This shouldn’t be difficult

Researchers claim they have come up with an algorithm that can predict the sexual orientation of males with up to 70% accuracy.

I think I can do one with 97% accuracy.

Simply predict that all testees will be heterosexual. Given that 97% of men are heterosexual, this should give me a 97% accuracy, no?

31 thoughts on “This shouldn’t be difficult”

  1. “Aren’t forty-something percent of young people saying they are bisexual now”

    Yes. But as 80% of young women are amongst them, it doesn’t impact our workings.

  2. You’d get a great false-negative rate but an appalling false positive rate. Algorithm fails.

    That said, this research is probably going to turn out to be bullshit. Sexuality is too complicated and nuanced.

  3. The best algorithm is the one we all use. Assume a man is not gay (negative) but revise that to gay (positive) if he talks in a gay voice, has gay mannerisms, etc. That’s a better method than Tim’s, and picks out more gays as well.

    Of course there’ll be a few false positives because not all men who act gay are gay, and a few (perhaps more) false negatives because not all gay men act in a stereotyped way. But it will give better results than any other method.

    As to that research, though, I agree with Matthew L.

  4. SQ2: “Some people have genetically caused infertility”

    If they leave no offspring then their line ends there. So no more genetic infertility from that source. In the general population the inheritance of infertility would end all lines with such inheritance. The inheritance of lower fertility would not extinguish such lines quickly –but over time the trait would get less and less common. As homosexuality would were it genetic in cause. I don’t see the point you are making unless it is sarcasm of some kind.

  5. “How can a condition be genetic when it most strongly works against its own survival and continuance ?”

    Very easily, Mr Ecks. Because it is a by-product of a successful genetic combination. So, if a particular trait is (say) twice as effective at reproducing, but has a 10% failure rate, that “failure” will continue, even though any one “failure” will (generally) not themselves reproduce.

  6. The obvious one now I think of it is Down Syndrome. Genetic as is caused by having an extra chromosome and most people with it are infertile I believe.

  7. Valid points.

    But across human history with thousands of mostly short-lived people with brutal survival conditions the norm, even a successful breeding trait that had a “dud” branch would work against survival. If you have three kids and have to spend time, energy, resources (hunting,gathering etc) to raise 3 kids but only 2 of them reproduce themselves then your survival potential is down by a third. Across those many generations your numbers would be reduced albeit more gradually.

    Down’s Syndrome is fairly rare after all. Certainly nowhere near the 2 to 4 % of the population believed to be homosexual.

  8. Yes it is. However, any genetic cause of homosexuality (if one exists) could be a non-hereditary genetic issue like Down Syndrome is. Perhaps some mutation cause by an environmental factor or just a random mutation… who knows?

    I’d agree it surely can’t be hereditary mind.

  9. Of course it could be hereditary, as gay men have a long long history of pretending not to be gay, getting married, having kids. It’s only very recently that they’ve started living their whole lives as gay men.

    Which raises the fascinating possibility: if it is hereditary, is it now about to die out?

    I don’t believe it is hereditary, but that would be a good idea for a novel.

  10. I don’t think it could be hereditary as such*. While gay men can and do have children any gene that gives a lesser chance of the organism reproducing is going to disappear over several generations losing out to the one that makes your young man wanting to bang anything in a skirt. Well at least that’s what I remember taking from The Selfish Gene; I’m no expert.

    * I suppose it could be hereditary but also a common mutation and survive… Again I’m no expert!

  11. How can a condition be genetic when it most strongly works against its own survival and continuance ?

    Well, consider sickle-cell disease, where (to simplify greatly) single alleles increase survival and reproduction in areas where malaria is endemic, but two alleles lead to life-shortening conditions.
    As for homosexuality, I saw a suggestion a few years ago – a researcher was studying the families of homosexual men to see if their brothers / cousins (etc) were more likely to be homosexual as well. I believe that was inconclusive, but what he did note was that homosexual men seemed to come from large families, and that the cousins from their mother’s sister were also from large families. His hypothesis was that there is a genetic combination which increases an individual’s interest in sex with men. When expressed in women, this obviously increases reproductive activity, and presumably increases the spread of that gene complex. It would seem that there is more to it than that – not all men from large families are homosexual, but it seems a plausible explanation.

