Speedtype library

Guides based on how the app actually works

This library combines maintained product references with the Speedtype Lab. The references define scoring, language handling, and a controlled practice routine. Lab articles answer narrower questions with source-backed examples, reproduction steps, and explicit limitations. All pages are written and reviewed by Saurabh.

Core references

Start with the maintained contract

Use these pages when a result or feature needs a stable definition. They explain what is counted, which comparisons are fair, and where an informal typing tool should not be treated as a certification.

Scoring reference

How Speedtype measures WPM, accuracy, and errors

The exact formulas, correction rules, capitalization behavior, worked test cases, and known limitations.

Open reference ->

Practice protocol

A repeatable accuracy-first typing routine

A two-week routine that records a baseline, isolates errors, and measures change without promising a result.

Open reference ->

Language guide

Using English and Marathi practice modes

Input setup, comparison rules, difficulty labels, Unicode handling, and how to report a passage problem.

Open reference ->

Speedtype Lab

Worked examples and product notes

These articles document capitalization, corrections, result fields, the race bot, passage review, bilingual comparison, and a reproducible personal baseline. They do not use universal WPM rankings or promise a specific improvement.

Visit the Lab hub

01Scoring / 7 min readWhy Speedtype Treats A and a as Different KeystrokesSee exactly how Speedtype grades capitalization, spaces, punctuation, and corrected input, with a case-mismatch example you can reproduce.02Scoring / 9 min readGross WPM, Net WPM, and Accuracy: Four Worked ResultsFollow four calculations from attempted keystrokes to gross WPM, net WPM, accuracy, and unresolved errors using Speedtype formulas.03Scoring / 8 min readWhat Backspace and Corrections Change in a Speedtype ResultLearn how attempted keystrokes, unresolved errors, completed words, and the current word affect a Speedtype result after corrections.04Product notes / 8 min readInside the Speedtype Race Bot: Pace, Progress, and Win ConditionsUnderstand the simulated race bot, its target WPM settings, pace variation, progress calculation, and the exact condition that ends a race.05Editorial process / 9 min readHow Speedtype Writes and Reviews English and Marathi PassagesA transparent account of passage authorship, minimum checks, difficulty labels, English review, and Marathi native-language approval.06Practice protocol / 9 min readA Seven-Session Typing Baseline You Can ReproduceUse a controlled seven-session protocol to establish your own typing baseline without relying on universal targets or guaranteed gains.07Language / 8 min readWhy English and Marathi WPM Should Be Compared SeparatelySee how script, input method, Unicode handling, and the five-character WPM unit affect fair comparisons between English and Marathi sessions.08Results / 7 min readHow to Read a Speedtype Result Card Without Chasing One NumberInterpret speed, gross WPM, accuracy, errors, keystrokes, and time together, then decide what your next controlled run should test.

How pages are reviewed

Product claims are checked against current implementation, tests, or maintained passage records. Each Lab article names that evidence, carries a review date, and includes a practical way to verify the behavior. Substantial source changes require another editorial review.

How to use the library

Read the relevant rule before comparing scores. Reproduce the smallest example in Practice or Race. If the app and article disagree, preserve the settings, target fragment, input sequence, and result fields so the report can become a regression test.

Open Practice Report a correction