**T**oday’s letter-to-the-editor appears in the *High Point Enterprise* of High Point, North Carolina, which is sometimes referred to as the “Furniture Capital of the World.” The letter is titled **The math of evolution doesn’t add up**. We always look for titles like that, and when we find one we’re never disappointed.

We don’t like to embarrass people (unless they’re politicians, preachers, or other public figures), so we’ll omit the writer’s full name and city. We will mention that his first name is Fred. We Googled for his name, and he may be the retired President of a furniture company. We’ll give you a few excerpts from his letter, enhanced with our Curmudgeonly commentary, and some bold font for emphasis. Okay, here we go:

I read with interest professor Mark Venable’s guest column on presenting the science for evolution.

Fred is talking about this: Evolution theory is backed by much evidence. We previously wrote about an earlier response to Venable’s column — see Creationist Wisdom #387: Tea Party Genius. Now it’s Fred’s turn to tell us what he thinks of it. He says:

The sciences (chemistry, physics, biology) can build a myriad of theories of the science of evolution. Science changes with time.

No, Fred, they can’t build “a myriad” of theories — not in this universe. Only one theory is consistent with all those sciences, and that’s the one we’ve got. And yes, unlike creationism, science *does* change as we learn more, which is why it keeps getting better. A “theory” that hasn’t changed in 3,000 years is — shall we say — perhaps a bit outdated. Let’s read on:

Did you ever notice the are few (if any at all!) mathematicians on the evolution bandwagon?I’ll tell you why.

No, we haven’t noticed that, but, here comes Fred’s answer to a non-existent problem:

Mathematically, a single person has a better chance of winning the Mega Millions lottery 50 times in a row than evolution occurring.

** BWAHAHAHAHAHA!** We started this blog with a three-part series of posts on this math “problem,” starting here: The Inevitability of Evolution (Part I). Fred continues:

Let’s assume that 1,000 events (events, mutations, evolutions) had to occur for me to evolve from some primordial goo.

It’s far more than 1,000, Fred, but let’s see where you’re going with this:

Each event had a 50/50 chance of success. Like flipping a coin, 50 percent chance of a head.Therefore, the odds are one-half times one-half and so on, a thousand times.Like flipping 1,000 heads in a row.Don’t try this on your home PC because it will probably blow up !

Actually, Fred, most mutations are useless, or harmful, so the odds against a “good one” are worse than 50-50, but ignoring that, your coin-tossing example makes sense only if you were to evolve all at once out of the goo. But you didn’t. Nothing did. The process took a few billion years, with uncountable zillions of “coin-tossing sequences” that produced failures and dead ends along the way. A few sequences did work out, and we are the result; but considering how many coins were being tossed, every hour of every day for billions of years, the result isn’t surprising.

And now we come to the end:

You can go with the science if you want to. I will stick with God and the math.

Have it your way, Fred. And thanks for the letter!

Copyright © 2014. The Sensuous Curmudgeon. All rights reserved.

I wonder if our math genius has tried to calculate the probability of his own existence, just from the possible combination of sperm and ova from his parents, let alone if we stretch that back through history. Why, it’s impossible that he’s here at all!

“Why, it’s impossible that he’s here at all!”

He probably isn’t.

Ah yes, the old improbability-of-producing-

-result canard. Quite apart from the obviously ignored fact that at any point there is a veritable myriad (to steal the letter writer’s word) of temporarily viableX’s as manifested by the rich diversity of life forms, there is also the observation that biological evolution involves “massive parallel processing” (to steal a computing science term) and hence proceeds undaunted by the “law of large numbers” (to steal a statistics concept).XMaths genius, indeed.

RationalWiki.com “Probability Misuse by creationists”

I love these “ah ha, gotcha” arguments. Without any awareness of how it can be reversed against their position. With creation, there is only one chance for everything to get right (well the flood is kind of a fix of stuff that didn’t), without the benefit of ‘parallel processing’. One deity had six days to get everything right while creating billions of stars, starlight, planets, atmospheres, and then thousands of creatures on just one of those planets. And two sentient beings with souls. Your move Fred.

Charles Deetzsays: “Your move Fred.”Fred has already made his move, and he will avoid the Lake of Fire. You, however …

“the[re] are few (if any at all!) mathematicians on the evolution bandwagon”

[citation needed]

TPKasks for a citation from Fred. He doesn’t give one, but will a counter-citation suffice? I went to NCSE’s list of Steves, and stopped counting at 20 — the last name I got to was Stephen Hawking.The Disco’Tute has its own in-house maths genius, the ineluctable Denyse O’Leary, who is making pretty much the same point, but possibly even more absurdly: Can All the Numbers for Life’s Origin Just Happen to Fall into Place?

Among other gems therein, she says of the “genetic code”:

Unlike other languages? So who invented English? Or Mandarin? Or…

Megalonyxmentions a post by Denyse O’Leary.I saw that, and she seems to be a rising star in the Discoveroid firmament. But her stuff is almost always too goofy, even for the kind of material with which I customarily deal.

