Risk Management—How the word “Probably” Influences your Decision Making

[Update: Take our new GPS Survey here] Last year, the FAA asked me to write a series of safety tips they could email to airmen registered at www.faasafety.gov. If you’re unfamiliar with the Web site, go there now to register to receive email notification of safety seminars in your area. You’ll also find useful online courses and information on the FAA WINGS program, a substitute for your flight review, previously known as a BFR.

My Safety Tip #3 brought more comments and...

diversity of opinion than anything I’ve ever written. Two-thirds of the comment were extremely positive, a sixth were neutral in that they agreed is some areas and disagreed in others and the rest were negative including one (which I’m not reprinting) that was downright hostile, something I rarely encounter in aviation.

I'm delighted the article got people thinking about risk management and a concept that might be helpful to them. Reader feedback forced me to rethink some aspects of the article. I’ve posted the original article below, followed by representative reader comments and my thoughts on what I would say differently.

Purge "Probably" From Your Flying Vocabulary
Just as a drug-sniffing dog alerts in the presence of an illegal substance, you should be alert anytime the word “probably” pops into your head when flying. Probably means that you’ve done an informal assessment of the likelihood of an event occurring and have assigned a probability to it. The term implies that you believe that things will most likely work out but that there’s some reasonable doubt in your mind. Because humans are in general optimistic, there’s also a good chance that you’ve overestimated the probability of success, as do the approximately 300 pilots a year who suffer fatal accidents.

If you ever think that your course of action will “probably work out,” you need to choose a new option that you know will work out. Even if you feel there’s a 99-percent probability that things will work out, that shouldn’t be sufficient justification for you to continue with a course of action. Would you play Russian roulette with a gun that had 100 chambers and just one bullet in it? I hope not.

Pilots can also be lured into a false sense of security if they’ve performed a risky behavior successfully in the past. One CFII, known locally by his colleagues as “Luke Skywalker,” had a reputation for always being able to make it into his local airport—which didn’t have an instrument approach—regardless of how bad the weather was. Having succeeded perhaps a hundred times, he may have felt justified in believing that he could always make his system work. The last time he tried, however, it didn’t work and he became a statistic.

In the San Francisco Bay area, about one-half of all VMC-into-IMC accidents occur in the Livermore Valley, probably because a marine layer of clouds frequently obscures the mountains that rise from sea level to about 4,000 feet. A common way to traverse this area is through the Altamont and Sunol mountain passes. I tell pilots that if they ever approach these passes and, based on visibility, think “they can probably make it through,” they need to make a 180° turn and land at an alternate airport. Undoubtedly, every pilot who crashed in this area thought that he or she would “probably” make it through—otherwise the pilot wouldn’t have continued.

You should always assess risk and prepare a Plan A, Plan B, and Plan C before you take off. That way, the first time the word “probably” flashes through your mind, you can instantly began to execute Plan B or Plan C. Fully thinking out these plans before you leave the ground leads to better decisions. Waiting until you know things are not going to work out is the wrong time to improvise a seat-of-the-pants decision.

Positive Feedback
This first comment is from the Chief Pilot of one of the largest corporations in America:
"I read the highlights of your article, sent out by the FAA's FAASTeam Safety Tip, 3/21/09, encouraging pilots to have a Plan B before they take off. Your point is a good one, and I totally agree."

From a Cessna Citation jet pilot:
"Well done, Max! I'm going to try to get all my flying companions to agree to challenge any use of the word probably!'"

Larry wrote:
"Outstanding article on use of the term 'probably.' As I replay some of my own 'bonehead' ventures throughout my flying career, all of the 'juicy' hangar stories would begin with (or soon involve) that very word."

Mark wrote:
"Hey Max, I just got your Safety tip from the FAA, about probably make it. Great one. I've been flying for 36 years, and I'll save that one in the back of my mind, not only for me, but for other pilots I talk to."

Neutral Feedback
Russ a high time CFI wrote in part:
"I also understand and agree with your point. I just look at it differently. Flying is all about managing risk. You take one risk after another, so you minimize the risk to an acceptable level.

