5 ms·

Can you solve Seth Godin's math problem?

14 points by Chrisroad 17y ago

apotheon 17y ago
That would only be a "math problem" if it came with enough numbers to do some calculations. Oh, wait -- he's probably assuming that the only worthwhile metric in the problem is fuel usage. Well, with that requirement, it's easy. Hell, I've even done the calculations for determining how long it would take to recoup my losses if I bought a motorcycle and used that instead of a car with decent fuel efficiency, given some assumptions about future gas prices, a while ago: http://sob.apotheon.org/?p=449 http://sob.apotheon.org/?p=449
jws 17y ago
If we americans worked in gallons/mile instead these sorts of decisions would be obvious and intuitive. No one goes out with the goal of burning a gallon of gas so measuring how many miles you can go by burning one is silly. mpg is the reciprocal of what they want and people are bad at reciprocals. I fixed his problem below: Suburbans: 23.5 liter/100km Priuseses: 4.7 liter/100km Suburban+: 18.0 liter/100km (save 5.5 liters) PriusPlus: 2.3 liter/100lm (save 2.4 liters) Would you rather save, 5.5 liters or 2.4 liters? (And you don't have to "go metric", just using gallons/1000miles would do fine.)
- 3pt14159 17y ago
  I live in Canada where we typically get both mpg and liter/100 km. I've found that, almost universally, my friends that report mpg have less fuel efficient cars than the ones that quote liters per 100 km when asked about a vehicle's fuel efficiency. I tend to think that one of the reasons that Americans tend to have less fuel efficient cars is that when they are evaluating alternatives 9 mpg and 10 mpg are "closer" or more equivalent than 47 mpg and 50 mpg. For example, if I go out and buy a Truck in Canada I could either compare Ford F150: 15.7 liters/100km (15 mpg) city, 11.2 liters/100km (21 mpg) highway. GMC Sierra 2500 4WD: 14.7 liters/100km (16 mpg) city, 11.76 liters/100km (20 mpg) highway. With the Sierra, you are using a full liter less of gas during city driving at the cost of only half a liter of highway driving, so unless I'm doing more than 66% of my driving on highway like conditions, I should go with the Sierra, but that information does not come across to gas conscious buyers that use mpg.
JimmyL 17y ago
The root of the problem is that we shouldn't be thinking in miles per gallon - we should be thinking in gallons per mile. The follow-up problem is that average person think miles per gallon is linear (as they would phrase it, "the difference between 1MPG and 2MPG is the same as the difference between 60MPG and 61MPG") because they don't know any better, whereas it's really an inverse/reciprocal relationship. I wrote a more though explanation of this - http://www.daniellanger.com/blog/2009/08/miles-per-gallon-vs-gallons-per-mile/ http://www.daniellanger.com/blog/2009/08/miles-per-gallon-vs... - the last time MPG came up in discussion, when there was some minor outrage over the Volt claiming 230MPG.
- lacker 17y ago
  we should be thinking in gallons per mile The average person is pretty bad at dealing with fractions, too. Are they going to get the difference between 0.05 gpm and 0.04 gpm? Seems pretty small, it's just 0.01!
  - JimmyL 17y ago
    Valid - gallons/100mi would probably be a more user-friendly measure. Another interesting argument, although I can't find a source offhand, was to force manufacturers to display a gallons/10,000mi - on the basis that since the average American drives something like 10Kmi per year, this would convey much more concretely the longer-term difference between fuel consumption.
- mmt 17y ago
  I'm going to agree with another commenter, however, that the "saving fuel" premise is flawed. Personally, I like MPG because it makes the range-on-a-tank-of-fuel calculation much easier, and I care much more about range than about fuel economy. I was sad when it turned out the original Honda Insight didn't get 70MPG, because, otherwise, it would have been interesting due to the 700 mile range.
  - JimmyL 17y ago
    I'm curious - why the fixation on range of a tank of gas? If we're talking about serious off-road vehicles, then I can see why you'd want to max out your range to make trip planning easier. But thinking of the majority of drivers doing the majority of their driving (especially in cars like a Honda Insight), they'd be hard-pressed to end up going somewhere that didn't have a gas station at least every 200 miles.
mnemonicsloth 17y ago
The premise is flawed. We start with: Let's say your goal is to reduce gasoline consumption. And then we start talking about fuel efficiency. IIRC, though, it's pretty well established in the economics literature that increasing fuel efficiency doesn't reduce gas consumption much, if at all. Increased fuel efficiency means lower effective gas prices, which people respond to by driving more. The net effect is about the same level of fuel consumption, with more cars on the road (i.e. more traffic and more accidents). If you want to reduce gas consumption, you have to increase the price. Good ways of doing this include taxing it, increasing demand for other stuff made from petroleum, or reducing supply.
- modoc 17y ago
  The problem with increasing the price of gas is you hurt the poor most of all. Honestly, jacking up the price of gas isn't going to change my driving habits at all, but doubling the cost of gas (for example) will absolutely impact the budget of a lower income person who is commuting to work in an older car. They can't afford a newer car, they can't afford to move closer into the city, and there isn't public transportation available. They have no options other than to cut back on things like food and braces for their kids, etc... (sure some folks blow lots of money on cigs, big screen tvs, beer, etc... - but honestly the majority of people who are just scraping by, are doing their best). Figure out how to make it affordable (while being safe) for people to live closer into major cities, and you'll have a real impact on gasoline usage.
  - roundsquare 17y ago
    Two things come to mind that might work: 1) Gas credits for the poor... but this might be hard to get right. 2) Cheap public transportation. This won't work everywhere, but I suspect it would work in big cities and their suburbs.
- pingswept 17y ago
  I think you're basically right, but I wonder about the details. For example, my own driving is not money-limited-- gas could double in price and I would drive just as much, because I drive so little already. Whether I pay $200/year or $400/year for gas doesn't make a difference. I suspect that the vast majority of miles driven are driven by people who would find cheaper alternatives, like buses or bicycles or email, if prices went up even 20%. But that's just speculation. Empirically, when gas costs doubled briefly last year, it seemed to me that traffic was still almost as bad. So, I mostly agree with you, but I'm not 100% convinced.
charliepark 17y ago
This is a solved problem. Seth actually linked to my explanation of it. http://charliepark.tumblr.com/post/169016492/in-seth-godins-post-this-morning-he-talks-about http://charliepark.tumblr.com/post/169016492/in-seth-godins-...
RiderOfGiraffes 17y ago
I find the easiest thing to do is assume we drive some distance, then work out how much gas is used. If we drive 1300 miles the numbers all work out easily (because it's divisible by 10, 50 100 and 13) Suburbans get 10 mpg: 130 gallons Priuses get 50 mpg: 26 gallons. Suburbans @ 13 mpg: 100 gallons (30 gallons improvement) New Prius @ 100 mpg: 13 gallons (13 gallons improvement) So clearly the best thing to do is scrap all the Suburbans and replace each with a newer style Prius.
petercooper 17y ago
m/10 - m/13 > m/50 - m/100
anigbrowl 17y ago
Here's a better solution: replace the suburbans with Priuses instead of buying into Godin's tired, second-hand false dichotomy. Some of us are wired for arithmetic, actually.
known 17y ago
If the goal is to reduce gasoline consumption I'd suggest * banning futures trading on oil world-wide. * provide free public transportation.
rubinelli 17y ago
It may look unintuitive, but it's just an optimmization problem. When you optimize, you always begin with the most resource-intensive element, because that's where most of the gains are.

