Comment

Michal Bartoszkiewicz

You should include the frequency of each word in the calculation – you probably treat the 't' in 'the' identically as the 't' in 'anthropomorphologically', but the former occurs slightly more often in (normal) English texts than the latter ;)
According to Wikipedia (http://en.wikipedia.org/wiki/Letter_frequencies) 't' is the second most popular letter with about 2.3 times the average.

Replies

Peter Bengtsson

yeah, you're right. I was lazy.