Murphy’s Corollary: anything that did not go wrong could not have gone wrong

From the paper, in Mike’s words, lightly edited; his indented asides are marked.

The law as we quote it #

The statement of Murphy’s Law that has gone into the popular lexicon is Anything that can go wrong will go wrong. That’s probably not what Murphy said, but that’s how we quote him today. And when we quote Murphy in this way, it is likely that we are doing it mostly for comic or ironic effect.

Because to be clear, nobody really believes this statement of Murphy’s Law. Almost anything could go wrong, and if it were really true that anything that could go wrong would go wrong, then most everything would go wrong. And most things don’t.

Where it came from: the rocket sled #

In the paper, an indented aside: material beyond the talk

The origin of Murphy’s law is generally attributed to a statement allegedly made by Capt. Edward A. Murphy, a development engineer from Wright Field Aircraft Lab working on Air Force Project MX981 at Edwards Air Force Base in 1949. Commonly known as the rocket sled tests, the project was designed to see how much sudden deceleration a person could withstand in a crash.

Paper, notes 1, 2 “Murphy’s laws origin,” excerpted from The Desert Wings, March 3, 1978; Nick T. Spark, “The Road to Murphy’s Law,” Improbable Research, 2003.

Murphy had brought a g-force sensor made of a bridge of four strain gauges. The gauges could be wired together in more than one way, and one of those ways would cancel them out and produce a zero reading. After just such a result, Murphy allegedly remarked in frustration, If there is any way to do it wrong, he will, referring to the technician who had wired the gauges.

Paper, note 3 “Murphy’s Law,” Wikipedia, accessed September 5, 2015.

The basic observation can be found at least 70 years before Murphy’s statement, and likely goes back much further, but today it is almost universally attributed to him.

What Murphy was really saying #

In the paper, an indented aside: material beyond the talk

The statement is obviously false, but only because the common phrasing omits one important word. What the law should say is: anything that can go wrong will go wrong, eventually. Given enough trials, if something can go wrong, it eventually will.

The value of Murphy’s observation is not in the relatively pessimistic implication of the Law as it is quoted today, but rather in the more meaningful way in which it informs design – both the design of things, and the design of procedures. What Murphy is really saying is that if we can design out the possibility of doing it wrong, we should. Our lives today are full of examples of this – the polarized electrical plug being just one.

The paper’s tragic illustration of the point: Case study · The F-86 aileron bolt.

The corollary #

But if Murphy’s Law were true, if anything that could go wrong really would go wrong, then a corollary would also be true:

Anything that did not go wrong could not have gone wrong.

Even though we don’t believe Murphy’s Law, we are surprisingly subject to being seduced by this corollary. If nothing did go wrong, then nothing could have gone wrong. We take a positive outcome as evidence that a positive outcome was inevitable.

That is what happened with our decision to use top landings to save time in production test flying. We had a long string of successes, and each one reinforced our perception that success was an outcome we could rely on. The problem with that perception is that our data set is too small, at least at the beginning of any such process.

Warning signs, suppressed #

In the paper, an indented aside: material beyond the talk

We had many indications during those fifteen years of successful landings that the decision to use this technique was of questionable safety. There were a couple of reasons we chose to land on the upslope of the side of the hill, below the top. One reason was that there was inevitably some turbulence, either rotor or convective or both, as we passed through an altitude of about twenty to fifty feet agl. Being weight shift controlled aircraft with relatively large wingspans and light wing loading (33 foot span, 1.1 to 2.0 lbs per square foot) hang gliders suffer from a progressive loss of control authority and response at lower speeds, especially as one approaches stall speed while slowing down. Approaching from below the top allowed us to approach at high speed for better control, and landing on the upslope greatly reduced the duration of time spent at low speed, reducing the probability of a loss of control. Typically, we would pass through the area of turbulence while still flying fast, then enter a region of wind gradient, where airspeed would drop dramatically as we rounded out to follow the slope of the hill, and then almost immediately flared for landing.

Many times, while passing through the region of turbulence and then entering the gradient, I remember thinking, this may not end well. And yet, with each successful landing, my next thought was, well that was fine, no problem – what could go wrong?

I had even observed other pilots having significant crash landings, sometimes even damaging gliders, yet I was able to rationalize those as having been due to poor technique – a poorly executed approach path, or poor control of airspeed on approach. It was only when it happened to me, following what I deemed to be a perfectly executed approach, that I was able to realize that my doubts about the safety had in fact been correct, and the suppression of those doubts had been Murphy’s corollary in action.

From the 2018 Gulfstream talk, not the paper.

Telling the same story in 2018, Mike added two sentences after the crash landings he had watched:

They crashed, but I wouldn’t have. Ego is a powerful enabler of rationalization.

@misc{meier2026corollary,
  author = {Meier, Mike},
  title  = {Murphy's Corollary: Anything That Did Not Go Wrong
            Could Not Have Gone Wrong},
  year   = {2026},
  note   = {Edition 1.0},
  url    = {https://meiersafe.com/murphys-corollary/}
}