Has anyone even attempted this calculation, given for one generation there are several hundred thousand eggs from the mother and millions of sperm per each spermatogonial cycle from the father; then per generation; yes, it’s a miracle that any of us are here 🙂

Douglas Easks: “Has anyone even attempted this calculation”I’ve posted about it before — see William Dembski’s Design Inference, but I never did the calculation. Quoting myself:

The average for a healthy young male is estimated to be 300-500 million spermatozoa, per, ah … event. To be on the conservative side, let’s say that a specific human zygote has less than a one-in-100 million chance of being conceived. And that’s for one particular fertile moment for the female. A month earlier or later, the zygote will be different. In other words, dear reader, considering the odds against your turning out to be precisely you, it’s obvious that your existence is quite improbable. Nevertheless, there you are.

The same improbability analysis applies to the conception of each of your parents, and their parents, and so on, going back as far as you care to go. The odds against the whole multi-generational drama is a factorial computation, with the mathematical conclusion that your existence is so very improbable as to be virtually impossible — by Discoveroid reasoning.

Maybe I can have one of my math colleagues give this to the class as a “homework” assignment 🙂

Douglas E, don’t be surprised if he doesn’t do it. It’s really a trivial calculation. It all depends on how many generations are included in the calculation. And even with only one generation, how do you decide which month is the one to use?Don’t forget that females start life with several million eggs. Even if they’re not all released at once, it’s a one in X million chance that any particular one will be released any given month.

Well darn,

Christine Janis, I never thought about that. I was just assuming that one egg per month was a given, and the variables were on the male side of things. You have expanded my worldview.Well, I learnt that it was a few hundred thousand – but a million is easier to work with. You would have to calculate a one in a million chance of any given egg for the fertile lifespan of a woman, say every 28 days for 40 years. So just the female number is quite big, and the male number will be much larger given the sperm ‘per event’ as Curmy says. Yes, it’s a fairly trivial calculation but it will give a very big number for the probability of just one person. Then times x, x, x, x…..

All in day’s work, Curmie!

Douglas

It depends how you’re counting. This is from a human fertility site

A baby girl is born with egg cells (oocytes) in her ovaries. Between 16 and 20 weeks of pregnancy, the ovaries of a female fetus contain 6 to 7 million oocytes. Most of the oocytes gradually waste away, leaving about 1 to 2 million present at birth. None develop after birth. At puberty, only about 300,000—more than enough for a lifetime of fertility—remain.

Aha 🙂 In my original comment, I was using the 300,000 available for ovulation.

Despite the lifetime odds of one in 300,000 for the egg, the fact that there are 300 million to 500 million spermatozoa wiggling toward that one egg sort of overwhelms the odds against the egg, The odds against that one lucky sperm are more than a thousand times greater than the odds against the egg.

Never mind that those ‘events’ usually occur more than once during a cycle, so that number will need to be multiplied by how excited said species is.

Yes, but to produce a

specificindividual requires justthisegg to be fertilised by exactlythatspermatozoon. Statistically speaking, these are compound probabilities because the selection of the egg and the male gamete that, er, comes first are events that are independent of one another (as far as we know). Thus, the chance of your parents producing preciselyyouis 1/300,0001/400,000,000, or about one in 120 trillion, so you’re even more improbable than you might be comfortable with. Still and all, the resolution to this apparent conundrum has its roots in something called the lottery paradox, so you could still reason that you, as yourself, are here by virtue of having won the Great Celestial Lottery oftimes… 🙄Oogity-Boogity!It all depends on how many generations are included in the calculation.The calculation is depends exponentially on the number of generations, but only linearly on the survival rate per generation. It you carry out the calculation back a mere few millennia, it doesn’t make any difference whether the number of gametes is 10 or 100 million.

>Don’t try this on your home PC because it will probably blow up !

Unless you are actually a mathematician and know that calculating log-likelihoods is both convenient and useful, with the added benefit that your computer doesn’t blow up. I have on screen in front of me at this very moment a log-likelihood of -1560.8, far below Dembski’s Universal Probability Bound and therefore impossible, and yet there it is.

The probability is actually irrelevant. Such arguments have an implied assumption that Design is the only possibly conclusion.

Whoa! All those sciency calculations on the probabilities of sperm and ova are impressive. But they omit consideration of the probability that your father may have just been a friend of the family, or that the hospital nursery gave your parents the wrong infant to take home.

Your calculations are all wrong. Many possible egg/sperm combinations would have the same combinations of chromosomes, i.e. be in effect twins of each other. Thus what you care about are UNIQUE combinations of chromosomes.

Each parent has 23 pairs of chromosomes and each gamete gets one chromosome from each pair, the odds of that being 50-50 for each chromosome per pair. So each parent can produce 2^23 = 8.4 million unique gametes. This assumes each chromosome in each parent is heterozygous in at least one allele, that is, the the two chromosomes in each pair are non-identical, which is highly probable unless you’re really inbred. This ignores chromosomal crossover, which is rare but increases the unique combinations. Each male emission should contain several copies of all possible unique sperm. The ovaries of a female fetus would contain most possible unique oocytes before they start dying off.

The number of possible unique sperm/egg combinations from a given father and mother would be, naively, 2^23 squared = 7 *10^13 = 70 trillion.

This number would be increased by crossover, and decreased if either parent is highly inbred (has a pair of identical chromosomes) or if the father and mother share an identical chromosome (possible if they’re related.)

Diogenessays: “Your calculations are all wrong. … This assumes each chromosome in each parent is heterozygous in at least one allele, that is, the the two chromosomes in each pair are non-identical, which is highly probable unless you’re really inbred.”That might explain the fact that in parts of the country where it’s common for one’s mother to also be one’s aunt, they ain’t no kin to no monkey.