How We Measure Typing Speed

English words range from single characters ('a', 'I') to multi-syllable terms ('unbelievable'). If typing tests counted literal whitespace-separated words, someone typing simple sentences would appear twice as fast as someone typing scientific literature. Standardizing one word to exactly five keystrokes (including letters, punctuation, and spaces) creates an objective, mathematically uniform metric comparable across any language or passage.

Raw WPM (Gross WPM) measures your total mechanical keystroke speed divided by five per minute, regardless of typos. Net WPM only counts characters that were typed accurately. Net WPM is the universally accepted standard for employment and civil service examinations because erroneous keystrokes do not produce useful work.

During your test, our engine takes an instantaneous Net WPM snapshot every single second. At the end of the test, we calculate the standard deviation and coefficient of variation (CV) across all snapshots. Consistency equals 100 minus CV. A score of 90%+ means you maintain a smooth metronomic rhythm, while lower scores reflect sprint-and-stall typing.

Backspace keystrokes do not count toward your correct character total, and correcting errors consumes valuable elapsed seconds. The 'accuracy tax' means fixing a single mistake typically costs 3 to 5 keystrokes (the typo, backspacing, and retyping). Maintaining 98%+ accuracy is almost always faster than typing fast with errors.

Our client-side engine records timestamped keystroke latencies and bigram transitions (such as H→E or T→H). After every completed test, it computes your slowest individual reach, your slowest two-key transition, and categorizes delays between wrong keys and hesitation pauses, giving you an exact bottleneck in plain English.