The Real Math of Facial Alignment

Mark has famously stated that he believes the odds of aligning 7 facial features on two random people is approximately 1 in 10 million. His math is very simple. He estimates that there is a 1 in 10 chance that one facial feature will align, thus multiplying 1/10 to the 7th power, you get 1 in 10 million.

I decided to put those numbers to the test. I went ahead and downloaded the 2d facial recognition set called “Aberdeen” that is found here.

It contains photos of 90 unique college students from Aberdeen sometime in the late 80’s or early 90’s. What I did was take one photo of a gentleman named Adrian (seen below) and match him up to each of the 58 other white males in the data set.

In planning this experiment I made 7 checkpoints on Adrian’s face where I would compare alignment by aligning the pupils and resizing the eyeballs to match in order to incorporate head size. If alignment was exact or very close, I marked that checkpoint with a “Yes” on a spreadsheet (which I will share upon request). The checkpoints were the eyebrows, the tip of the nose, the subnasale, the crista philtri (top of the lips), the labiale interious (bottom of the lip), the gnathion (chin) and the bottom ear lobe.

After comparing all 58 white males, here are the numbers I got.

None of the photos matched on all 7 checkpoints. One photo had 4 checkpoints, and four of them had 3. Here are the totals below:

4 checkpoints – 1/58 or 1.7%

3 checkpoints – 4/58 or 5.2%

2 checkpoints – 16/58 or 27.6%

1 checkpoint – 28/58 or 48.3%

0 checkpoints – 9/58 or 15.5%

What I learned is that certain alignment checkpoints were much more common than others, which were rare. When pupils are aligned, there is a higher than average chance that the eyebrows will align as well. Since humans have similar head sizes, alignment of the chin was fairly common. What was rare was alignment of the other features. See below:

Eyebrows – 36/58 or 62%

Chin – 21/58 or 36.2%

Bottom Lip – 5/58 or 8.62%

Subnasale – 4/58 or 6.89%

Top Lip – 3/58 or 5.17%

Nose Tip – 2/58 or 3.44%

Bottom of Ears – 2/58 or 3.45%

Multiplying the fractions you get 181,440/2,207,984,167,552 or

1 in 12,169,225 chance that all 7 facial features will align. Mark guessed pretty damn close.

Now of course I understand that 58 can be considered a small sample size, and it is. But I believe 58 random college students (who are of a similar age, race, and ethnic background) is a decent enough sample size to give you a ballpark estimate. Using a confidence interval calculator, we can be 95% sure that the average facial alignment checkpoint mean will fall between 1.08 to 1.54.

In the future I would like to do the same experiment but only between father and son or (non-identical) brothers to see what alignment percentages we should expect. In the meantime, let’s take a look at the number of famous people in the USA and how many we should expect to align on all 7 features.

This article on Deadspin tried to guess the number of famous people in the USA and they landed on 10,000 (with admittedly little research, but I think it is a decent ballpark estimate). Link

Now let’s include all the famous people who have died over the last 50 years. Let’s be liberal and double our number to 20,000. Hell, let’s make it 25,000.

According to the numbers above, there is a 1 in 487 chance that we will find just one matchup between two random celebrities that aligns on all 7 checkpoints. Hopefully this will help you look at our facial alignments in a different light. It will also force us to ask questions about why so many celebrities matchup with 5 or 6 different checkpoints so often, when it is so rare with random people. My suspicion is that 4 or 5+ checkpoint alignments are rare even amongst family members.

13 thoughts on “The Real Math of Facial Alignment”

Straight, you really knocked this one out of the ballpark. That last bit about how many famous people there are was really the last crucial piece of information to see what kind of likelihood we are dealing with here. I curious: about how many ‘attempted’ match-ups do you and Mark go through on average before you find one? I realize you’re not just doing it randomly — you’re making educated guesses and following hunches — but still it would be interesting to get a sense, even a rough estimate.

Like

1. straightfromthedevilsmouth says:

Thanks daddieuhoh, I’m glad you appreciate these types of posts.

Based on intuition, we get alignment on maybe 1 in 25 matches. When I try to bulldoze by picking a news guy (say Wolf Blitzer), estimating what year and age range his previous musician identity would have died at, I narrow it down to about 100. From there I found our guy (which Mark hasn’t confirmed yet) by the letter C, so maybe after 10-15 matches. If I’m wrong, it’s usually because 1 or 2 alignments were very close, but just off.

There are times when I don’t find a match using that method, like Dr. Drew, but I will still find 4 or 5 matches where we’re just a hair of the nose or chin off. So we’re still seeing a shocking number of alignments with 5 or 6 checkpoints.

