Picture the moment. You have just rotated off runway 07 at Wilson, gear and flaps clean, nose up, climbing through 350ft, and the engine note changes. Not a splutter, a genuine loss of power, and the aircraft that was climbing is now sinking toward a city.
The instinct that fires first in almost every pilot is the same one: turn back, land where you took off from, where you know the surface and the length. That instinct is the thing most likely to kill you, and the data on why is not close.
The Short Version
- An engine failure after takeoff (EFATO) below about 1,000ft AGL is rarely survivable if you try to turn back to the runway, and it is highly survivable if you fly the aircraft straight ahead to a controlled landing.
- An NTSB review of light-aircraft turnback attempts from 500ft found straight-ahead landings succeeded 100 percent of the time and turnbacks succeeded only 62 percent of the time, with most failures ending in a stall/spin.
- The FAA's own maths on the "impossible turn" shows why: the turn itself takes longer and costs more altitude than instinct suggests.
- Wilson's elevation of roughly 5,536ft and Kenya's highland strips reduce climb performance, which means less altitude for the same distance down the runway when the engine quits.
- The fix is not a better turn. It is a decision made and briefed before every single takeoff, so there is no decision left to make when the engine actually quits.
What EFATO actually is
Engine failure after takeoff, EFATO, is not the same emergency as an engine failure in the cruise. In the cruise you typically have altitude, options, and time to think.
On climbout you have almost none of the three. You are slow, close to the ground, in a nose-up attitude that will bleed airspeed fast if you do not lower the nose immediately, and the ground below you is whatever happened to be under the departure path.
The FAA's Airplane Flying Handbook, Chapter 18, treats this as its own distinct emergency for exactly that reason. Its guidance for a total power loss immediately after takeoff is blunt: lower the nose to maintain flying speed first, then look for the most suitable landing area within your gliding distance, and land essentially straight ahead with only small changes of direction to avoid obstacles.
Straight ahead does not mean the exact extended centreline. It means the arc of terrain roughly in front of you, not a manoeuvre that turns your flight path back the way you came.
The impossible turn, and why it is called that
Pilots have been arguing about the 180-degree turnback since powered flight began, and the FAA eventually put out a safety pamphlet, FAA-P-8740-44, The Impossible Turn, specifically to settle the maths.
A genuine 180-degree turn back onto the runway needs more than 180 degrees of heading change, because you also have to correct back onto the extended centreline from wherever the turn placed you. The pamphlet works through a standard-rate turn plus the extra heading change needed to re-intercept, and gets a total manoeuvre time in the region of 80 seconds.
During those seconds a typical light aircraft gliding at a normal descent rate loses far more altitude than pilots instinctively expect, easily over 1,000ft in that time. The FAA's own conclusion, worked through their example starting points, is stark: below roughly 600ft AGL you are gambling, and even at 800ft things are deteriorating fast.
That is before you account for the startle response, the extra seconds most pilots lose simply processing that the engine has actually failed before they even begin the manoeuvre.
What the accident data actually says
The FAA's maths is theory. The accident record is what happened when real pilots tried it.
An NTSB General Aviation Crashworthiness Project review, examining 535 light-aircraft accidents involving 1,267 occupants, specifically studied the outcomes of engine failures at 500ft AGL and below. Pilots who flew straight ahead and accepted whatever was in front of them landed under control 100 percent of the time, 35 out of 35 cases.
Pilots who attempted the turnback succeeded only 62 percent of the time, 69 out of 112 cases. Nearly two out of every five turnback attempts failed, and the review found the majority of those failures ended in a stall/spin, which is close to unsurvivable at that altitude.
AOPA's Air Safety Institute has reached the same conclusion from a different direction, and it is worth being precise about what their research actually says. It is not that a turnback is always impossible.
Their "Reality Check" flight testing found the manoeuvre is nearly impossible in aircraft with high glide speeds and high wing loading, and can be flown successfully most of the time in lighter trainer-type aircraft, but only with sufficient altitude, a briefed plan, and prior practice with a Certified Flight Instructor (CFI). Without those three things in place before the engine quits, the odds collapse back toward the NTSB numbers above.
Sitting alongside both of those studies, AOPA's own accident database work puts the fatality rate for a forced landing where the pilot kept the aircraft under control at roughly 10 percent. That means roughly 9 out of 10 pilots who maintain control all the way to the ground, wherever the ground turns out to be, walk away.
The single biggest variable separating those pilots from the ones who do not survive is not the failure itself. It is whether they kept flying the aircraft instead of trying to make it go somewhere it could not reach.
Kenya's altitude makes the trap tighter
Every number above was generated by pilots flying from airports far lower than most of Kenya's training fleet operates from, and that matters more than it sounds like it should.
Wilson sits at an elevation of roughly 5,536ft AMSL. Its main tarmac runway, 07/25, runs about 1,530m, with the shorter 14/32 used when wind or congestion requires it, both short by the standards of a major international field.
