces you to count every door, not just “press on or crash.”
• Set your bail-out points on the ground before engine start - the cheapest, most effective decision you can make all day.