"I think that you should continually ask yourself whether it is safe to continue the flight. And when you ask yourself this question you can never answer 'yes'. There is always some risk. If your answer is 'probably', then that means the risk is low and you should continue the flight, but if you answer 'I don't know', then you should definitely not continue the flight."

Ben, another CFI wrote:
"Good work overall on the "safety tip", but one thing bothered me a little bit. I agree with what I think your point is—which is the importance of recognizing the warning signs that it is time to change the plan [Yes Ben, that was indeed my point]—I believe some people will get the wrong message. You suggest that there can be a course of action of which you can be certain will work out [Oops, I guess I did]. I have not found that to be true, and In fact such misconception is what leads people to grief—'I'm sure I can make, it's only 3 more miles to the airport...'

"I know (well I think I know) that your point is that some people over-estimate the chances for success when they get into a risky situation. I'm pretty sure that was your point—'probably' is a clue you've over-stated your chances and should reconsider. I have never departed terra firma without a plan, and always know that the plan may need to change completely—never have I been certain as to where the flight would conclude. There are always unconstrained variables."

Negative Feedback
This commenter expresses more disagreement than agreement:
"The criticism of the use of 'probably' in the context in which you explain it is correct [Thank you], however, the advice that the word not be used is absurd."

Finally (and no this is not the hostile one):
"This is a very BAD article [I’m sure it could have been better, but BAD??]. It is always wise to minimize the risk associated with negative consequences [We agree here]. It is not possible to zero the risk out [I agree]. To imply that it is so provides a false sense of security and is detrimental to good decision-making in general and that for aviation, specifically.”[I think I get your point]

Max's Feedback
Just as my article stimulated readers’ thinking, their comments got me thinking. Perhaps the title was misleading. The title said to “purge” the word “probably” from your vocabulary, but the point of the article was that you should look at “probably” as a red warning flag that should cause you to pause and think whether there are any other better options available to you. Sometimes hyperbole is a useful tool. Hopefully the title grabbed your attention and got you thinking. Even if perhaps it stretched the article’s main point to an extreme.

When I said “you need to choose a new option that you know will work out,” that also included hyperbole and I think most readers recognized that, though I know some interpreted it 100% literally. Of course, there’s no way to eliminate all risk and that’s true in aviation. Perhaps I could have stated it more clearly by saying “you need to choose another option, hopefully one that reduces risk to the greatest possible extent."

Applying risk management to each flight is the best way to reduce the accident rate. I’m glad the article generated a large response, since it means people were actively thinking about risk management tools. Hopefully it gives you an additional concept to use as you develop your own approach to risk management.

Comments

[B]

Funny that the point flew over some people's heads (get it? flew over? har har)... to focus on eliminating a word from your vocabulary wasn't the issue at hand. "Probably" could be "maybe", "perhaps", or anything else. It's the -attitude- that Max is talking about here.

[Mark Fraser]

You are trying desperately to link breaking FAR's as a normal decision making process for pilots. CFII's that break regs (like your Luke Skywalker example) are irresponsible idiots, not rational decision makers. You also try to link irresponsible behavior like Russian Roulette with rational flight decision making based on "best outcome" of an action in a situation.

Your argument reminds me of proponents of gun laws. Gun laws (like FAR's) don't stop idiots and they don't help already responsible rational people.

As far as I can tell, you can't convert idiots so all the rest is preaching to the choir.

[Max Trescott]

Mark, I sure hope breaking FARs isn't normal for any pilot. My personal standard is to never knowing break an FAR; I hope others do the same. Regarding Russian Roulette, my point was that even a 99% probability of a favorable outcome--which is sufficient in many other areas of life--is totally inadequate for flying.

I agree we cannot convert idiots. Hopefully we can give other pilots additional decision making tools early enough in their careers to prevents them from getting into trouble. I think there are licensed pilots who were never told of the risks of say, flying in the mountains, IFR, at night. That doesn't make them idiots (though perhaps they lack curiosity to have not figured that out), but it does indicate that we have a pilot training system which sometimes fails to give pilots all of the tools they need to become safe pilots. Sadly, a lot of training is geared to meeting the minimums standards outlined by the PTS. But the PTS barely addresses decision making and judgment. Yet these factors address 80% of all accidents. So yes, some idiots can never be helped. But many newly minted and even experienced pilots can learn a few new tunes while being preached to in the choir.