That’s one of the reasons besides twins we think something very weird is going on in that world, and Mark will have an article on that by Wednesday. I was watching a Youtube video yesterday where a pretty obvious disinfo guy was speaking, and I couldn’t get over how much he looked like Ryan Reynolds. Things like that make me wonder how far and wide this IVF extends. It also better explains their Doppelganger project from last year.

Like

1. I got the same weird vibe after reading Mark’s post about Katy Perry, and then through Google finding at least five other celebs who look like they came out of nearby test tubes: Zooey Deschanel, Siwan Morris, Emily Blunt, Francesca Brown, and to a lesser extent, Mia Kirschner. I suppose makeup and hair style account for some of the resemblances. But not all of them. Maybe there’s some poor guy chained up at the Spook IVF clinic who has developed carpal tunnel syndrome.

Like

1. straightfromthedevilsmouth says:

Those are just the pretty celebs. Who knows how many talentless and ugly lookalikes in that Katy Perry batch have been assigned to theaters, colleges, youtube shows, politics, etc.

Like

Oh, and in case somebody reading doesn’t understand why you Mark initially thought to multiply 1/10 to itself 7 times when looking at 7 features, it’s due to the math of coincidences. Basically, you take the likelihood of each event happening separately (here it is the likelihood of ‘matching’), then multiply them together. So if I flip two coins, what is the likelihood they will both come up heads? Well each coin has a .5 chance of coming up heads. So .5*.5 is .25, meaning that I have a 25% chance of two flipped coins coming up heads, or 1 out of 4 times.

Like

1. I am glad Straight expanded on this. I did something else as well – it would never stand up in court, but I used one-chance in ten on seven different features knowing that the actual alignment probabilities vary widely from that. It is a “rule of thumb” that predicts, usually, the variances will scatter in either direction, so the result would be close. After I read Straight’s analysis, I said “phew!” I was even conservative.

Like

3. By the way, this has come up before … who originally spotted this? John Holmes and Captain Sully match up fairly well. It is unusually difficult to find a straight-on photo of Holmes, so I ended up with a very small one that I had to blow up 450% or so, so it is imprecise. But I look at this and think “Needs further research,” but want to farm it out to anyone who might be interested in finding photos of Holmes where he was not at work.

Like

1. straightfromthedevilsmouth says:

John from Texas brought this one up.

Like

4. hillcountry says:

Greetings,

I filled a bunch of pages of Word docs with photos and interview screen captures for two days and found out how much time this must take you guys, particularly with the variety of permutations out there, from touched up photos and surgery to reversed or blurred pictures. And all of that on top of doubles, twins, zombies and maybe triples, or twins and doubles for each etc.

Here’s a couple of possibles.

Gram Parsons – Pierce Brosnan. Brosnan is easily ten years older than his Wiki dates. It was the eyes and the left nostril. The same thing happened near the end of that zombie search as the one I did today, near the end of my stamina I was suspecting Brosnan is twinned. Maybe it’s an aspect of mental fatigue, but there are some details that one can’t explain away very easily. Anyway, it’s an interesting feature of the process, getting side-tracked.

The other possible search started from hearing a guy on a pod cast mention that Cyndi Lauper looked like Jill Stein. Not sure about the Lauper twin after a couple of hours doing side-by-sides, but I’m convincing myself that there’s two Jill’s on the campaign trail. Neck mole, chest freckles, scarfs, necklace, other details. There’s a strange painting of Jill way down on the Google Image page that has a distinctive line through it and a skin shade difference on either side, representing a facial analysis. She’s got her arms folded in a masonic fashion and the words to the upper left of her head are “It’s In Our Hands”. Lower right says “2016 – Jill”. Bored cube guy?

It would make sense that if they’re going to all the trouble of running the Green Party, they’d probably want to have a back-up. She’s memorized her script pretty well from the video interviews I watched. I was trying to figure out if there’s one Jill for TV and one for the streets.

That Louis F – Sam C article and match is great. An interesting YouTube out there is one where Alex Jones is interviewing Louis F right inside Louis’ home, across a dining table. A zombie interviewing a zombie? Farrakhan pretty much dominates, so maybe Hicks was a Cooke fan.

Found another one where Farrakhan is lecturing a Navaho tribal council on their turf. I’ve got an old paperback with him going up into a UFO mother-ship. I think it was published by the NOI. I remember those cats from college days in the early 70’s when they wouldn’t even sell you one of the newspapers they were hawking on campus. It’s been one very long mind-f*** on all of us, for sure.

Like

1. da bear says:

Mark,

Here is a possibility for Pierce Brosnan that I just thought of, although I have yet to do a facial match — Jim Morrison. Similar faces and smirks. Plus not just anybody gets to play James Bond, I would assume.

Like

2. Thanks guys for the suggestions. I’ll add your suggestions to the list. You’re right hillcountry about the time and fatigue factor. It is taxing.

Like