Reading ease, measured honestly
These formulas count syllables and sentence lengths. They do not read — which is exactly what to keep in mind.
Reading ease, measured honestly
These formulas count syllables and sentence lengths. They do not read. A clear sentence full of long words scores badly and a meaningless one full of short words scores well — which is exactly what to keep in mind.
What each formula counts
Flesch Reading Ease = 206.835 − 1.015 × (words / sentences) − 84.6 × (syllables / words). Higher is easier; 60–70 is described as plain English.
Flesch–Kincaid Grade = 0.39 × (words / sentences) + 11.8 × (syllables / words) − 15.59, expressed as a US school grade.
Gunning Fog = 0.4 × ((words / sentences) + 100 × (complex words / words)), where complex means three or more syllables.
Syllable counting is approximate. It uses vowel groups with corrections for silent e and common endings, which is what every implementation of these formulas does. It is wrong on some words, and every score here inherits that.
The formulas cannot read
All of them are arithmetic on word length and sentence length. Neither is meaning. "The cat sat on the mat" scores beautifully; so does a paragraph of nonsense made of short words.
Their real use is as a smell test on your own writing: a Fog index of 18 usually means the sentences have grown, and shortening them usually helps. Writing to the score produces choppy prose that reads worse than what you started with.
WCAG 3.1.5 is AAA, and narrower than it looks
It asks that where text requires reading ability beyond lower secondary level, a simpler alternative is available. It does not require every page to hit a grade level, and no automated score demonstrates compliance.
Plain language is not short words
It is one idea per sentence, the actor before the verb, and a term defined the first time it appears. Those are the changes that help, and none of them shows up in these numbers.