Predicting random violence by mathematics

[Ichingcarpenter]

The researchers looked through public data of military fatalities during the period of 2003 to 2010 in Iraq as well as a ten year period of fighting throughout Afghanistan. Using an open-access software, the researchers discovered that after an initial gap between the first two attacks, the attacks that follow seem to come faster and more frequent, but begin to follow a mathematical pattern.

This gradual increase and predictability of attacks seems to point to the facts that, essentially, practice makes perfect. The more the insurgents execute attacks, the easier and more predictable they become. The interval between the first two attacks is the key to predicting future attacks. The researchers were able to devise an equation, based on this interval, which is able to estimate the course of subsequent attacks.

The researchers tested this equation on over 3000 fatal terrorist attacks throughout the world and in all tests; the escalation of subsequent attacks estimated by the equation closely matched the actual outcomes.


SNIP


http://www.physorg.com/news/2011-07-random-violence-mathematics.html

[Syrinx]

1. practice makes perfect

Self-important researchers justify their craft by expressing common sense in fancy language.

That also explains why competent programmers aren't allowed to work until they get expensive degrees in computer science.

[caraher]

2. That's a pretty anti-science attitude there

This wasn't published by some rinky-dink journal, it was in Science - which is a good sign that people with some actual knowledge in the relevant fields found it important.

There's also a world of difference between merely observing "practice makes perfect" and developing a mathematical model that captures exactly how this works out, in detail, in the real world. Disdain for this result is somewhat like dismissing the value of accurate weather forecasts by noting that any fool knows it's hot in summer and cold in winter.