[mcgowan]

Mr Trescott:
I just read your 'probably' article concerning a crosswind takeoff at max demonstrated velocity.
As I understand it, Max Demonstrated Crosswind velocity is NOT a limitation, but rather the greatest crosswind available to be demonstrated during the certification process. The aircraft may be capable of much more, or only that much. Without reading the engineering reports a pilot cannot know.

Also, it IS possible for more than half the pilots to 'be above average' mathematically. If for instance there were 10 pilots, and 9 of them scored 60points out of 100, and one scored 10, the average would be 55. 90% would be above average. You meant that only half could be above THE MEDIAN.

Get 'em right, please.

[RJ]

I am the chief flight instructor at a nice school in Northern Ca. I try my best to teach students the proper ways to fly and to manage risk the best the can by remembering to fly the airplane, not panic, and to always have a backup plan.
With this in mind, a student cancelled her private pilot checkride because of winds. She said she was not comfortable and did not want to risk failing the checkride and possible damage to the aircraft. Demonstrated crosswind for the Cessna 172N is 15 knots. Crosswind at the airport at the time her checkride was to take place exceeded that limit. I praised her decision not to go as did her flight instructor. Problem started when to owner of the school complained and questioned me as to why the student did not want to take her checkride. I simply stated that it was her decision and the reason she gave. After all, she is pilot in command. I was berated for not in some way forcing this student to take her checkride. What would I be teaching this student and her flight instructor in doing so? That's it's OK to risk running off the runway and damaging the aircraft just so you can take your checkride? I don't think so. I will always stand by a persons decision not to fly. Weather environment changes. Things will change. In this case, why take the risk.
Your article was great. Drove my point home.

[Howard Williams]

Great Article Max! I'm a CFI in Houston, TX. When planning a lesson with a student and we say we can 'probably' do this, then we do something else, probably doesn't cut in aviation. And for those who wish to take everything out of context or really disagree with this concept, I just hope you are not the subject of the next NTSB investigation or make the next issue of IFR!

[Max Trescott]

RJ, Good for you! CFIs should always stand up for what they know to be right, even if the boss doesn't want to hear it. Candidly, that's what you're being paid for--to teach good judgment, to exercise it yourself and to defend that good judgment when others may try to trample upon it.
Howard, as a well known airline pilot and author wrote to me today, “How will this look at the hearing when you are sitting at the end of the long table and everyone else has a glass of water?” A great phrase I'll try to remember the next time I even contemplate doing something unwise.
McGowan, you challenged me to "Get 'em right" and I contend that I did. No where in the article did I use the word limitation. Median is a type of average so it's not incorrect to use the term average in this context (though I admit, median would have been more accurate, which was the type of average I was thinking of). Of course you did exactly the same thing in your comment--you used the word average while describing "the mean." However if we used the mean, if every pilot scored a 100 we would all be perfect pilots, which isn't a very useful concept. Using the median is a far more useful concept, since half the pilots will always better than the other half. That 80% of the pilots think they're above average suggests that 30% of them have overestimated their skills, which could lead some of them to overestimate their ability to handle difficult situations.

[MJ Renolds]

Max: "my point was that even a 99% probability of a favorable outcome--which is sufficient in many other areas of life--is totally inadequate for flying."

Me: If 99% is TOTALLY inadequate for decision making in General Aviation by amateur pilots then you are at odds with the FAA and the entire FAR Part 91. Even FAA acceptable scores for pilot certification are not issued only when a student scores ABOVE the 99 percentile.

I hate to say it, but the FAA is being realistic and reasonable when it comes to regulating amateur aviation flight operations. You're not.

[Max Trescott]

MJ, I'm sorry if I wasn't clear. I wasn't talking about the passing grade required to get a pilot license. 99% is of course more than adequate. I'm was talking about in-flight decision making--something which the PTS doesn't test very well. My point is that if in flight, you make a decision which has only a 99% probability of success, than on one out of every one hundred such flights, you will have an accident--which is totally unacceptable in my mind.

[Howard N LaPierre]

Max

Another way of saying it is: "If you think you can make it, you have made your first mistake"

Howard