The 4 Big Myths of Profile Pictures

January 22, 2010 2:11 PM

"In looking closely at the astonishingly wide variety of ways our users have chosen to represent themselves, we discovered much of the collective wisdom about profile pictures was wrong."
posted by Groovytimes (51 comments total)

*goes and hugs wife gratefully*
posted by Burhanistan at 2:14 PM on January 22, 2010 [10 favorites]

One my my current profile pics. I'm not sure if this is flirty face or unsmiling?
posted by mullingitover at 2:18 PM on January 22, 2010 [5 favorites]

Part of me assumed that the 'collective wisdom' about profile pictures had more to do with quality of contacts than quantity.
posted by Adam_S at 2:19 PM on January 22, 2010

ahem
posted by theora55 at 2:26 PM on January 22, 2010

theora55, you realize it's not actually a dupe, right?
posted by 0xFCAF at 2:27 PM on January 22, 2010

"she actually gets the most messages by flirting directly into the camera"

y' don't say.
posted by longsleeves at 2:30 PM on January 22, 2010

I find these data dumps are endlessly fascinating, and all the more so because the authors are examining real users.

If only someone could run a similar analysis for Craigslist casual encounters. Did you know that posters who say "I'll let you shit in my hair" get twice as many replies as those who post a picture of themselves with their pet (unless they're having sex with their pet)?
posted by Admiral Haddock at 2:34 PM on January 22, 2010 [7 favorites]

Every time one of these OK Cupid posts appears, I want to hate it:

"Another one of these? Don't these people have anything better to do than to post crappy long-ass entries about dating infometrics?"

*skims blog post*

"Hey, this is interesting."

*keeps reading*

"Hey! Boobs and Cleavage!"

*more reading*

"Huh. Cool."

But they really are interesting, and often surprising. Only 4% of men and 9% of women of women smile in their profile pictures? Why wouldn't people be more apt to smile and make eye contact with the camera if they're trying to meet people to hook up with? The freakin' MySpace shot is what brings all the boys to the yard, so to speak? Should I update my MeFi profile pic to be trendy? :D
posted by zarq at 2:34 PM on January 22, 2010 [1 favorite]

What about having a gun in my profile pic?
posted by qvantamon at 2:37 PM on January 22, 2010 [2 favorites]

I read this earlier today and thought it was fascinating. It's funny how well the male shirtless photos seem to do despite how often women's profiles say they don't like them.
posted by ludwig_van at 2:41 PM on January 22, 2010

I love this series. I think we all have very strong opinions about online dating, and ideas about how one should or should not do it. And then they add data. Not perfect data, and not data that's representative of all situations. But data nonetheless, which beats out "my old roommate said..." stories all day long.
posted by Forktine at 2:41 PM on January 22, 2010 [2 favorites]

It's funny how well the male shirtless photos seem to do despite how often women's profiles say they don't like them.

Only for the younger women, whereas cleavage shots get better results with age. Paging Dr Freud to the courtesy phone...
posted by Forktine at 2:43 PM on January 22, 2010 [1 favorite]

Wow, they're going to be setting up blind dates? That is awesome in a "the future really does allow us to be lazier" kind of way.
posted by m0nm0n at 2:44 PM on January 22, 2010

What about having a gun in my profile pic?

This seems to work.
posted by mazola at 2:45 PM on January 22, 2010 [2 favorites]

What about having a gun in my profile pic?

I was sat on a toilet in mine (til I deleted my profile), but I was smiling!
posted by Sova at 2:47 PM on January 22, 2010

The collective wisdom is what people wish was what worked (smiling, wearing a shirt, etc), while people persistently flout this wisdom in favour of doing what actually works.
Mom raises you to be nice and polite to women. She thinks the world would be better if more people treated women like that. And sure, the world is better, but the cost that you're not getting second dates - the bad boys have the appeal. :-)

That's what I love about these OKcupid investigations - it blows the lid off our wishful-thinking, and gives us a glimpse of the real world, no-matter how unexpected or ugly or self-incriminating it might seem.
posted by -harlequin- at 2:56 PM on January 22, 2010 [5 favorites]

I would think this is skewed (for both genders) by the people who are basically looking for some action. Basically, anyone with boobs or six packs showing will get quite a few "Oh, hi, wanna bump uglies?" (the okcupid equivalent of the craigslist dick photo). That gets reflected on the second graph. What it says is that cleavages get lots of messages but not lots of meaningful messages.

The conclusion isn't really "it's better for you to put b00bs in your profile", but more like "if you put b00bs in your profile you'll get a lot more messages, but you'll have to filter a lot". (unless, of course, you're also just looking for some action)
posted by qvantamon at 2:58 PM on January 22, 2010 [2 favorites]

What about having a gun in my profile pic?