  12. Bloke not in Cymru

    Given historical studies that show how few men used to be able to spread their genes vs women maybe there’s some genetic usefulness in part of the population not wanting to reproduce

  13. Occam’s razor says that we’re all bisexual, and monosexualism – both hetero and homo – is a matter of psychological/cultural pressure.

  14. Dave: You have a way of starting conversations that ends conversations.

    How exactly does Occam’s Razor make this newsworthy statement?

  15. 70% accuracy? Who did the “research”, Mystic Meg? The other term for such “accuracy” is guesswork.

  16. From a natural selection perspective, you don’t have to pass your own genes on if you assist in the passing on of your relatives genes ( many of which you share with your relatives).

    Ant colonies etc work on this principal.

  17. Homosexuality could be hereditary if its expressed by having two copies of the same gene. If you only inherit one copy you can pass it one to your decendents but it doesn’t get expressed. If you mate with somebody else with one copy and your offspring inherits a copy from both parents, they will have two copies and it will be expressed.

  18. So Much For Subtlety

    Squander Two – “Of course it could be hereditary, as gay men have a long long history of pretending not to be gay, getting married, having kids. It’s only very recently that they’ve started living their whole lives as gay men.”

    Or to put it another way, homosexuality is an invention of the modern period and before it was described as a medical condition, it did not exist. Some men had sex with other men. But virtually everyone had sex with women.

    “Which raises the fascinating possibility: if it is hereditary, is it now about to die out?”

    If there is a test that is 70% accurate in utero? Certainly.

    Bloke not in Cymru – “Given historical studies that show how few men used to be able to spread their genes vs women maybe there’s some genetic usefulness in part of the population not wanting to reproduce”

    Useful to whom? Genes don’t survive if they allow mates to shag the blonde in the corner. They are, to coin a phrase, selfish.

    Dave – “Occam’s razor says that we’re all bisexual, and monosexualism – both hetero and homo – is a matter of psychological/cultural pressure.”

    Occam’s Razor would note that to all intents and purposes homosexuality does not exist, is of no interest to anyone very much, and is biologically improbable, and so would assume that some people just have a psychological problem with the female sexual organs.

    It is the obvious and most simple explanation.

    johnny bonk – “From a natural selection perspective, you don’t have to pass your own genes on if you assist in the passing on of your relatives genes ( many of which you share with your relatives).”

    Which relies on the highly dubious assumption that having a gay relative helps in any way whatsoever. Given most gays are grossed out by women and women’s biology in general, I find that hard to believe. How does having an uncle who is a friend of Dorothy help passing on your genes?

    jgh – “Homosexuality could be hereditary if its expressed by having two copies of the same gene.”

    It could be. But any combination of genes that reduced the number of offspring by as little as 1 in 10,000 – so someone with this gene could expect to have 0.01 fewer children that someone without it – would be bred out of the population over the course of a few thousand years.

  19. Bloody hell does nobody actually read the original articles. They aren’t saying homosexuality is inherited, but that environmental factors in the womb and childhood can create different epigenetic markers in otherwise identical twins.

    Tim’s method would get 100% false negatives, these guys are saying they only get 30% false negatives.

  20. Funny Tim.

    I did a paper about 15 years ago on what (statistics help) determine which baseball players will get elected to the Hall of Fame.

    You can get a very good (accuracy in the high 90s) basic model by forecasting that none of them will get in.

    For the curious, the biggest determinant is a long career: if they keep you around year after year, it’s probably because they haven’t been able to find anyone better. Stay longer than almost everyone, and you’re probably a Hall of Famer. The tough ones are the short careers, of say, Sandy Koufax.

Leave a Reply

Your email address will not be published. Required fields are marked *