Laplace’s Rule, aka The Rule of Succession
I don’t know why it’s called the “rule of succession.” But it’s called Laplace’s Rule because it was developed by Pierre-Simon Laplace in 1774.
It’s also called the Laplace-Bayes Estimator, because it’s related to, or a result derived from, Bayes’ Theorem of 1764.
Say there is a event which has a chance of occurring, but you don’t know what its probability is. If you make a set of tries at the event (say, n tries), and it occurs some number of times (call that m), what can you say about the probability?
Laplace’s Rule says the most likely probably for the event is the fraction (m + 1)/(n + 2).
For example, if you have interviewed five customers, and three of them complain about a particular problem, what is the most likely percentage of all customers who suffer from that problem?
Laplace’s Rule gives us the answer – it’s 57%.
- (3+1)/(5+2) = 4/7, or 57%
(Ha! I just noticed I did the math wrong in the episode itself.)
Algorithms To Live By
I learned about Laplace’s Rule from a great book called Algorithms To Live By: The Computer Science of Human Decisions, by Brian Christian and Tom Griffiths.
I highly recommend the book. A delightful read, full of interesting computer science and math ideas, most of which are directly applicable either to life or to product management or both.
Other podcast episodes related to being “data-driven”
- 93: How To Show Quantifiable Results – Even If You Don’t Have The Numbers
- 92: Get “Data-Driven” With The The Most Powerful Computer For Analytics
- 86: Data-driven Product Management
If you are just getting into product management, or you’re busy leveling up into more senior PM positions, don’t forget there are a ton of resources in my previous episodes – on storytelling, persuasion, prioritization, working with developers, working with marketers – and on my website The Secret Product Manager Handbook.
And if you want to accelerate your product management career, go to calendly.com/nilsdavis/consultation for a FREE 45 minute coaching and discovery call. (No obligation.)
Don’t forget to subscribe!
If you want to help me out, and more importantly, help other product managers find and enjoy this podcast, please subscribe on your podcast system of choice. And consider giving me a review on iTunes or “starring” the episode on your player. It helps other people find the podcast, and makes me a little happier as well. I might even read your review in an upcoming episode.
If you have comments, questions, or complaints about this episode or the podcast, please leave a comment down below and I’ll get back to you immediately.