Are accidents and incidents Zipfian?

In this post, I shall discuss the possibility of modelling (or rather predicting) the occurrence of anomalies in a system using Zifp’s law. Also, I will bring up empirical evidence from time to time to tip the argument in my favor.

Allow me bring up the evidence first. The case of the Malaysian Airlines tragedy, Air Asia and Egypt Air were all likely succumbed to this probability model. How so? The first crash involving a Malaysian Airlines plane occurred one fateful day of March, 2014 and it was the first time an MAS plane was involved in a major hull loss incident. Soon after in July that same year, another MAS plane was downed by Ukrainian forces.

We see that when something which is unlikely of happening happens, a similar something is exponentially more likely to happen again.

The incidents involving Egypt Air and Air Asia were no different. An MSR plane is first hijacked and soon after, one of its planes crashes. Again, an AXM plane crashes and soon after a plane operated by the same airline is involved in a runway excursion incident. Now you might wonder why I have chosen to constrain my argument to those of aviation-related incidents and accidents – and I have a perfectly good reason why because they are very rare making them the apt candidate for our discussion.

Let’s open up our minds a little now, I propose we might be able to predict the state of the human race’s understanding of the universe several years from now based on instantaneous trends in research and data collection. Say we make a breakthrough in chemotherapy and assuming this event might obey Zipf’s law, wouldn’t it be safe to say that we are more likely to make another similar breakthrough with the trend in chemotherapy breakthroughs moving upward exponentially. This way, we could model how we might end up in looking several years into the future, pretty much like how smartphones took the consumer electronics’ industry by storm. It would have been near impossible to linearly extrapolate from the 1980’s or so up until now and say what technology might have looked like.

I believe it is possible to explain the progress of intelligent civilizations in a similar way as demonstrated above.

Zipf’s law is more than just modelling events linearly. One could say that Zipf’s law allows us to make decisions and/or predict what the state of a system overall ahead of time by taking into consideration what is happening and what is likely to happen within the system. To answer my original question if accidents and incidents are completely Zipfian is plain stupid without any hard evidence, however I’d be comfortable with exploiting this trend to keep myself safe and make careful decisions.