Python Crash Course, 3rd Edition: a review with no score, on purpose

Python Crash Course, 3rd Edition is the book most people mean when they say they bought a Python book. This is a review of what it is, what it costs and what is in it — read off its publisher's page on 24 September 2026 — and it deliberately carries no score, for a reason set out at the end.

The specification

• Author: Eric Matthes.
• Publisher: No Starch Press.
• Published: December 2022, third edition.
• Length: 552 pages, lay-flat binding.
• ISBN-13: 9781718502703.
• Price: $49.99 for the print book, which includes a free ebook, or $39.99 for the ebook alone in PDF, Mobi and ePub.

No Starch describes it as "the world's best-selling guide to the Python programming language". That is the publisher's claim about its own book, and it is repeated here as a quotation rather than as a fact we verified.

What is in it

The book is built in two halves, and the second half is the reason to buy it.

The first half is the fundamentals: variables and data types, lists, dictionaries, conditionals, loops, functions, classes, files and exceptions, and testing your code. This material is not scarce — freeCodeCamp covers the same ground in 540 free steps, and covers it interactively.

The second half is three projects carried through to something that runs: an arcade game, a set of data visualisations, and a web application with user accounts that you deploy. That is where the book earns its price. Courses are good at exercises and poor at endings, and this is a book almost entirely about endings.

The publisher lists the third edition's updates as including VS Code for editing, the pathlib module for file handling and pytest for testing — which tells you how the tooling has moved rather than the language.

What the December 2022 date means

Paper does not update. That is the trade you accept with any book, and it is worth being concrete about it rather than nervous.

The Python fundamentals in the first half have not moved in a way that breaks the book — the language is stable and deliberately so. What will drift is everything around it: editor screenshots, library versions, the deployment step in the web project, and the exact behaviour of third-party tools. When you meet a discrepancy, the documentation is right and the book is old, and knowing that in advance saves an evening of doubting yourself.

There is also a purchasing trap: the first and second editions are still on sale, and their listings look almost identical. Check that the page you are buying from says third edition before you pay.

Who it suits

It suits someone who has done some interactive learning, can write a loop, and has noticed that they have never finished anything. The three projects are the cure for that, and they are laid out at a pace that assumes you will get stuck.

It suits someone who reads better away from a screen, which is a real and undervalued preference.

It does not suit someone who has never written a line of code and wants their hand held through the first fortnight. A book cannot tell you why your particular code is broken, and that feedback loop is exactly what a beginner needs most. Start somewhere interactive and free — the free Python routes worth your time covers them — and buy this when you want structure rather than reassurance.

The free alternative, stated plainly

Automate the Boring Stuff with Python is free to read online under a Creative Commons licence, in its third edition, with a free workbook companion. For a beginner it is arguably the better first book, because its examples are chores you actually have rather than programs written to demonstrate a concept.

The two books are not really competitors. Automate the Boring Stuff answers "what is Python for?" and Python Crash Course answers "how do I finish something?". If money is tight, read the free one, do a free course, and buy the paid one later if you find you want the projects. Nothing about that order makes you a less serious learner.

Why there is no score on this page

Every review elsewhere on this site puts a number on the platform it covers, because those numbers score something specific and checkable: whether a company is straight with you about money. We opened those pricing pages, clicked every billing panel and counted.

We did not read 552 pages. So what you have above is the book's specification, its publisher's own description, its contents as listed, and our judgement that the structure suits a particular kind of reader — with the judgement labelled as a judgement. Putting a 7.5 at the top of this page would have looked more authoritative and meant less.

That is the same rule that keeps salary figures, completion rates and time-to-employable claims off this site entirely, and that governs how every price here is read and dated.

The three projects, and which to do first

If you buy this book for the second half — and that is the reason to buy it — the three projects suit very different people, and you are not obliged to take them in order.

The arcade game is the most fun and the most self-contained. It teaches you to hold a program with several moving parts in your head at once, which is a genuinely new skill after a diet of exercises. Start here if you have never finished anything.

The data visualisations are the most immediately useful to people who came to Python for work reasons: the output is a chart you can show someone, and the loop from data to picture is short enough to stay motivating.

The web application is the most valuable and the most likely to age. It builds something with user accounts and deploys it, which is the closest the book comes to real work — and it is the chapter most exposed to that December 2022 date, because deployment platforms redesign themselves constantly. Expect to translate a few steps, and treat the translating as part of the exercise rather than as the book letting you down.