Editors Reads
Algorithms to Live By by Brian Christian and Tom Griffiths — book cover
Editor's Pick intermediate

Algorithms to Live By — The Computer Science of Human Decisions

by Brian Christian and Tom Griffiths · Henry Holt and Co. · 368 pages ·

4.4
Reviewed by Daniel Fry

Computer science algorithms offer surprisingly practical guidance for everyday human decisions — from optimal stopping to the explore-exploit tradeoff to how to sort your email.

Check Price on Amazon (paid link) Opens Amazon · Prices subject to change

Editors Reads Verdict

A rare book that takes ideas from computer science and applies them rigorously and usefully to human decision-making. The optimal stopping chapter alone — including the 37% rule for apartment hunting, hiring, and relationships — is worth the price.

4.4
Check Price on Amazon (paid link)

What We Loved

  • The 37% rule for optimal stopping is immediately applicable to real decisions
  • Explore vs. exploit gives a mathematical grounding for why novelty matters more when young
  • Explains complex algorithms accessibly without dumbing them down
  • Each chapter stands alone as a useful idea

Minor Drawbacks

  • The human-behaviour examples don't always map cleanly onto the algorithmic analogies
  • Some chapters feel more like standalone essays than a unified argument
  • Mathematical rigour occasionally sacrificed for accessibility

Key Takeaways

  • The 37% rule: spend 37% of your search window observing, then commit to the next option that beats all previous ones
  • Explore vs. exploit: early life favours exploration; later life favours exploiting known goods
  • Sorting is expensive — often it is better to leave things unsorted and search when needed
  • Forgetting is not failure — it is the brain's least-recently-used cache operating correctly
  • Randomness is sometimes the optimal strategy when the cost of computation exceeds the value of a perfect answer
Book details for Algorithms to Live By
Author Brian Christian and Tom Griffiths
Publisher Henry Holt and Co.
Pages 368
Published April 19, 2016
Language English
Genre Technology, Science, Psychology
Difficulty Intermediate
Best For Anyone curious about how mathematical thinking can improve practical decisions, and readers who want to understand what computer science has to say about life choices.

Algorithms as a Guide to Life

Algorithms to Live By begins from a deceptively simple observation: computers face the same fundamental challenges as humans — how to allocate limited time, how to sort through information, when to stop searching and commit. The solutions computer scientists have developed for these problems are mathematically optimal, and Brian Christian and Tom Griffiths argue they are also practically useful for humans facing equivalent choices.

The result is a book that is more rigorous than most popular science writing and more readable than most academic work — a tour through twelve algorithmic ideas and their human applications.

The 37% Rule

The book’s most immediately actionable idea comes from the optimal stopping problem: how long should you look before deciding? The mathematical answer — derived from the secretary problem in probability theory — is the 37% rule. Spend the first 37% of your available time or candidate pool purely observing, then commit to the next candidate who exceeds everyone you’ve seen so far.

This strategy maximises the probability of choosing the best available option and applies cleanly to apartment hunting, hiring decisions, and the question of when to stop looking for a better option in any domain. The precision of the answer is itself the point: optimal stopping has been worked out, and the answer is not “as long as possible” or “trust your gut.”

Explore vs. Exploit

A second central idea addresses the tension between trying new things and returning to what you know works. Multi-armed bandit theory — developed originally for clinical trials and casino mathematics — shows that the optimal strategy shifts with time horizon. When young and with a long future, exploration makes mathematical sense. As time runs short, exploitation of known goods becomes dominant. This gives a formal justification for why settled preferences are rational with age rather than merely conservative.

Final Verdict

Algorithms to Live By is the book that makes readers want to learn more mathematics. Its central insight — that optimal decision-making has been formalised in many domains and the answers are available — is both humbling and useful.

Our rating: 4.4/5 — One of the best popular science books of its decade. Practical, surprising, and genuinely illuminating about how to think.


Reading Guides

Frequently Asked Questions

What is "Algorithms to Live By" about?

Computer science algorithms offer surprisingly practical guidance for everyday human decisions — from optimal stopping to the explore-exploit tradeoff to how to sort your email.

Who should read "Algorithms to Live By"?

Anyone curious about how mathematical thinking can improve practical decisions, and readers who want to understand what computer science has to say about life choices.

What are the key takeaways from "Algorithms to Live By"?

The 37% rule: spend 37% of your search window observing, then commit to the next option that beats all previous ones Explore vs. exploit: early life favours exploration; later life favours exploiting known goods Sorting is expensive — often it is better to leave things unsorted and search when needed Forgetting is not failure — it is the brain's least-recently-used cache operating correctly Randomness is sometimes the optimal strategy when the cost of computation exceeds the value of a perfect answer

Is "Algorithms to Live By" worth reading?

A rare book that takes ideas from computer science and applies them rigorously and usefully to human decision-making. The optimal stopping chapter alone — including the 37% rule for apartment hunting, hiring, and relationships — is worth the price.

Ready to Read Algorithms to Live By?

Check the current price on Amazon.

Check Price on Amazon (paid link)

Prices and availability are subject to change. See Amazon for current price.

Affiliate Disclosure: As an Amazon Associate I earn from qualifying purchases. Clicking Amazon links and purchasing may earn us a small commission at no cost to you. Our reviews are editorially independent — affiliate relationships do not influence our ratings or recommendations. Product prices and availability are subject to change; see Amazon for current pricing.
#algorithms#decision-making#computer-science#mathematics#cognitive-science

Review last updated:

Skip to main content