That's fine, but it has to go off by the third date.
posted by nave at 3:00 PM on January 22, 2010 [40 favorites]

Women met per 10 attempts for guys with no shirt is approximately 9.

That number would be so much lower on eHarmony.
posted by water bear at 3:02 PM on January 22, 2010 [2 favorites]

Sova: women in the toilet are cutesy. Unless of course there's a turd in sight.
posted by qvantamon at 3:08 PM on January 22, 2010

Only for the younger women, whereas cleavage shots get better results with age.

Better relative results, you mean.

The graph clearly shows new contacts decreasing with the age of the women in question, but if they show cleavage, the number of new contacts doesn't decrease as rapidly as if they show no cleavage.

I'll leave you to draw your own conclusions from this.
posted by UbuRoivas at 3:11 PM on January 22, 2010

I'm still a little stunned by learning that women are most successful when they look directly at the camera (i.e. at you) whereas men are most successful if they look away (i.e. not at you).

You could build a really depressing picture of an entire society just from that information alone.
posted by ErikaB at 3:22 PM on January 22, 2010 [4 favorites]

Am I looking at the camera? Or away? Only I know for sure.
posted by Splunge at 3:27 PM on January 22, 2010

what's a cleavage?
posted by infini at 3:31 PM on January 22, 2010

Sova: women in the toilet are cutesy. Unless of course there's a turd in sight.

I can't believe I'm going to do this.
posted by jabberjaw at 3:39 PM on January 22, 2010 [1 favorite]

what's a cleavage?

It's when a woman is shown in a photo cleaving a big chunk of meat. It shows she is a good cook and can make a mean steak. Men dig that.
posted by qvantamon at 3:39 PM on January 22, 2010 [2 favorites]

what's a cleavage?

$20, same as in town.
posted by UbuRoivas at 3:44 PM on January 22, 2010 [7 favorites]

It's funny how well the male shirtless photos seem to do despite how often women's profiles say they don't like them.


The Situation would not be surprised.
posted by Bookhouse at 3:44 PM on January 22, 2010 [1 favorite]

Is it possible to get a real statistician to help okcupid determine a p-value or a dang confidence interval for once? Because really the difference between 5.91 and 5.92 does not seem that significant to me, and I need to complain about the lack of scientific merit in these studies.
posted by sarahnade at 3:46 PM on January 22, 2010 [3 favorites]

What is he looking at? Slashdot? Or Engadget?

In our favour, taking your photo while reading MeFi gives you an attractive blue sheen.

This was an oddly interesting read. More proof that you can (social) science anything, really.
posted by Dandeson Coates, Sec'y at 3:48 PM on January 22, 2010

What's with the weird graphs where the x-axis meets the y-axis at some arbitrary point instead of 0?
posted by EndsOfInvention at 3:58 PM on January 22, 2010

I think the 'arbitrary point' is the mean.
posted by UbuRoivas at 3:59 PM on January 22, 2010

What about having a gun in my profile pic?

KlangKlangston's got that covered... one of the most hilarious profile pictures I've seen.
posted by crapmatic at 4:04 PM on January 22, 2010

Does the universe implode if I update my profile with a picture of myself sitting on the crapper holding a big pistol?
posted by bukvich at 4:40 PM on January 22, 2010

What's with the weird graphs where the x-axis meets the y-axis at some arbitrary point instead of 0?

They are centered on the average, in order to show whether each factor pulls you above or below the level of success you would otherwise expect, and by how much it pulls you up or down.

Putting the y-axis at zero would be the real case of a meaningless arbitrary point :)
posted by -harlequin- at 4:50 PM on January 22, 2010

At least from the perspective of online-dating, and perhaps social media in general, the MySpace Shot might be the best way for a woman to take a picture.

This isn't that surprising if you understand human psychology. The MySpace shot features one distinctive element, namely the subject's arm that's holding the camera up. Why does this matter? It usually indicates that the person is stuck taking a picture of themselves, presumably because they have no friends and are more often that not desperate for companionship. It's one of the leading "slut indicators" of our times.
posted by dhammond at 5:03 PM on January 22, 2010 [3 favorites]

Is it possible to get a real statistician to help okcupid determine a p-value or a dang confidence interval for once? Because really the difference between 5.91 and 5.92 does not seem that significant to me, and I need to complain about the lack of scientific merit in these studies.

This would be one of those places where p-values don't make sense. They have an absurd amount of data, and there probably is a real small difference (which may not be causal). As the amount of data goes up, any difference no matter how small becomes nominally significant.

I think they they know where to find stats people, but the random errors are not the problem here.
posted by a robot made out of meat at 6:08 PM on January 22, 2010 [2 favorites]

Is it possible to get a real statistician to help okcupid

If you weren't aware, they're Harvard mathematics graduates (among other things). They're not entirely unfamiliar with the territory :-)
posted by -harlequin- at 6:28 PM on January 22, 2010 [1 favorite]

It usually indicates that the person is stuck taking a picture of themselves, presumably because they have no friends and are more often that not desperate for companionship. It's one of the leading "slut indicators" of our times.

