S t a t i s t i c s
By Andrew Harris 6 December 1999 ©
Ivan Panin was probably the earliest writer to mention statistics in relation to Bible numerics. However his treatment of statistics has left a lot of misconceptions in its wake. For this reason I intend to examine and correct what Ivan Panin did before going on to some of the more recent work that has occured. This should also serve as a reasonable introduction to proper statistical analysis for the average reader.
Ivan Panin's Numerics
One of the most widely quoted of Ivan Panin's examples is the features of seven in the first verse of Genesis. He usually only mentions a list of 14 features of seven in Genesis 1:1, however in various of his writings there are different features mentioned in the list of 14. The following, is a composite list from several of his writings.
Ivan Panin would then say that the chance twenty features all being multiples of seven is one in 7 x7 x7 x7 x7 x7 x7 x7 x7 x7 x7 x7 x7 x7 x7 x7 x7 x7 x7 x7. Or one chance in 720. Or one chance in 79,792,226,297,612,001 or approximately one chance in eighty thousand million millions. With such a small chance of occurring randomly Mr. Panin would claim that it must be design.
Can you see the problems with the list? Lets look at feature 12, the numeric value of 777. Feature 13 then says that this has a 7 for the units part of the number. Feature 14 then tells us we have 7 tens and feature 15, 7 hundreds. But, hang on, we already know this, that is what the number 777 means. In effect we are being told the same thing twice. Really we should not count the 7 units, 7 tens and 7 hundreds as separate facts. So Ivan Panin's list should be shortened to only 17 features of seven.
In mathematics this need to not repeat information when you are counting up is called independence. All the facts need to be independent. For example if we know that a + b = 10 and we know that a = 6, then we also know (or can figure out) what b is. Thus even though there are three facts in this equation, a = 6, b = 4 and a + b = 10, only two of them are independent. Another way of saying this would be: a and the total are free (free of restrictions, independent) to be any value they want. But as soon as they are decided, then b is fixed, b is not free to be any value, but must be one specific value. It does not matter which two of the three facts are independent, the third is always dependent.
Lets go back to Mr. Panin's list, there are more features that repeat information we already know. Feature four is already known from features 1,2 and 3. Feature eight is already known from feature seven.Feature eleven is already known from features nine and ten. Feature twenty is already known from feature eighteen and nineteen. So if we remove all the repeat information we have only 13 features of seven. So, if we calculate in the same way as Mr. Panin, one chance in 713 or one chance in 96,889,010,407 is still good odds for design, right?
Now lets look at how Mr. Panin calculates his statistics. It is true that the chance of 13 features all being multiples of seven is one in 96,889,010,407. But only if they are the only features to be found! Clearly they are not. In the above list of features, why is the numeric value of the entire verse never mentioned? The numeric value of the verb is mentioned, the numeric value of the words which are not nouns is mentioned. The numeric value of the first and last letters of each word is mentioned, the numeric value of the first middle and last letters of the verse is mentioned. There are some glaring omissions. The numeric value of the nouns is not mentioned, the numeric value of the first and last word is not mentioned. Is it possible that these are not mentioned because they are not features of seven?
So, we have thirteen features of seven, but many other features that are not features of seven. Mr. Panin not only does not mention the features which are not multiples of seven, he does his statistical calculation as if they did not exist. This is clearly wrong. How then should we calculate the odds of these features of seven occurring by chance alone?
In a systematic examination of 41 independent features in this verse, 14 independent features of seven were found. That means 27 independent features were not features of seven. So what is the chance of getting 14 features of seven out of a total of 41? Fortunately there is a mathematical formula that will calculate this for us. We will look at the formula and the calculation in detail later, but for now lets just go to the answer.
It turns out that the chance of this many features of seven in the 41 we examined is one chance in 7,143. This is still quite remarkable, and good evidence of design rather than random chance. However it is a very different number to what Mr. Panin would have calculated. He would have said it was one chance in 714 or one in 678,223,072,849.