Unless noted as otherwise, the odds presented on the Codes I have found are the approximate odds of finding just the main term in a given block of text.
The odds are approximate in that they do not take in account the individual odds of the different letters. For example some letters are used much less often than others in the visible text (just as in English - the letter 'z' is used much less than the letter 'e') and in the terms searched for. In an extreme example, the odds on the term "zzzzz" are the same as the term "eeeee" when searched within the same range block of text, because they are both 5 characters.
"The Keys to the Bible" simply figures the odds of finding a term using a particular number of letters within a text block of a particular number of random characters in order to arrive a the "Standard Deviation".
In some cases the odds of the main term and the additional terms showing together, and takes into account the proximity of the two (or more) terms. If this is the case, it will be noted.
I think I only have two matrices where the odds of multiple terms is calculated, this is in the second and third Codes in this paper:
Which has the odds of finding the term "Meraioth" (a Biblical meaning 'rebellion') showing 6 times at a 2 skip in that matrix, right next to one of the seven times that name shows in the visible text, an R factor of 13.63209 = 1 in 42,863,372,988.234.
The term actually shows 15 times in that entire matrix which The Keys to the Bible gives an R factor exceeding 29 - very unlikely to happen by random chance.
The third matrix in that paper has an R factor of 5.57980 = 1 in 380,010.256 of being random.
"Standard Deviation is the measure of the variability (dispersion or spread) of any set of numerical values about their arithmetic mean" - manual to The Keys to the Bible
The Standard Deviation is the difference between the expected Number of occurrences and the actual number occurrences.
Standard Deviation = (expected number of occurrences - the actual number found) square root of expected number of occurrences
Here is the example presented in the manual for The Keys to the Bible:
Expected number of appearances: 4 Minus the actual number found: 3 Difference between expected and found: = 1
square root of the expected occurrences √ 4 = 2
difference between expected and found 1 square root of expected occurrences 2
Equals a Standard Deviation = 0.50
A low Standard Deviation, such as 0.50 is indicative of a random occurrence.
As the Standard Deviation rises above 1 the odds increase exponentially.
Standard Deviation Approximate odds
0.50 = 1 in 2 1.46 = 1 in 5 1.74 = 1 in 10 1.99 = 1 in 20 2.10 = 1 in 30 2.53 = 1 in 170 2.90 = 1 in 300 3.62 = 1 in 5,000 4.20 = 1 in 50,000 4.57 = 1 in 500,000
By the time the Standard Deviation approaches 5 the odds of it being a random occurrence surpasses 1 in 1 million, and the program ceases to calculate, and just leaves it at "less than 1 in 1 million" of being a random occurrence.
In order for a search to have statistical value, from a scientific point of view, the search must be done 'a priori', that is, all terms must be decided upon in advance and the spelling must be decided in advance. And the search range must be decided in advance.
Any terms or extensions observed after examining the Code do not have statistical value from a scientific point of view.
Of all the Codes I have found the least likely to be a random occurrence is "Peter the Roman" searching in the Torah. The Standard Deviation of this Code, done 'a priori', is 20.14, surpassing 1 in 100 million of it being a random occurrence.
"In a nutshell" if the odds are 1 in 1000 then it should be necessary to run 1000 searches in random or "monkey text" to duplicate the find.