Whew, that felt good.

It seems I spend most of my time these days defending quantitative techniques in support of risk assessments. One of the most common arguments by folks who support subjective, qualitative opinions is that the numbers are not absolute; they are often based on assumptions, can be manipulated or "gamed", and generally can be refuted. All true enough in many (though not all) cases.

What I don’t understand is why people think their current subjective, qualitative approach is any different. One of the key reasons we focus on arguing about quantitative approaches is precisely the benefit I want – it makes everyone more objective if we can begin to evaluate the veracity of our assumptions, collect evidence on outcomes, and factor this information into our decision-making process.

Numbers provide a level of clarity and precision that you can’t get with qualitative approaches, and they are at least as accurate. This is an important point: with quantitative approaches, you cannot do any worse than you are doing today in your risk assessment, and, given the numerous biases that humans are known to exhibit, there is a tremendous upside. Anyone that is "gaming" numbers is doing it to support their own subjective, qualitative (likely-biased) approach.

Quantitative methods don’t need to be absolute (and never are when dealing with probabilities of future events) to be useful. Using numbers that other people disagree with doesn’t make you a liar any more than two people who disagree on what "due diligence" mechanisms to deploy makes one of them a liar. It provides an opportunity for collaboration and consensus that is much more specific than subjective, qualitative approaches.

The numbers themselves are not the point; they are the means to a better end.