As kind of a follow-up to my article a few weeks ago about Quantity over Quality, this post by Seth Godin made some very interesting observations about how to improve efficiently.

Question: What to Improve?

We all have things in our playing that are pretty good.

It could be a comfortable range, dynamics, or keys.

Many times in practice we try to take the stuff we do pretty well and make it perfect.

But math shows us that that may not be the best way to actually get good.

Seth’s Questions

• You have a daily commute. Half of it is dirt road, half is paved. Your current car can go 10 MPH on the dirt road, and 50 MPH on the paved road. Which of the following is a better choice:
  • A car that can drive 22 MPH on dirt roads but still 50 MPH on paved roads?
  • A car that can drive 200 MPH on paved roads, but still 10 MPH on dirt roads?

• You have a steel-processing factory with 2 machines, each working 100% of the time. One machine processes steel with 20% accuracy and the other machine 80% accuracy. Which is a better upgrade:
  • Upgrade the 20% machine to 30%.
  • Upgrade the 80% machine to 100%

• You own two cars and want to improve mileage. You have an SUV that get 14 MPG and a sedan that gets 35 MPG. Should you:
  • Upgrade the SUVs to 35-MPG sedans?
  • Upgrade the sedans to electric cars (100,000 MPG)?

The Math Doesn’t Lie

For the math-averse, here are the answers to each of these questions, with the math to back it up.

Spoiler alert: It’s always better to improve the straggler.