Ask yourself a question: where do most stalls occur? Take a moment. Write down your answer.
Almost everyone probably wrote down “in the base-to-final turn.” The ubiquitous stall scenario is overshooting the turn from base leg to final approach, and (perhaps subconsciously) adding too much rudder to try to slew the airplane’s nose into alignment with the runway centerline in a skidding turn.
The resulting overbanking tendency may incite the pilot to apply aileron opposite the turn. The upward deflected aileron on the wing outside the turn decreases that’s wing’s angle of attack compared to the wing inside the turn. If the pilot also pulls back on the elevator control in this turn—another instinctive response to an overshoot—the inside wing may reach its critical angle of attack. It suddenly stalls while the outside wing is near its maximum coefficient of lift. The airplane snap-rolls toward the inside of the turn with nowhere near enough altitude for the startled pilot to recover.
A base-to-final turn gone bad is a deadly Loss of Control – Inflight (LOC-I) scenario. However, LOC-I in the base-to-final turn is one of the least common stalls in the accident record.
The truth about stalls was quantified by AOPA’s Air Safety Institute in a 2017 study titled “Stall and Spin Accidents: Keep the Wings Flying.” This report “analyzes 2,015 stall accidents between 2000 and 2014, and concludes with recommendations for prevention, recognition, and recovery from stalls while offering ideas on a shift in focus for stall awareness, prevention, and recovery.”
Using the AOPA-ASI data, which in turn derives from NTSB conclusions, I created some images that describe the true nature of traffic pattern stalls.
The first image details real-world stall data on the arrival end of a visual traffic pattern. The commonly cited base-to-final turn, and stalls in the turn from downwind to base leg, together account for only 3.8% of all NTSB-reported stall events. Now these stalls, when they do occur, are quite deadly: 66% of the downwind-to-base stalls are fatal, and 80% of base-to-final turn stalls result in death. That stands to reason; if a stall occurs in one of these places there is little room to recover. Still, these most commonly considered turns are low-probability, high-severity events.
Stalls on the downwind leg or the wings-level portion of the base leg almost never occur, only 0.8% of the reported LOC-I events in the circuit. A little over half of these resulted in death, still a low-probability, high-severity event.
Stalls after completing the turn to final approach are almost twice as common as stalls in the turns. Still, they account for only 6.1% of traffic pattern stalls, 40% of them fatal. This becomes a low probability but moderate severity type of event.
Stalls in the landing flare are much more common than any of the others on the arrival end of the pattern: 21.2% of the pattern stalls total. Close to the ground, these stalls usually do not devolve into spin rotation, and vertical movement stops before the airplane accelerates to a deadly descent. We call these stalls a hard landing—only 8% of stalls in the flare kill people. These are high probability but relatively low severity events.
Put them all together and LOC-I in a visual arrival account for 31.9% of all traffic pattern stalls. Another commonality: these are generally power-off stalls, the type most pilots and their instructors are far more comfortable practicing and tend to practice more often.
This second image plots AOPA-analyzed NTSB data to show stalls during a go-around and during the initial climb. This is the surprising part: takeoff and go-around stalls, power-on stalls, are far more common than power-off stalls during the approach and landing. About 18% of the reported stalls happened during a go-around. Because these LOC-I events are close to the ground, a quarter of these stalls are fatal…but three-quarters of them are not.
Many types of airplanes, when trimmed for final approach speed, have an elevator trim setting that is more nose-high than the takeoff trim setting. Some types are trimmed very nose high on final approach. Meanwhile, in many airplanes adding power causes an upward pitch movement.
So at the beginning of a go-around, many airplanes will pitch up into a high angle of attack. It may take forward pressure on the controls to fly the correct initial attitude and airspeed. Pilots who do not practice go-arounds routinely may not be prepared for the control inputs necessary to avoid a stall.
However, half of all traffic pattern stalls happen during takeoff and initial climb. 40% of these losses of control prove fatal. These are high probability, moderate severity events.
The major commonality here: these are power-on stalls. You know, the ones that are uncomfortable to fly, and may seem unrealistic is flown the way they are prescribed in the Airman Certification Standards. For the stalls that will get you, full power stalls during takeoff or a go-around, are often flown with some flaps and with (in retractable gear airplanes) gear extended.
Such an airplane, combined with nose-up trim, may reach the critical angle of attack at a pitch attitude much lower than is required to fly an ACS-style power-on stall (pg. 43). The “dirty” airplane configuration often results in a more dynamic, more dramatic departure from controlled flight than a 
clean, ACS-style power-on stall. And full power adds to the rapid departure from controlled flight, compared to the often reduced-power power-on stall taught at altitude—power application can introduce yaw and roll, and countering that movement with aileron (a common response) sets the pilot up for that same skidding-stall scenario we discussed back in the turn from base to final.
My third diagram interpreting AOPA’s report compares where we think we’ll stall to where we actually stall, based on NTSB accident history:
About half of all NTSB-reportable stalls are power-on stalls during takeoff and in a go-around. Almost 90% of all stalls—add the hard landings to the power-on stalls—happen over or beyond the runway. We think if we’re going to stall it will be in the pattern before the final approach. We actually stall over the runway and on the departure end.
We spend a lot of time and effort teaching the power-off stall, avoiding accelerating the stall (pulling back on the controls, which increases G load and therefore angle of attack for a given pitch attitude) and emphasizing rudder coordination to keep both wings at the same angle of attack, avoiding the snap-roll scenario. This emphasis may be why the most commonly cited stall scenario, the base-to-final turn, is in reality one of the least common stalls in the accident report. Don’t stop training, practicing and thinking about these stalls. Training and awareness work.
We need to add training and awareness of stalls that occur over and beyond the runway, and practice realistic simulations of a power-on stall in the landing and takeoff configurations, to guard against the most common stalls. Get as comfortable recognizing and recovering from these stall scenarios as you are with power-off stalls more commonly practiced, to avoid the traffic pattern loss of control threat.
Professional CFIs should take special notice of the *real* pattern threat – high power/high nose – and train these areas more assiduously for pilot proficiency and confidence. Fly SAFE out there (and often)!
Thanks to Tom Turner, a charter (and lifetime) member of SAFE, for sharing this important article here. This was originally published on Tom’s Mastery Of Flight Training website which publishes weekly “Flying Lessons” (subscribe for free). Tom is also the author of many books and articles as well as Executive Director and Chief Pilot of the American Bonanza Society.
Join SAFE to support our safety mission of generating aviation excellence in teaching and flying. Our amazing member benefits pay back your contribution (1/3 off your ForeFlight subscription)! Our FREE SAFE Toolkit App puts required pilot endorsements and experience requirements right on your smartphone and facilitates CFI+DPE teamwork. Our CFI insurance was developed by SAFE specifically for CFIs (and is the best value in the business).
The Captain agreed, and handed control of the aircraft to the First Officer (FO). The Captain (CA) then began coordinating with company dispatchers over the radio while maintaining contact with the FA to track the passenger’s condition. The FO assumed “single pilot” control of the jetliner, running checklists, setting up for a visual approach and “working the radios” with Air Traffic Control (ATC), and flying the airplane—alone—while the CA continued to monitor the medical emergency and relay information to company dispatchers.
This one I probably don’t have to explain. I’m talking about the “Miracle on the Hudson,” as described in the recent theatrical release Sully. I was able to see the film recently, and am working on the assumption it accurately portrays at least the inflight portion of the story it tells (I suspect some other aspects of the story were embellished to heighten tension, notably the conduct of the NTSB investigation and hearing, which likely was treated with some license—a movie needs an antagonist, and there’s no tension from blaming the birds).