Sometimes a comment appears that’s worthy of its own post.
Vitrivius;

I think that one of the biggest problems that is posed by innumeracy is the inability of the innumerate to reason with big numbers. For example, some time ago some people were complaining that the oil patch uses 330,000,000 liters of water a year from the Red Deer River. 330,000,000 liters! Ohh, be vewy afwaid!
But wait just a minute, how big is 330,000,000 liters in the Red Deer River context? Well, the river flow rate varies from 10 to 100 cubic meters per second. Assuming only 10 cubic meters per second, that’s still 10,000 liters per second. Which means the oil patch uses 33,000 seconds, or about 9 hours, or about 0.1% of the river’s annual flow.
Why weren’t those people complaining about the oil patch wanting to use 0.1% of the river flow? Why do they say 330,000,000 liters instead? Are they trying to hide a fraudulent agenda behind big numbers?
Ok, now let’s consider the case of atmospheric CO2 vapor. Humans produce about 50 giga-tonnes of atmospheric CO2 vapour per year. Be vewy afwaid!
But how big is 50 giga-tonnes of atmospheric CO2 vapour in the green house gas global warming context? Well, the portion of atmospheric CO2 vapor that is produced by humans is about 3% or 0.03. The portion of green-house gasses that is CO2 is about 1% or 0.01. Thus, the portion of green-house gasses that is human produced CO2 is about 0.03% or 0.0003. The heat-trapping effectiveness of CO2 compared to the average of green-house gasses is about 10% or 0.10. Thus, the portion of the green-house gas global warming caused by human CO2 is about 0.003% or 0.00003 or 30 millionths. Even if we stop producing any CO2 at all, 99.997% of green-house gas global warming will still happen (even if we ignore the Borrello Limit and the saturation effect).
Here’s one more case to consider, this one with a twist: the Canadian long-gun registry. Some people say, well, a billion dollars, you know these big projects are expensive, there’s nothing to fear. But how big is a billion dollars in the context of the long-gun registry?
Well, given that loaded staff costs of the sorts involved in such a project are about $100,000 per year, on the order of 10,000 man years have been spent on the registry. But there’s no legitimate way to spend 10,000 man-years on such a project. Therefore, some part of that money was spent illegitimately. In this case, innumeracy is used to hide a valid fear: that with a probability of 100%, our tax dollars are being wasted (or worse). (At tinyurl.com/ybnozb you can read my detailed analysis of the long-gun registry scam, which received Mark Steyn’s letter of the week award.)
Smoking, since it has been mentioned, is another such case. We are told that smoking kills. Nonsense; having a smoke won’t kill you. Smoking a lot for a long time has a tendancy to reduce one’s life expectancy by a few years. The fear-mongers tell us to be vewy afwaid. The numerate ask themselves: am I willing to spend a few end-years of my life on the overall cause of enjoying all the other decades of my life?
My point is: measured values mean nothing until they are scaled to a relevant context. People who fear-monger by deliberately hiding relevant context scaling are fraud artists.

Or my favourite – if you think that $90 is a lot to spend for the fuel required to move three tons of pickup truck down the highway for 300 miles – try mailing the sonofabitch sometime.