Human Factors: A Lethal Game of Chance

Risk is an element of flying we often either ignore or mentally mitigate. Here is an experienced pilot who failed to come to grips with reality.


Would you like to play a game of Russian Roulette? To make it more interesting, what if we load five of the six chambers. If you think those odds are poor, then consider that flying VFR into IMC conditions are about the same.

Too often many general aviation pilots place themselves in IMC situations and then fail when it comes to Aeronautical Decision Making (ADM)? Over the years, the lethality of an IMC accident has had a similar rate as the deadly game above—which consistently hovers around 80-percent.

The overall general aviation accident rate is down. But, as a group, we need to step up our ADM game whenever weather is involved. Here is an example of a straightforward IMC accident with ADM being the root cause.

The Aviator

The 35-year-old pilot held a commercial certificate with single- and multi-engine land ratings, with an instrument rating. He also held a mechanic’s certificate with airplane and powerplant ratings with inspector authorization.

The NTSB reviewed the pilot’s logbooks and determined that during a seven-year period prior to the accident, he had flown 700.1 total hours of which 648.8 were as pilot in command. These hours were further broken down as 498.3 hours in single-engine airplanes; 205.2 hours in multi-engine airplanes; 577.4 hours of cross-country time; 95.6 hours of night flight; 60.5 hours of dual; and 33.3 hours in the make and model accident aircraft, a 1964 Cessna 182G.

Additionally, the NTSB noted the pilot flew 45 hours of actual IMC and 11 instrument approaches during the six months before the accident. One would have to assume he was instrument proficient.

However, investigators questioned if the pilot had a current flight review. They were only able to find evidence for a single such flight being logged during the five-year period preceding the accident—and this entry was over four years old.

Nor did the NTSB note the pilot having added a flight certificate or rating within the two years prior to the accident either of which would have fulfilled the requirement for a recent flight review. This leads us to believe the pilot was flying without a current flight review—raising questions about his judgment and decision-making ability.

Las Vegas Trip

The pilot was the principal in an aviation business located at Kickapoo Downtown Airport (KCWC), in Wichita Falls, Texas. According to a company employee, the pilot, a fellow business partner and the business partner’s wife departed at about 10:00 a.m. local on a November flight to Las Vegas, Nevada—a straight-line distance of over 800 miles.

According to investigators, there was no record of a weather briefing for the flight. But, with today’s internet, it’s not uncommon for a pilot to self-brief, sometimes with sites that don’t record the pilot’s access. However, considering the flight’s distance relative to the pilot’s reported experience, it would have been prudent to go further by speaking to a flight service briefer or getting a formal online briefing. This is especially true since the distance and the time involved would likely have the Cessna crossing areas of IMC and/ or other adverse weather conditions.

Moreover, no flight plan was filed. While not a legal requirement for a VFR flight, it would be wise for ATC to have a record of one’s basic information and intentions for a trip of this length, especially over desolate terrain. At the least, a request for flight following services should have been in order. Having another set of eyes and ears to assist with weather and traffic alerts is an added safety net.

Instead, the aircraft departed on its journey with the transponder set in Mode A, instead of in Mode C, so no altitudes were captured by ATC. The decision to set the transponder to Mode A was likely spurred by the Cessna not having the required pitot static system and transponder-encoder tests in over three years.

At this point, we need to seriously question this pilot’s judgement and his ability to follow basic aviation rules.

The Weather Encounter

The passenger’s camera recorded the undercast as it began to thicken.

A photograph taken by a passenger in flight shows the airplane traversing desolate mountainous terrain in VFR conditions below a cloud deck with weather in the distance. Another photo shows the airplane flying over a solid undercast. It could not be determined where the photos were taken. However, given the time of year, and the temperatures aloft (at the required terrain clearing altitudes) why would a pilot fly an aircraft, not certified for flight in icing conditions, over a solid undercast? Yet another indicator of questionable ADM. This action would become the catalyst that set the accident chain in motion.

As the airplane approached its planned midpoint stop, Double Eagle II airport (KAEG) in Albuquerque, New Mexico—near the Sandia Mountain range—it deviated from its westerly course by first turning southwest, then north. This was likely a reaction to a wide area of IMC weather with light precipitation near the destination. The icing severity and probability charts showed a potential for icing in the area starting at 7500 feet and increasing in severity and probability with altitude.

It was also the first time the pilot contacted any ATC facility during the flight. The pilot called the Double Eagle control tower, when 25 miles to the east, stating he was “descending out of 13,000 feet, trying to get over weather but we couldn’t get high enough to make it work, kind of in between layers.” The pilot requested an ILS approach into the airport, but erroneously reported his distance from KAEG as less than five miles. The pilot was given a frequency and was told to contact Albuquerque Approach for a short range IFR clearance.

Flight path (yellow line) and key locations are noted amongst the weather.

When the pilot radioed Albuquerque Approach, he again erred when reporting his position as five miles east of KAEG. This hampered the controller trying to find the aircraft on radar. It took some time for approach to find the Cessna, but with help from Albuquerque Center the aircraft was finally identified. It was in an area with a minimum en route altitude of 11,500 feet and having a more than 60-percent probability of light to moderate ice.

Due to terrain, radio communications with the pilot were poor, so the controller had to relay information through a passing air carrier flight. The information included an altimeter setting. But transmissions received by the controller were worrisome… “pretty hairy… I can see the ground… I’m just trying to maintain visibility right now.”

It’s obvious that stress and possibly hypoxia were taking their toll on the pilot’s decision-making ability and his basic attitude instrument flying skills. It was also the likely cause of the pilot not performing two basic actions—updating the altimeter setting and more importantly turning on the aircraft’s pitot heat. The NTSB noted the failure to set these two items when investigating the wreckage.