14 points by Chrisroad 17y ago

That would only be a "math problem" if it came with enough numbers to do some calculations. Oh, wait -- he's probably assuming that the only worthwhile metric in the problem is fuel usage. Well, with that requirement, it's easy. Hell, I've even done the calculations for determining how long it would take to recoup my losses if I bought a motorcycle and used that instead of a car with decent fuel efficiency, given some assumptions about future gas prices, a while ago: http://sob.apotheon.org/?p=449 http://sob.apotheon.org/?p=449

jws 17y ago

If we americans worked in gallons/mile instead these sorts of decisions would be obvious and intuitive. No one goes out with the goal of burning a gallon of gas so measuring how many miles you can go by burning one is silly. mpg is the reciprocal of what they want and people are bad at reciprocals. I fixed his problem below: Suburbans: 23.5 liter/100km Priuseses: 4.7 liter/100km Suburban+: 18.0 liter/100km (save 5.5 liters) PriusPlus: 2.3 liter/100lm (save 2.4 liters) Would you rather save, 5.5 liters or 2.4 liters? (And you don't have to "go metric", just using gallons/1000miles would do fine.)

3pt14159 17y ago
I live in Canada where we typically get both mpg and liter/100 km. I've found that, almost universally, my friends that report mpg have less fuel efficient cars than the ones that quote liters per 100 km when asked about a vehicle's fuel efficiency. I tend to think that one of the reasons that Americans tend to have less fuel efficient cars is that when they are evaluating alternatives 9 mpg and 10 mpg are "closer" or more equivalent than 47 mpg and 50 mpg. For example, if I go out and buy a Truck in Canada I could either compare Ford F150: 15.7 liters/100km (15 mpg) city, 11.2 liters/100km (21 mpg) highway. GMC Sierra 2500 4WD: 14.7 liters/100km (16 mpg) city, 11.76 liters/100km (20 mpg) highway. With the Sierra, you are using a full liter less of gas during city driving at the cost of only half a liter of highway driving, so unless I'm doing more than 66% of my driving on highway like conditions, I should go with the Sierra, but that information does not come across to gas conscious buyers that use mpg.

