Stop using averages. Use percentiles instead

Bill Gates walks into a pub. Instantly, the average wealth of each person in the pub is 45 million dollars.

But really, averages are bad

The primary purpose of averages is to measure changes over time in the same sample group or cohort. It is in this application, or more so misapplications, by using averages for different purposes that the three most common errors occur.

Using Percentiles

Our average, 559ms, misses the mark entirely

Calculating percentiles

  • p50 values = [ ̶1̶2̶0̶,̶ ̶1̶2̶0̶,̶ ̶1̶3̶4̶,̶ ̶1̶5̶5̶, 300, 867, 980, 1800]
  • p90 values = [ ̶1̶2̶0̶,̶ ̶1̶2̶0̶,̶ ̶1̶3̶4̶,̶ ̶1̶5̶5̶,̶ ̶3̶0̶0̶,̶ ̶8̶6̶7̶,̶ ̶9̶8̶0̶, 1800]

What percentiles tell us

p50 and p90 tell a more accurate story

Settings goals

Downsides to all this

--

--

--

I work for the Internet. By day, a front-end web developer with a passion for typography and design. By night, I’m sleeping.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

A Brief History of Medical Image Registration: Part 3

Sales dashboard using Power BI

Storytelling with data

Customer churn prediction with Keras and Pandas

Decomposition of Variability

Simple Modeling of NBA positions using the K-Nearest Neighbors Machine Learning Algorithm

Marginal vs Conditional Probabilities — Visualizations and Models in R.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Ricardo Magalhães

Ricardo Magalhães

I work for the Internet. By day, a front-end web developer with a passion for typography and design. By night, I’m sleeping.

More from Medium

Fraud Prevention: Exploration-Exploitation Tradeoff in AI-based Systems

How Machine Learning could help design more customized Mental Healthcare Treatments

Machine Learning for Mental Healthcare Treatments

Intro to Hadoop and It’s Core Components

Diving into Data!