It is unknown, but ice was likely also accumulating on the airframe. The remainder of the flight was an erratic flight path indicative of spatial disorientation. There were no survivors after ground impact at an elevation of 7634 feet about 22 miles east of KAEG.

The NTSB determined the following probable cause: “The pilot’s continued visual flight into instrument meteorological conditions, which resulted in spatial disorientation and a loss of control.” The report went on to list additional factors, the first on the list under the heading Personnel Issues was: “Decision-making/judgment – Pilot (Cause).”

ADM And Risk Management

The extent of the worsening weather pattern is apparent.

The concepts of aeronautical decision making, and risk management go hand in hand. Unfortunately, they also get brushed aside with the common misconception that both processes are just plain old-fashioned common sense that can be done on the spur of the moment. Nothing can be further from the truth, as both these processes are systematic approaches to making decisions and managing risk. Both require forethought on the part of the pilot-in-command.

The FAA uses some trite mnemonics such as the 3Ps, DECIDE and IMSAFE for these models— that tend to make it easy to dismiss the concepts. But these memory aids are minimalist versions of robust decision making and risk management techniques used by the airlines, the military, and large corporations.

Consider scheduling time with an instructor (or a more experienced IFR aviator) to create a personal minimums worksheet and then run some flight scenarios that require decision making. Do this over some refreshments on a day when you can’t go fly—for pilots, talking aviation is the next best thing to flying.

Doing this on regular basis, especially before any significant trip, can only help. Incorporate this activity into a flying club or aviation organization meeting. No one has all the answers, and group discussions can help.

The idea is to keep ADM and risk management as our most important resource when flying rather than a chore to be considered when convenient. It needs to be spring loaded to help us make good decisions before we aviate—and while we fly.

Article is based on NTSB Number: CEN16FA042

Amand Vilches is a commercial pilot and instructor who lives in Brentwood, TN. He is a past Nashville District FAASTeam Honoree. His extensive background in risk management and insurance allows him to bring a unique perspective to flight instruction.

This article originally appeared in the November 2019 issue of IFR Refresher magazine.

For more great content like this, subscribe to IFR Refresher!


  1. Like many pilots I would have read this accident report and have said that ‘it couldn’t happen to me’. But before any of the aviation community do that, I would strongly recommend a review of some of the published work by Professor Sidney Dekker – particularly his book ‘The Field Guide to Understanding Human Error’ (currently in its 3rd edition). If you want the quick version type ‘Field Guide to Understanding Human Error’ into a well known search engine and then click ‘Videos’. There is a series of 5 videos which all together come to about just under one hour of viewing.

    Prof Dekker talks about the principle of ‘Local Rationality’ – the simple fact that we all make what we think at the time to be reasonable decisions – based on our goals, our knowledge, and the focus of our attention. And this applies ‘in the moment’ as the decisions are made. It is not a retrospective judgement that can easily be reached by others sitting in a comfortable chair with a nice cup of coffee in their hand

    So many accident reports start at the end of the process (the final event – in this case multiple fatalities) and then work back through the decision tree working out when the subject of the accident zigged (when they should have zagged), and zagged (when they should have zigged). This process almost inevitably suffers from many biases – it’s retrospective, often involves counterfactual statements, it involves making judgements using criteria that are only applied after the event and it often suffers from judgments that are proximal (close to the incident) and fails to look at the wider picture. I strongly suggest reading Dekker’s book for a more comprehensive description of these issues.

    It’s not invalid to come these judgments ‘after the event’. But it can be of limited value in educating ourselves because it is so easy for us to say ‘I would never have done that’ – I wouldn’t have skipped my BFR, I would have checked the weather, I would have filed a flight plan, I wouldn’t have made that flight at all…..

    That is certainly useful learning because it shows us circumstances that can get ourselves into trouble. A long time ago when I was in the student bit of the RAF when I was at college there was a poster in our ops room that that stated – ‘A superior pilot is one who uses his superior judgment to avoid situations requiring the use of his superior skill’.

    Dekker describes the principle of getting inside the pipe (sometimes called the Dekker tube). When we are inside the pipe then all we know is what we see in front of us at that moment in time. We are assimilating information from what sources we have and making decisions based on that data. If we could sit outside the pipe (as others do when they sit in judgement over our decision making) it would be obvious to us that we are moving step by step toward an adverse outcome. When we read accident reports such as this then it is so easy to say – why didn’t he do this, or that? – because we know the outcome.

    But the pilot of this aircraft, at the time the decisions were being made, did not know the outcome. Because if he did then he would have done something different – the principle of ‘local rationality’.

    Please let me be clear – I am not seeking to ignore the issues that this kind of incident raises. 3 people are now dead as a result – and likely some loved ones left desolate. But I am also pretty confident that the pilot in command did not depart that day intending to die.

    But the real learning from this event would come from getting inside the pilot’s mind and flying this flight alongside him in real time trying to work out what he he thought he was seeing at any moment (and I am not just thinking of outside the window) and then his decision making as a result of that situational assessment. Tragically that will never be available to us as (and similar to so many GA incidents) the data stream available for analysis is so minimal.

    But before we sit in comfort making retrospective judgement of the actions of others I do encourage you to read Dekker’s book. I worked as an orthopaedic surgeon in a UK hospital for 24 years and toward the end of my career got involved with investigating adverse incidents in our own hospital to try and identify learning and system changes required. This book turned my approach to these investigations upside down and the reports I wrote afterward were completely different to the ones I wrote before.