Gay Bar Combinatorics

[snousle]

Here's an interesting mathematical problem.

Imagine that patrons of a gay bar are either bearded (B) or shaved (S). Each person either likes beards on other men (L) or dislikes them(D). So there are four kinds of people - BL, BD, SL, and SD, all occurring at some rate. These two qualities are probably not independent of each other, nor completely correlated.

Assume that your chance of getting laid is a function of the number of people you're mutually compatible with. A BD would get along with a SL, while having only a one-way attraction with a SD or BL, and no attraction at all with another BD. And so forth, through all sixteen glorious combinations.

The question is, given the rates at which these four kinds of people attend the bar, is it an advantage for a person who likes beards to grow one, or vice versa? In other words, in order to get laid, should one become the image of what you are attracted to, even if that isn't what you would normally choose yourself? Or should you do the opposite? If large numbers of people make such a choice, does the population converge to all BL and SD over time? Maybe it could change to all BD and SL if the initial conditions were different.

It's a surprisingly complicated problem and I haven't been able to get my head around it very well. It's fortunate that sub-optimal cruising strategies are still reasonably effective.

Comments

[mudcub.livejournal.com]

I'm guessing the percentages of the gay population as something like this:

BL 0.35
BD 0.05
SL 0.1
SD 0.6

That is... the majority of guys are beardless and dislike beards on other guys. I think it would be rare to not have a beard, but like them on other guys, unless you are a bottom who is looking for a biker daddy, or for some reason such as work, unable to grow one yourself.

So, it's much better to not have a beard (with 65% of the community liking you: BD+SD = 0.65). Otherwise, with a beard, you'll only get 35% of the population.

But I really like having a beard... and I like guys with one. Put me strongly in the DL category!

[mudcub.livejournal.com]

The beardless community is much larger than the bearded community. That's my experience walking around... though certain bars change this. At the Lone Star, the statistics might be:

BL 0.7
BD 0.05
SL 0.249
SD 0.001

Note the small size of beard dislikers. Why would you go to the Lone Star if you don't like bears? For this bar, it's much better to have a beard (BL+SL = 0.949 or 95%80%). Plus, you'll probably get laid, as well.

[snousle.livejournal.com]

Particular environments are going to have really different numbers. The question is not so much about one particular place or characteristic, but the broader question of how systems like this work.

I could imagine that there might be places where bearded/shaven is sort of like gender, with overwhelmingly BD/SL pairings, and would remain that way because it would be a disadvantage for individuals to switch. Butch/femme in Latin American countries is arguably like that.

[bluebear2.livejournal.com]

So that's why it rarely happens. Oh well. At least there's beer too.

[jpeace.livejournal.com]

There's a confounding factor that you're forgetting known as the Bilinear Optimal Obfuscator - Zeta Epsilon (BOOZE). As it's evaluated to lim 03:00 AM, the resultant hookup product asymptotically approaches 1. 00

[sig-info.livejournal.com]

In the last 10-20 years, the percentage of bearded men has been stable, so it seems (discounting the next cultural vogue for facial hair) a temporary equilibrium has been reached on the B/S axis. The question is really whether BL predominates over BD (or SL over SD, depending on your preference). That's amenable to study in the field—prepare your research grant proposal now, before everyone's doing it.

Of course, that's ignoring the psychological aspects of B/S. Some people won't change to please others.

[furr-a-bruin.livejournal.com]

Just because I like playing with tables and sometimes it's interesting to visualize such things....

	BL	BD	SL	SD
BL
SL
BD
SD

Color Key
Mutual Attraction	One Way - Column to Row
One Way - Row to Column	No Attraction

[sig-info.livejournal.com]

I should add: this isn't really combinatorics, it's population dynamics. Probably wouldn't take someone trained in the field long to whip up some diffyq's to model the problem. I wonder there's some curve in the BS/LD plane that separates different attractive basins, and what such a curve would look like. Hmm...

[furr-a-bruin.livejournal.com]

Your initial presentation assumes getting laid is the only motivational driver... even amongst gay men, that's not necessarily true. As sig_info points out, there are men who will not change their B/S state for internal reasons, however it might affect their chances of getting laid. I'm certainly one of those - I've been continuously bearded since I was first able to grow a decent beard as a teen, and the only reason that might ever change is medical necessity.

[snousle.livejournal.com]

All models are inadequate. But if you don't start by trying to understand a simplified model, understanding a more realistic one isn't going to happen on its own!

[furr-a-bruin.livejournal.com]

Quite so. I just found the underlying assumption that getting laid is all that matters somewhat amusing. ;)

Of course, trying to generalize from ME would lead to a massively erroneous model for a wider population....

[bluebear2.livejournal.com]

On the topic of facial hair. Check out this site I stumbled on to: (I don't know how I got there. Honest.)

http://uwf.edu/wlees/CREWPHOTOS.html

[backrubbear.livejournal.com]

This feels simple but I'm probably looking at this a slightly different way than you are.

Presume that L or D is a binary and invariant quality.

The property of a given person wanting to get laid, B or S is not stated as a preference/like property, it's simply stated as a current state.

Given that, for a single individual in a given population of L and D to increase their chances of getting laid they'd simply become B or S as appropriate. This presumes there's a way to determine L or D; perhaps presume people flag one way or the other.

Given these inputs and the ability to instantly become bearded or shaven, then system would converge instantly to the optimum laid case.

I'm sure that's not the system you were thinking of. It's more complicated than that. Change the presumptions a bit and the complex system becomes more interesting:

Presume a fixed population with initially no knowledge of a given person's L/D.
Presume no flagging.
Presume a limited ability to discover L/D in another person - one that is not influenced by your current B/S state. Presume also an infinite memory to recall a given person's L/D.
Presume still the ability to instantly be bearded or shaven.

It will take some number of rounds for the system to converge but it should still optimally converge but it's still not terribly realistic.

Presume the system is not closed and at any given round some percentage of the population is replaced at random with a different L/D. I don't believe the system converges even in astable state.

However, if you adjust the system so population is fixed but there are a number of locales, I suspect you'll end up with some astable convergence for different local optimums.

The above is at least testable using cellular automata. I don't have the clue to be able to predict system convergence from the above properties without running the simulations. :-)

Even more frustrating is that:
- People don't have perfect memory.
- L/D aren't usually binary qualities. I might like a guy with a little fuzz while you might prefer guys that look like billy goats. Our preferences are probably ranges with perhaps notch points.
- B/S similarly aren't binary qualities.
- You can adjust your state to S pretty much instantly but B takes time and thus would bias convergence of the system.
- People have a personal preference for B/S. Adjusting for that degree of preference in the CA would be interesting.
- Mobility among locales will bias people with low personal preference B/S only if they have knowledge of the current state of the locale.

This might be amusing to model at some point.

AAKK!

[beastbriskett.livejournal.com]

No wonder you like the green. It slows the cacophony in your brain ;)

[snousle.livejournal.com]

I'm struck by the symmetries that come out when you rearrange the order like that. That had not occurred to me.