Numbers are just relationships

August 10, 2009 · Filed Under Uncategorized 

My friend and colleague August Jackson says that in most management contexts, fake numbers will win out over real logic. Pie charts sound warm and delicious, while cold hard facts sound like they are tasteless with an unpalatable texture, which is often the case. So deep is our attachment to numbers that often, we don’t even care how they were calculated or what they mean.  At least statistics count as “quantifiable,” which for many is superior than common sense that lacks numbers.

And yet I love statistics, any form of abstraction that helps us understand a rapidly changing world. They often help us understand more about the analysts and what they chose to study, but it’s usually better than nothing. And when they are really useful they help us unpack dangerously inaccurate beliefs that we hold, usually from listening to media stories picked for their sensational value rather than their informational content.

crimeExample: If you listen to the major media, you will think that the United States is at an all-time peak of violence, a cesspool of shootings and abductions and freak crimes. A quick look at statistics will show you that actually, violent crime has dropped in a statistically-significant way since 1993. News reportings of violent crime are up, but the actual numbers are down.

That’s useful to know. Now we can stop feeling like the world is out of control and that humans are getting worse, and we can start asking why our TV stations and newspapers are invested in us feeling scared. By looking at some simple statistics, our understanding of the world just got a little better. Isn’t it cool to be an analyst, and not just accept other people’s view of the world at face value?

As a competitive intelligence aficionado, this dynamic tension is my life - how do we use numbers, understand what they mean, and see beyond them, while thinking clearly and logically? It’s not easy, but it’s what makes being an analyst both an art and a science.

bestchartevar

Not all sets of statistics are useful in helping us understand our world. I got thinking about this when the guys at Fark pointed me to this graph of the Forex for pounds sterling/US dollars under the caption “This will make it all clear.”

OK, kinda funny. What is this stuff? Is it about money? Wealth? Life getting easier or better? What will it help you predict?

This is my tension with a LOT of numbers bandied about these days, especially on the subject of the putative economic recovery.

The questions are in no short supply.

These are all questions being asked incessantly these days. The answers are typically being given in number form. BIG number form. Well, this bank owes five trillion in derivates, so we spent two trillion in borrowed money to stimulate other numbers, especially stocks and housing prices. So, um, we know lots of you don’t have jobs and find your wages dropping, but here are some big numbers to look at to make you feel like a recovery is in the works. You want this to recover, don’t you?

Here’s the problem: I don’t really understand what a trillion is, and neither do you. Moreover, I’m pretty sure I don’t understand that Forex chart above, and for that matter I’m not sure why stocks are up.

Here’s another, bigger question: What does any of this matter? At the end of the day, these numbers are here to serve human beings, not the other way around. Their adoption less than 100 years ago was supposed to bring a scientific bent to management and make things more efficient. However, these days our statistical measures have less to do with actual human behavior than ever. Sure, perhaps they describe the behavior of a few thousand people in Manhattan, London City, and Tokyo, but they have less and less to do with people waking up, making coffee or tea, leaving the house and working with fellow humans to provide value to each other. Our statistical measures are supposed to help us better understand our relationships with one another, but they are rapidly become a self-justifying system of overcomplex gambling. We watch these numbers with baited breath, but I am less and less convinced that the return of our numbers will mean a return to a pleasant humane life.

As such, perhaps some non-statistical questions may be the only ones that suffice on the subject of the global economy. Instead of waiting for irrational and abstract statistics to show you what the world looks like, try these non-quantifiable measures:

  • What kind of food are you eating? Is it good?
  • Do you get to eat with friends and loved ones?
  • How many hours a week are you working?
  • Would you prefer to work more or less?
  • How many of your family and friends are sick right now?
  • Can they get the care they need?
  • How often can you take part in some form of leisure activity – camping, dancing, music, polo, whatever?
  • How much time will it take you to get to the beautiful space (park, museum, mountain range) near your home?
  • Do you know your neighbors?

Yes, you can makes these questions into statistics, but I dare say they are less abstract than the S&P 500. They will skip the phenomenon of “numbers for numbers’ sake” and ask about how we are doing as human beings, relating to one another, even in an economic sense. So it’s not the numbers that are the problem, but the relationships they are describing.

As analysts, let’s make sure we’re trying to understand the right relationships.

Tags:

Comments

  • John Larson
    Nice! I like the "soft" questions you pose much better for two reasons. First, they are MUCH more relevant to the actual quality of my life. Second, I have power to impact the answers. It is WAY more fruitful to concern myself with those issues than with the thinly veiled gambling that is our current cultural economics focus.
blog comments powered by Disqus