Melançon Enterprises • a multidivision multinational since 1980

Important counter-intuitive (for some of us) math in Seth Godin's latest, "The Magic of Low-Hanging Fruit:

Imagine that half the cars in the US get 10 miles per gallon. And half get 40 miles per gallon. Further stipulate that all cars are driven the same number of miles per year.

Now, you get one wish. You can give every low-mileage car a new set of spark plugs that will increase fuel efficiency by 5 mpg, up to 15. Or you can replace every 40 mpg car with a car that gets 75 mpg, an increase of 35 miles for every gallon driven.

Which is better?

It turns out that the 5 mpg increase is far better for overall mileage than the 35 mpg increase, even though it's smaller both as a percentage and absolutely. That's because the 10 mpg hogs use up so much gas. They're the low-hanging fruit, not just easy to fix, but worth fixing.

I had to break out the calculator to confirm this:

100 / 10
= 10

100 / 40
= 2.5

100 / 15
= 6.6666666666666667

100 / 75
= 1.3333333333333333

The cars are equally split, all the cars are driven an equal amount, the miles per gallon increase is better for the more efficient cars, even the percentage increase over the already higher miles per gallon is better.

Yet 10 + 1.3 is not as good as 6.7 + 2.5.

What matters here is not percentages but absolute improvement, the amount of gasoline used.

It's the law of diminishing returns. If you had half your cars getting 1,000 miles to the gallon, doubling or tripling or quadrupling that performance is way less valuable than getting an extra half-mile per gallon out of half your fleet getting one mile per gallon, because that's so abysmal that the vast majority of your gas usage will come from that half. The super-efficient half of your cars are already approaching zero in the amount of gas they use. No matter how spectacular their performance increase, their total potential impact on gasoline used is small because they use so little already.

The same holds for the more realistic scenario Godin describes, but it's important to note that usually the bad will be more common – not half the fleet but three quarters, although maybe not quite the range of bad (even SUVs mostly get at least 15 mpg already, right) to good.