Textalyzer, a free online tool, combines Microsoft Word-style readability statistics with detailed “instance counting” for up to (it says) 1000 words. That means it will display a table with up to the 1000 most commonly used words in your document; I doubt anyone would ever need that many, but it’s there if you need it.
Textalyzer also can count instances of the most commonly used two- and three-word phrases, tell you the average syllable count of a text, and search (and count) an arbitrary phrase.
The interface takes a moment to get used to, and Textalyzer doesn’t work well with blogs. Since post titles, etc, often display multiple times on a page (repeated in sidebars and so on), wordcount rankings aren’t accurate. But by cutting and pasting sections of a novel, story, or article into Textalyzer, you can get an idea of what words you use (or overuse).
Somewhat relatedly, has anyone used this?