QUOTE (deja @ Oct 28 2009, 05:26 PM)

Actually it's incorrect. It's far lower than that. The paragraph lengths don't matter. There will be *some* character at the beginning of each line, so your estimate must rely solely on the character. And 26^7 is correct only if you make the assumption that every letter has an equal chance of appearing. As we know, X and Z, Q aren't very common, while F, C, T, etc. stand a much higher chance of appearing in a word. There are only a few hundred words that start with Z, and only a few words (less than 50) used practically and about 70 words that start with X, about 5 in common use, where by contrast there are tens of thousands of words that start with F and Y.
It's a very low chance, but it's not as low as you have denoted.
EDIT: Seems I was beaten to the punch by Loki, oh well here's my arithmetic anyway:
Frequency of most commonly used first letters in the written English language (Calculated from Project Gutenberg Selections):
a 8.167%
b 1.492%
c 2.782%
d 4.253%
e 12.702%
f 2.228%
g 2.015%
h 6.094%
i 6.966%
j 0.153%
k 0.772%
l 4.025%
m 2.406%
n 6.749%
o 7.507%
p 1.929%
q 0.095%
r 5.987%
s 6.327%
t 9.056%
u 2.758%
v 0.978%
w 2.360%
x 0.150%
y 1.974%
z 0.074%
So our odds become...
(0.02228)(0.02758)(0.02782)(0.00772)(0.01974)(0.07507)(0.02758) = 5*10^-12
or 1 in 200 billion. Even
more unlikely than the odds without using statistical analysis of common words through Project Gutenberg (as it turns out, all the letters used except for 'o' are below average probability for first letters).
This also doesn't factor in the fact that the two words were neatly arranged in separate paragraphs, which could add a couple more factors of 10 to decrease the odds. It also doesn't factor in the fact that short words are less likely to start a new line in a word processor than long ones. This would drop the probabilities of 'a' and 'i' and other letters more commonly found in short words and probably have a negligible or slight decrease in the probability of this being a coincidence. Then there is also a slight increase due to a handful of messages (maybe 10 or so) that would have caught our attention.
The overwhelming odds against this being a coincidence did not seem to deter Aaron McLear, a spokesman for the Governor's office. In an email responding to comments on the hidden message, he said: “It was just a weird coincidence. I suppose when you do so many vetoes, something like this is bound to happen.”