As I have said, Perez is a numerologist. Indeed, he is a very typical numerologist who could be held up as an example of what it is that numerologists do and why it's stupid. Let's look in detail at what Perez is doing, as seen in his paper "The “3 Genomic Numbers” Discovery: How Our Genome Single-Stranded DNA Sequence Is “Self-Designed” as a Numerical Whole".

**How To Do Numerology**

Like most numerologists, Perez has two basic materials. First, he has special mathematical constants, such as φ, π, e, the Fibonacci numbers, and indeed any integer whatsoever.

Second, he has a large data set of numbers drawn from the real world: in this case codon frequencies in various genomes. Of course, numerologists differ one from another by using different data sets, it could be the measurements of Stonehenge or the Great Pyramid or the solar system, the only necessity is that the data set should be reasonably large.

He is then free to form expressions by adding, subtracting, multiplying, dividing and exponentiating the special numbers; and to do the same with the numbers from the data set.

When he finds that one of the expressions formed out of the raw material of the data set is approximately equal to one of the expressions formed out of the special numbers, he can declare this to be a deep insight into biology (or, for other numerologists, whatever field they took their data set from).

**Why This Sucks**

He's playing a game he can hardly lose.

In the first place, he has wide latitude to make expressions out of his special numbers. Here are some of the expressions he gets excited about in this one paper: 1, 2, 3, 4, 5, 8, 144, 1/2, 3/2, 8/5, 5/3, φ, 2φ, 1/φ, 2/φ, φ^{2}, φ^{3}, φ^{5}, 2φ^{5}-1, 1+(φ/2), 5/2φ, φ^{1/3}, (φ+10)/9, (4/φ )-1, 5-2φ, (3-φ )/2, 2/φ^{2}, πeφ, πeφ/10, πeφ/20, 3-φ, 6/(3-φ ), 1/π, 1/φ-1/π ...

Obviously he's not limited to *these* expressions in particular, (φ+11)/6 would be every bit as good as (φ+10)/9; or 7/(4-φ ) is just as valid an expression using special numbers as is 6/(3-φ ).

But of course you can approximate *any number at all* to *any required degree of accuracy* by means of expressions in the form (φ+a)/b or c/(d-φ ) for integer choices of a, b, c, and d. There might be a practical difficulty in finding the right integers if the appropriate integers are large, but in principle it can always be done.

So his ability to find expressions of this form which match expressions made out of his data set doesn't particularly tell us anything about his data set. It tells us what any man could do if he had enough time on his hands.

Similar remarks apply to his manipulations of the data set. He's free to add, subtract, multiply, divide anything he likes by anything he likes.

But what's more, if he makes some expression out of his data set, and he *can't* get it to equal some expression made of special numbers, he doesn't have to mention that. He can just go on and try something else. He doesn't have to tell us about his failures. He awards himself a point (so to speak) whenever he can get two expressions to match, but doesn't dock a point from his score when he fails.

And he himself can see no significance in his failures. After all, he doesn't have a *theory*. Nothing *has to be* true for him to feel satisfied with his work. He is not in a position where if he divides this datum by that datum and the answer isn't φ, he has to say; "Oh, in that case I was wrong about everything, forget it".

He is, then, playing a game he can't lose. But it isn't worth playing, because, as I said, his ability to score points in this futile game tells us nothing about the data set. You could play this game and win with *any* sufficiently large data set. All Perez has proved, then, is that he's got too much time on his hands.

---

I don't think this sort of thing adds luster to the Intelligent Design movement. I guess it doesn't make it any worse either, it's all about on this intellectual level.