If you're a slut, wouldn't that mean you have lots and lots of, er, companions willing to take your photograph? Sitting at home, alone, in front of your computer? You're slutting it wrong!
posted by Never teh Bride at 7:44 PM on January 22, 2010 [2 favorites]

The shirtless males get lots of hits from other shirtless males.
posted by Peach at 7:48 PM on January 22, 2010

Actually, my theory about the MySpace shot is that it creates a zone of apparent privacy between the viewer and the creator, but then I'm pomo like that.
posted by dhartung at 9:50 PM on January 22, 2010 [4 favorites]

nave: "What about having a gun in my profile pic?

That's fine, but it has to go off by the third date."

Wow, that comment was simply stupendous. I'm kind of in awe!!!

What's with the weird graphs where the x-axis meets the y-axis at some arbitrary point instead of 0?

harlequin: They are centered on the average, in order to show whether each factor pulls you above or below the level of success you would otherwise expect, and by how much it pulls you up or down.

Putting the y-axis at zero would be the real case of a meaningless arbitrary point :)

I completely disagree! There already exists the median line in each graph; the reason to have 0 be the "arbitrary" point is because it conveys additional important information, namely the significance of the variation.

Looking even at the first bar chart about "The Effect of a Woman's Facial Attitude", it looks like those no eye contact flirty women are in dire straits- people are running away in terror! But if you put that chart on a y-axis of 0, you'd realize that they're still getting ~7.5 new contacts a month, compared to the best possible facial attitude nets them ~9.3. That relation between the very best and very worst would look something like this:

[][][][][][][][][][][][][][][][][][][]
[][][][][][][][][][][][][][][]

That seems far less tragic a case for our off-screen flirters than the original chart suggests- yes there is a noticeable difference but it's less than 2 extra contacts a month, and it's not like these women should go join a convent what with their dating prospects being so dim. While it doesn't contradict their overall point to use those truncated charts, it does visually overstate the importance of the differences.
posted by hincandenza at 11:09 PM on January 22, 2010

But their stats are just based on quantity, not quality.

Big diff.

Less value.
posted by Vaike at 12:18 AM on January 23, 2010

So according to my calculations, the optimum photo should be taken with a webcam, with me looking away from the camera and not smiling, with my shirt off while doing something interesting with an animal.
posted by Mike1024 at 4:07 AM on January 23, 2010

it might just be as simple as trying to codify what earnestly and seriously demented means
posted by infini at 4:19 AM on January 23, 2010

My theory on the effectiveness of MySpace shots was 1) Photos taken from slightly above the face looking down are more flattering -- they emphasize eyes and cheekbones and 2) Photos taken from this angle look directly down into the cleavage, accentuating it and making the boob objects appear larger.
posted by foxy_hedgehog at 4:37 AM on January 23, 2010

Reading the article would have put the kybosh on that theory at least at okcupid since they found the results remained the same even having filtered out cleavagey myspace-shots.
posted by Iteki at 5:53 AM on January 23, 2010

I'm not terribly surprised at the picture viewpoint results (women get more responses looking directly at camera.. men when looking away)... because it seems like we as a society are conditioned to expect females to be "approachable" (more social leaning) and expect males to be more solitary or independent.

As far as the apparent effectiveness of the Myspace angle, cleavage shots and male shirtless shots...I'm not really surprised at that either,.. but I'd also agree that the type/quality of results you get with those shots will differ from results you'd get from more convention/interesting pictures. (I'd guess most people who use dating sites probably already know this.. .but maybe I'm giving them to much credit)

I like the fact that OKCupid does these occasional data dumps... but I'm usually left wondering how accurately (or not) it represents online dating as a whole. As someone who has used a variety of different dating sites, I find some commonality, but the type of crowd each site draws is usually different (in the same way MeFi's crowd is different from Reddit, and different from Gizmodo.. and different from Myspace). In other words I think its interesting.. but I'm not sure they hit a high enough threshold of conclusions to build any practical new strategies for more effective online dating.
posted by jmnugent at 9:33 AM on January 23, 2010

I've used OKcupid.

Okcupid is full of kind of weird, geeky people. Or it might be that it only matches me with weird geeky, people. Which I guess means that it's kind of awesome.
posted by empath at 9:44 AM on January 23, 2010

with my shirt off while doing something interesting with an animal.

FYI that might be a felony in your state.
posted by qvantamon at 2:12 PM on January 23, 2010

Empath is right. OKcupid users are a very different group from say, Match.com users.
posted by Chrysostom at 6:47 AM on January 25, 2010