A data skeptic episode titled "Crypto" surely this will be about cryptography, well, no actually. This is the 64th episode of the show, that's two rays to the eighth. Over the last year and a few months, I've had the pleasure of sharing many episodes related to data, to science, machine learning, statistics and so forth, and I've had a few people ask me, why is this the data skeptic podcast? Not just the data science podcast. I explained a bit of my intentions in my first episode, but this capstone is a fine place to revisit. Data science is a broad umbrella of tools, algorithms and techniques used to extract knowledge from data sets that are often large and sometimes unstructured. Scientific skepticism is the practice of asking whether claims are supported by data, an empirical evidence, and are reproducible. A large part of being a skeptic is not being afraid to entertain impossible ideas. This is why organizations like the James Randy Educational Foundation have challenges, in their case the million dollar challenge, to be awarded to anyone that can demonstrate supernatural or paranormal abilities under testable conditions. Will they ever find something? They haven't yet. It would be cool if they did, but if they don't, maybe there's things to be learned along the way. Aside from these two areas being things I'm passionate about, they are also sets of ideas that I want to bring together. I've seen more than a few cases of people claiming they've demonstrated something and that their claim is intrinsically correct because they used machine learning or a statistical model. Not true. This appeal to machine learning fallacy is dangerous. Data scientists have a responsibility to be stewards of our skill sets and help people make smarter, better informed decisions. The tagline for this show, which by the way will be retired as of this episode, has been conversations at the intersection of data science, skepticism and empirical validation. Half my shows are shorter mini episodes where I try and convey a very, very high level understanding of a data science technique to my non-data scientist wife Linda. These are intrinsically data science alone shows I suppose, but for my interviews I try to keep things at that intersection of these two topics. The pendulum swings back and forth. As of claims made by the global consciousness project as discussed in episode 29 are certainly more skeptical, albeit with a data component. My recent conversation with Andrew's drop in on video game analytics certainly leans towards the opposite polarity. This episode will swing strongly to the skeptical side for today, but it won't be without some statistical commentary and a bit of exploratory data analysis. I think it's entirely within the data skeptic domain to discuss the probability of improbable things. I'm fascinated by outliers and what they teach us about our data sets. With this in mind, this episode, crypto, won't be about cryptography. We'll be discussing crypto zoology. The study are reports of scientifically unconfirmed species. In particular, this episode is about Bigfoot or what I'd like to call the Bigfoot hypothesis. Let's talk about what this episode will and won't be. It will not be a complete review of the history of Sasquatch. It will not be an examination of the rich folklore. It will not be a complete review of evidentiary claims. This will not be a diatribe on skeptical investigation methods. I want to tell a story about evidence, about data, and how tightly intertwined beliefs and statistics are, or at least should be. After all, I believe firmly that the US dollar is a relatively stable currency. But I don't believe that as much as I believe the sun will rise tomorrow, and the next day, and the next. So today, I ask, what is the probability, however small, that Bigfoot might be real? You know what I'm working on, right, this audio document or anything? Yeah. You know, it's about, right? Big fun. But what is your honest thoughts on that? I don't know any thoughts. I don't know anything about Bigfoot. I mean, do you think I'm wasting my time? Do you think it's silly? Mom, I think it's cute. I think it's gonna be a really good show. Oh, thank you. That's what I was expecting me to say. Why did you think I would say? Well, I don't know. I thought maybe you'd be like, "Check your ass." No, no, no. Like, your audience doesn't necessarily, or data science people, they don't care about Bigfoot. And the skeptics people already know there's probably not Bigfoot, so maybe you might be like, "No one cares." No, it's about the journey. That's right. Any good story is about the journey. I'm glad you saw that. Thank you. A lot of people would tell you, there's no value in these things at all. They're just completely trivial, fringe, baloney. And that's Daniel Loxden. I write and illustrate books for kids. Mom and little science for grownups. Junior Skeptic Bound into the Back Skeptic magazine. When I talk to Daniel, he was no stranger for these ideas. That you and I should probably not be talking about it. We should go ahead and get a job and get our hair cut. So why are questions like this worth asking? These things have always seemed terrifically interesting to me, and that would be my first argument. Paranormal beliefs in general are almost universal across our species. Cryptozoological beliefs are very, very common, even in the educated west. But mostly I just think they're fascinating parts of the human story. Huh. Bigfoot. Okay. I knew when I was going to take on this project, I was going to need help and some advice from friends and former guests of the show. Many listeners will recognize that as the voice of Noelle Cio Saldana from Episode 53. In addition to having a math and data background, Noelle also has a degree in physical anthropology. Do. Yeah. So I asked for her open-minded take on this. So it's like probability. Is this even plausible? Complete like hypothesizing. Like there's no reason that we don't can't have like some enormous bipedal thing that just offshoot somewhere. And just because that the evidence doesn't exist doesn't mean it's not necessarily there somewhere. Okay. So I'm