Higher field elevation means lower air density, which means a naturally aspirated C152 or C172 climbs noticeably worse off Wilson than the same aircraft would off a sea-level strip, before you even add a hot Nairobi afternoon and the density altitude penalty that comes with it. Our piece on density altitude in Kenya's highlands covers the full mechanics of why that penalty bites; the relevant point here is narrower and sharper.
A degraded climb rate means less altitude gained for the same distance travelled down the departure track. If the FAA's maths already shows the turnback needs altitude the aircraft has not yet had time to build at sea level, a highland strip gives you that altitude even later. Nanyuki, Eldoret, and Kericho sit higher again, and the same arithmetic gets worse, not better, at every one of them.
None of this changes the manoeuvre's physics. It changes how much runway and time you actually get before you cross the altitude where a turnback becomes a coin flip, and in Kenya that point arrives sooner than pilots trained on sea-level numbers tend to expect.
The briefing that replaces the decision
The pilots who survive an EFATO are, almost without exception, not the ones who made a brilliant decision at 400ft. They are the ones who had already made the decision on the ground, before the takeoff roll even started, so there was nothing left to decide in the two or three seconds they actually had.
A pre-takeoff safety brief, said out loud, every single flight, does that work in advance. It does not need to be elaborate.
- Below [your aircraft's briefed decision altitude, typically 500 to 800ft AGL for a light single]: land straight ahead, small heading changes only, gear down and flaps as configuration allows.
- Above that altitude, with the numbers actually in hand: a turnback may be an option, but only if you have practised it at height with an instructor and know your aircraft's real numbers, not a guess.
- In any case: fly the aircraft first. Airspeed before altitude before anything else, every time.
Say the altitude number aloud on every single takeoff roll, not just the ones where something feels risky. The habit is worthless if it only shows up on the day you remember to think about it, because that is exactly the day your brain will be too busy processing the actual failure to invent a plan from scratch.
Straight ahead is not giving up
Here is the one opinion in this piece and it is an earned one: too much flight training frames the straight-ahead landing as the consolation prize, the thing you do when the "real" option, the clever turn back to the tarmac, is not available.
That framing is backwards and it gets pilots killed. The straight-ahead landing is not the fallback. It is the option with a 100 percent success rate in the actual accident data cited above, against a manoeuvre that fails nearly two times in five.
An off-field landing under control, at flying speed, wings level, onto Kenyan farmland, a murram road, or open ground near the departure end, is not a crash. It is a successful application of the emergency procedure, and it is the outcome every EFATO briefing should be aimed at producing, not a backup plan for when the turnback does not seem safe enough.
What instructors owe this decision
A student's only real exposure to this scenario is usually a single "simulated engine failure" pulled at a safe altitude, briefed well in advance, in calm conditions with an instructor already anticipating the reaction.
That is nothing like the real event, and every CFI knows it. What actually builds the reflex is repetition of the pre-takeoff brief itself, said aloud on every departure until it is as automatic as the pre-landing checks, alongside honest, altitude-safe practice of the straight-ahead glide and landing configuration rather than only the turnback.
Debrief students who skip the verbal brief the same way you would debrief a missed checklist item. The habit that saves a life is built on ordinary training days, not invented from memory on the one day it matters.
What pilots admit on the forums
EFATO threads on pilot forums have a recognisable shape once you have read enough of them. Very few pilots claim they would calmly execute a perfect turnback if it happened tomorrow.
The more common admission is smaller and more useful: pilots who say they only started taking the pre-takeoff brief seriously after they heard about, or were close to, an accident that could have gone the other way with one different decision. The other repeated thread is instructors saying the students who freeze hardest are the ones who never briefed the failure out loud, because their first real exposure to the decision came at the worst possible moment to be improvising one.
Neither pattern is dramatic. Both point at the same fix: say the brief, every flight, until it needs no thought at all.
AngaBrief's Aircraft and Environment sections prompt a pilot to record performance-affecting conditions, including field elevation and density altitude, before every flight, but the tool does not fly the aircraft and it does not make the emergency decision for you. AngaBrief is not a dispatch authority, and the go/no-go call, in the climb-out as much as on the ramp, always rests with the pilot in command and their instructor.
Key Takeaways
- Below roughly 600 to 800ft AGL, a turnback to the runway after engine failure is not a safer option than landing straight ahead. NTSB data shows straight-ahead landings succeeding 100 percent of the time against 62 percent for turnbacks.
- The FAA's own "impossible turn" maths shows the manoeuvre needs more altitude and more time than instinct suggests, roughly 80 seconds for the full turn and re-intercept.
- Wilson's elevation of about 5,546ft and Kenya's highland strips reduce climb performance, meaning less altitude is available at any given point down the departure track than sea-level numbers assume.
- Decide and brief your EFATO plan out loud on every takeoff, not just the ones that feel risky, so there is no decision left to make when the engine actually fails.
- A controlled straight-ahead landing is not a fallback. It is the option with the strongest survival record in the actual accident data.