when not to stall

 Out of 250 that took place during instructional flights between 2000 and 2014, nearly two-thirds (166) occurred on dual lessons—and those were twice as likely to be fatal as stall accidents on solo training flights, 46 percent versus 23. Perhaps just as counterintuitive is the fact that fatalities were rarer in accidents involving primary students than those in which the pilot under instruction already held a certificate, regardless of whether the flight was dual or solo. In fact, the lethality of stall accidents was lowest on authorized solos by student pilots at 17 percent—and highest during advanced dual instruction, where it reached 52 percent.

If you’re still skeptical that post-solo students are safer on their own than flying with your CFIs, we applaud your common sense. Percentages are extraordinarily slippery things, perhaps second only to pie charts in their potential to mislead the public. It always pays to ask: Exactly which is a percentage of what?

Digging into the data a little further, we see that student solos accounted for 44 percent of all stall accidents during primary training. We’re willing to bet that solos don’t make up anything close to 44 percent of students’ flight time—and more than half those crashes happened during landing attempts or go-arounds, with no fatalities in either. (Another 20 percent, including one-third of the fatal accidents, were departure stalls.) With a CFI on board and/or a certificated pilot receiving instruction, landings and go-arounds only account for about one-quarter of stall accidents. So any intuition that students are susceptible to low-altitude, low-energy upsets that more experienced pilots usually avoid has some grounding in reality.

Further support comes from the fact that during dual instruction, takeoffs are the setting for 30 percent of stall accidents in primary training but just 12 percent in advanced. In both cases this is among the more lethal categories, with fatalities in more than one-third; survivability is largely a function of the altitude at which the stall breaks. A mush back onto the runway after an overly aggressive attempt to climb out of ground effect inflicts far milder deceleration forces than a nose-low impact from 50 or 100 feet above ground level.

Solo students rarely crash while practicing maneuvers, partly because the maneuvers they’re authorized to practice solo are relatively benign, and partly (we suspect) because many choose not to spend much of their solo time practicing them. On dual lessons and during solos by certificated pilots, however, ASI’s “maneuvering” category—which includes everything from spin training or eights on pylons to ordinary turns in the traffic pattern—was consistently the most lethal, with fatalities in more than half. It accounted for more than a third of fatal stalls during primary dual, half of those during advanced dual, and 70 percent during advanced solo airwork. Notably—and as we’ve reported before—very few occurred while deliberately practicing stalls (though one private pilot candidate and a designated pilot examiner were killed when a Cessna 172 apparently failed to recover from a stall during a checkride). Rather, the context was typically situations in which concentration on meeting exacting test standards drew attention away from airspeed, attitude, and angle of attack.

In that light, it’s not surprising that the single largest number—half of all fatal maneuvering stalls during advanced instruction, for example—occurred while practicing emergency procedures, most particularly simulated engine failures. Twin-engine aircraft were perhaps even more vulnerable than singles: Nearly 40 percent were unrecoverable spins during single-engine work in twins, presumably after airspeed was allowed to decay below V_MC. In singles, attempts to reach the chosen landing spot lead to overly aggressive turns and efforts to stretch a glide even as altitude and the potential to recover dwindle. Perhaps fixed-wing instructors should guard the controls as closely as helicopter CFIs do during practice autorotations—and be quicker to call for go-arounds once things no longer seem to be working out quite right.