My BON-nie Lies Over the Ocean...

How much energy are Americans wasting by having it lie idly over their bellies? Fat has energy right? Americans are supposedly obese, right? So, it's a fair question: How much raw fat energy are we not using?

Now I love math, despite an early childhood enmity with the subject. These days I'm an amateur mathematician, which just means I occasionally do math in my spare time for shits and giggles. So the other day amid the candidates' talk about energy I decided, tongue placed firmly in cheek, to figure out roughly how much energy Americans waste by eating too much. It was a calculation in the same Swiftian spirit as "A Modest Proposal".

It was fun to play with, and play is the right word. Indeed, the point of this post is to prove that so-called back-of-envelope/back-of-napkin/Fermi calculations are riddled with inaccuracies. Hereafter I will refer to this type of calculation as a "BON".

In my last post I state that a few BONs rendered an episode of Knight Rider an affront to science. As if this was important somehow.

Anyway; sometimes you see someone (usually arguing on the Internet) who is using a BON, completely without irony, as the last word in an issue. Pointing out the weaknesses in the equation always garners the same response:



So while I love BONs as much as the next guy, it's good to be aware of their limitations. Frankly trying to use a BON to prove a point is like trying to use a kid's plastic carpentry set to build a house: At best, you can make a close approximation of what building a house might look like.

Using my BON for the amount of energy retained as American body fat, we can see how the calculation might break down depending on the real-world circumstances:

((A1-B1)(C1)+(A2-B2)(C2))*3,500(0.239)= Amount of energy in kW h

Where
A= 1. Male average weight (in pounds) 2.Female Average weight
B= 1. Average male ideal weight 2. Average female Ideal weight
C= 1. Male population 2. Female population

3,500 is the average amount of energy in a pound of fat in Kilocalories.
0.239 is the conversion factor for kilowatt hours

There are obviously other ways of calculating this, and plenty of other factors can be plugged in for further accuracy, but this is a BON and not a complete study.

Of course, I calculated ideal weights using height weight charts. So here are the weaknesses:

Average weight: I assume a standard distribution in both men and women.

Average BMI/Ideal weight: I assumed (X3) average standard distributions of height, weight, and age

Other: BMI, and statistics and concepts of ideal weight do not extend to children, I look at the US population as being comprised of adult, average Joes and Janes. I assume this is isn't going to have a major effect.

For every assumption I make, I add a sometimes indeterminate amount of error. To quote a Steven Seagal movie, "Assumption is the mother of all fuck ups!" I would never stand in front of a roomful of people, and present this figure with a straight look on my face. At best, it will allow me to perform, real, well researched calculations, and come up a number perhaps somewhere within the same order of magnitude. That's all. Sometimes when you're playing with big enough numbers related to niggling little factors that are small enough, your algorithm might do as is. Most of the time though, an algorithm used to solve a well posed problem (this is important) will always benefit from more data.

By the way, based on various stats of mixed reliability, my BON result is 6.21 X 10^12 kilowatt hours or, 6.21 petawatt hours. This is about enough to power all of the US for about a fifth of one year. Which doesn't sound like much until you remember that the US is the number one consumer of electrical power in the world. This number also fails to consider the amount of energy expended in the harvest, transport, and delivery of food. But that's another BON for another day.