BELL CURVE

Hang around airports long enough and you’re bound to hear talk about newly rated pilots becoming more dangerous after 200 hours — and then less dangerous after 500. The implication is, those who survive this cocky phase become truly safe at last. My own empirical evidence as a flight instructor supports such a notion more than it doesn’t, but what do I know? Wondering if there were actual evidence, I cranked up the google machine and quickly found that, like so much of what ‘they’ say, the legend rings true… Half true.

The government publication DOT/FAA/AM-15/3 obviously was written by statisticians for statisticians and is spectacularly unreadable, but it does include graphs for assorted categories of accident probability that reveal striking similarity across the board. This one sample is representative of the whole set.

screen-shot-2019-08-14-at-10-41-59-am

In each category, accidents within pilots’ first 100 hours are essentially zero, but increase manyfold by 200 hours, and continue to increase even more rapidly to 300. After that the increase continues at a slower rate until indeed peaking at 500 hours — but in every category the bell curve descends much slower toward the high-time side of the graph.

Several factors influence this data in opposing ways. It’s reasonable to expect that pilots with five hundred hours are more at risk than those with five thousand. But common sense also dictates the longer anyone flies the greater their chance of an inevitable mishap. From the perspective of DOT/FAA/AM-15/3, “…there are simply fewer pilots at higher total flight hours, hence fewer accidents.” Who knows to what extent low-time accidents eliminate less-safe pilots from the data pool? The authors grapple with with these and other potential contradictions by analyzing data in multiple ways, and not surprisingly, draw differing conclusions about what they call the ‘killing zone’ of heightened risk.

From a maze of exotic formulas and statistical jargon, they emerge with two deductions that carry the same message even as they disagree. One defines a killing zone from 50-350 hours, and the other from 100-2000. Either way, the authors note that “relatively high risk for an individual pilot may extend well beyond the 2,000-hour mark before leveling off to a baseline rate.” In other words, ordinary pilots with less than 5000 hours spend most of that time in the killing zone — and it takes much longer to leave the zone than to enter it.

Look again at the graph. There’s still data dribbling in far out to the right… I personally know of more old-timers than I care to count, who flew well into the five-digit side of this graph accident free, then took to the air, regrettably, just once too often. It’s what life in the air has in common with life on the highway: the killing zone is everywhere all the time, unless we do whatever’s necessary THIS TIME to make it otherwise…