JimmyL 17y ago

The root of the problem is that we shouldn't be thinking in miles per gallon - we should be thinking in gallons per mile. The follow-up problem is that average person think miles per gallon is linear (as they would phrase it, "the difference between 1MPG and 2MPG is the same as the difference between 60MPG and 61MPG") because they don't know any better, whereas it's really an inverse/reciprocal relationship. I wrote a more though explanation of this - http://www.daniellanger.com/blog/2009/08/miles-per-gallon-vs-gallons-per-mile/ http://www.daniellanger.com/blog/2009/08/miles-per-gallon-vs... - the last time MPG came up in discussion, when there was some minor outrage over the Volt claiming 230MPG.

lacker 17y ago
we should be thinking in gallons per mile The average person is pretty bad at dealing with fractions, too. Are they going to get the difference between 0.05 gpm and 0.04 gpm? Seems pretty small, it's just 0.01!
- JimmyL 17y ago
  Valid - gallons/100mi would probably be a more user-friendly measure. Another interesting argument, although I can't find a source offhand, was to force manufacturers to display a gallons/10,000mi - on the basis that since the average American drives something like 10Kmi per year, this would convey much more concretely the longer-term difference between fuel consumption.
mmt 17y ago
I'm going to agree with another commenter, however, that the "saving fuel" premise is flawed. Personally, I like MPG because it makes the range-on-a-tank-of-fuel calculation much easier, and I care much more about range than about fuel economy. I was sad when it turned out the original Honda Insight didn't get 70MPG, because, otherwise, it would have been interesting due to the 700 mile range.
- JimmyL 17y ago
  I'm curious - why the fixation on range of a tank of gas? If we're talking about serious off-road vehicles, then I can see why you'd want to max out your range to make trip planning easier. But thinking of the majority of drivers doing the majority of their driving (especially in cars like a Honda Insight), they'd be hard-pressed to end up going somewhere that didn't have a gas station at least every 200 miles.

mnemonicsloth 17y ago

The premise is flawed. We start with: Let's say your goal is to reduce gasoline consumption. And then we start talking about fuel efficiency. IIRC, though, it's pretty well established in the economics literature that increasing fuel efficiency doesn't reduce gas consumption much, if at all. Increased fuel efficiency means lower effective gas prices, which people respond to by driving more. The net effect is about the same level of fuel consumption, with more cars on the road (i.e. more traffic and more accidents). If you want to reduce gas consumption, you have to increase the price. Good ways of doing this include taxing it, increasing demand for other stuff made from petroleum, or reducing supply.

modoc 17y ago
The problem with increasing the price of gas is you hurt the poor most of all. Honestly, jacking up the price of gas isn't going to change my driving habits at all, but doubling the cost of gas (for example) will absolutely impact the budget of a lower income person who is commuting to work in an older car. They can't afford a newer car, they can't afford to move closer into the city, and there isn't public transportation available. They have no options other than to cut back on things like food and braces for their kids, etc... (sure some folks blow lots of money on cigs, big screen tvs, beer, etc... - but honestly the majority of people who are just scraping by, are doing their best). Figure out how to make it affordable (while being safe) for people to live closer into major cities, and you'll have a real impact on gasoline usage.
- roundsquare 17y ago
  Two things come to mind that might work: 1) Gas credits for the poor... but this might be hard to get right. 2) Cheap public transportation. This won't work everywhere, but I suspect it would work in big cities and their suburbs.
pingswept 17y ago
I think you're basically right, but I wonder about the details. For example, my own driving is not money-limited-- gas could double in price and I would drive just as much, because I drive so little already. Whether I pay $200/year or $400/year for gas doesn't make a difference. I suspect that the vast majority of miles driven are driven by people who would find cheaper alternatives, like buses or bicycles or email, if prices went up even 20%. But that's just speculation. Empirically, when gas costs doubled briefly last year, it seemed to me that traffic was still almost as bad. So, I mostly agree with you, but I'm not 100% convinced.

charliepark 17y ago

This is a solved problem. Seth actually linked to my explanation of it. http://charliepark.tumblr.com/post/169016492/in-seth-godins-post-this-morning-he-talks-about http://charliepark.tumblr.com/post/169016492/in-seth-godins-...

RiderOfGiraffes 17y ago

I find the easiest thing to do is assume we drive some distance, then work out how much gas is used. If we drive 1300 miles the numbers all work out easily (because it's divisible by 10, 50 100 and 13) Suburbans get 10 mpg: 130 gallons Priuses get 50 mpg: 26 gallons. Suburbans @ 13 mpg: 100 gallons (30 gallons improvement) New Prius @ 100 mpg: 13 gallons (13 gallons improvement) So clearly the best thing to do is scrap all the Suburbans and replace each with a newer style Prius.

petercooper 17y ago

m/10 - m/13 > m/50 - m/100

anigbrowl 17y ago

Here's a better solution: replace the suburbans with Priuses instead of buying into Godin's tired, second-hand false dichotomy. Some of us are wired for arithmetic, actually.

known 17y ago

If the goal is to reduce gasoline consumption I'd suggest * banning futures trading on oil world-wide. * provide free public transportation.

rubinelli 17y ago

It may look unintuitive, but it's just an optimmization problem. When you optimize, you always begin with the most resource-intensive element, because that's where most of the gains are.