Coding typing test

Type the snippet exactly as written, including indentation and line breaks. Press Enter at the end of each line.

0wpm

100% acc

0:00Time

SELECT u.id, u.email, COUNT(s.id) AS sessions, AVG(s.net_wpm) AS avg_wpm FROM users u JOIN typing_sessions s ON s.user_id = u.id WHERE s.created_at >= DATE('now', '-30 days') GROUP BY u.id, u.email HAVING COUNT(s.id) > 5 ORDER BY avg_wpm DESC LIMIT 20;

Just start typing.Tabrestarts

delete

tab

[

]

caps

;

return

shift

Left handRight hand

Snippet: Join with aggregation (sql)

Why your code speed is lower than your prose speed

It is supposed to be. English prose is roughly two percent punctuation; source code is often twenty to thirty percent symbols. Those symbols live on the shifted number row and in the little-finger column, which are the two areas ordinary typing practice never touches.

A developer typing 90 words per minute of prose and 45 of code does not have a typing problem in general. They have a symbol problem specifically, and it is fixable in a way that generic speed practice will never address.

What gets measured here

Four numbers, because improving one of them does not improve the others.

Code WPM - the same five-character convention, applied to code.
Symbol accuracy - how often you land brackets, operators and punctuation correctly.
Bracket accuracy - specifically the paired characters, which are the most commonly transposed.
Capitalisation - how reliably you produce shifted characters, tracked separately because the key statistics fold case.

The habit that costs developers the most

Pressing shift with the same hand that types the character. It works fine at moderate speed and becomes a hard ceiling above it, because the hand has to leave home position for every capital and every shifted symbol.

Underscore and the shifted number row are where this shows up worst. If your snake_case identifiers feel slower than they should, this is almost always why.

Common questions

Does autocomplete make this pointless?

Autocomplete handles identifiers, not structure. You still type the brackets, the operators, the punctuation and the capitalisation yourself, and those are exactly the characters this measures.

Should I practise code or symbols?

Symbols first if your symbol accuracy is below about 95 percent. Practising whole snippets when the underlying keys are unreliable mostly rehearses the unreliability in a more expensive way.

Keep going

JavaScript typing practicePractise typing real JavaScript: arrow functions, template literals, destructuring and nested braces.Python typing practicePractise typing real Python: colons, underscores, decorators, comprehensions and significant indentation.Symbol typing practiceTargeted practice for brackets, operators, quotes and the shifted number row - the keys ordinary typing practice never trains.The full curriculumThirteen levels from posture to professional typing.