hearing not impossible. That's a good start. Maybe this thing actually did exist. Welcome back to the show from Episode three, Nathan Janos, the very first guest of the show. Call it Bigfoot Sasquatch Yeti. What are we talking about here? Historically, I think that Yetis might have been actually other species of hominids. I think we're finding out that the hominid branches are much more diverse and we maybe thought and maybe there's co-occurring branches. We only scratched the surface for some things like, you know, archaeologically. Who knows? What happened? We find things all the time about species that have been preserved since the Jurassic era that haven't really evolved much and haven't really changed much. So who knows? One of the things that is consistent about our knowledge of life forms on earth is that we're constantly being surprised. I would say that our planet would support a Yeti. Okay. This is not where I thought we'd be at this point. You didn't mention exactly what your definition of the... It depends on, first of all, how you wanted to find... You know, all the different names we have for this character all around the world. Abominable snowman or Sasquatch Yeti. The American version. Like if we want to say the probability of Bigfoot existing, considering he's bipedal, do we know that Bigfoot is related to the apes? I feel like this definition of stuff is going to kill me. So for sure Bigfoot would have to be a primate, probably of the hominid family. We'll talk more about speciation later. But for now let's just say it's a large modern day hairy bipedal creature with really big feet. So the idea of a large bipedal ape living in the US is just really absurd. Oh, I forgot about geography. And welcome back to the show episode 35 guest Sharon Hill. However, it's not impossible. Okay, but I'm going to need some sort of reference point here. But I think it's more likely than us being visited by space aliens. Yeah, in most instances on most planets where there's two-legged things walking around and somehow some level of hair, that it's going to be very likely that there's a Yeti on most planets like ours. Okay, that's not what I meant by revisiting geography. But it's interesting that we're talking about aliens here, because I reached out to some people in the skeptical community to get some perspective on this topic. Hello, this is Richard Saunders from the Skeptics Own Podcast. Bigfoot. I think it's more likely that Bigfoot exists than extraterrestrials visiting the earth. However, I think that the prospect of finding extraterrestrials or their being extraterrestrials elsewhere in the galaxy of the universe is more likely than Bigfoot existing. I agree with that. Upper bound. What do you think is more likely? Extra-terrestrials or robot uprising? Great question. I don't know. My actual answer was like, I think Bigfoot's maybe less likely than winning the lottery? I don't know about winning the lottery. I mean, it depends on which lottery. So less likely than getting killed by a champagne cork type of the thing. This is kind of fun. Let's do one more. This is Susan Gerbick from the G-S-O-W Project. Listeners will recall Susan from her appearance in episode 13. I grew up believing in Sasquatch because my father took me to see the movie when I was a teenager. And why would Hollywood lie? In 1976, we had only the public library in the Encyclopedia Britannica to get our information from. This is why it's so important to have great information on the internet. And Wikipedia as a source of all knowledge is where to start. Cryptozoology is still drawing crowds. In April 2015, the Bigfoot Wikipedia page had 59,572 viewers. Kyle asked me to rate Bigfoot on a plausibility scale. And I would say that Bigfoot is more possible than spontaneous human combustion. Stop, drop, roll. But less likely than facilitated communication. Both Wikipedia pages were rewritten by G-S-O-W. Okay. I know I said less one, but just one more. I would say that I find Bigfoot to be less likely than, or less apt to exist than finding some new monkey that we've never found before. Not in the US, though. I really don't think we're going to find any big mammals left in the US. Not so common, yeah. That's true. Just to make sure I have the pronunciation corrected. Leven de Vries. De Vries. De Vries. Leven. Leven. I'm here with Leven de Vries. Hopefully I got that pretty close, Leven. Yeah, it's fine. Leven is a super cool guy. I'm really lucky he agreed to come on the show. For my massive thesis, I wanted to study primates in the wild. So I went to the Central African Republic to study the largest known group of Agile manganese. I tried to figure out why they live in such a big group. I first learned about Leven during his second expedition when he launched the elusive red colobus monkey expedition on Indiegogo. Because actually there's virtually nothing known about his species of red colobus monkey. Since the 1980s, people have mentioned that there should be someone doing a survey to look for them because the last sightings are from the 1970s. But these sightings were not really confirmed. There was no proof and it was done by non-primitologists. So we don't know for sure if it was the same species. This is a really interesting point about how we weigh new evidence. If a primatologist, someone who's qualified and trained comes along and presents evidence, that carries a lot more credibility than stories that the villagers tell. But we can't completely discount these reports either. Perhaps they're mistaken and the species has gone extinct. But we can't completely trust that these sightings weren't over the red colobus monkey either. But probably it was. And so there was really nothing known about these species. It was even not known if they still existed. And so that's something I wanted to find out with this expedition. I want to make something very clear. I'm not trying to make a false dichotomy. I didn't ask Leven to come and be interviewed because I want to say if he can find this monkey then someone else can find Bigfoot. We don't have much a priori reason to believe in Bigfoot. But in the case of the red colobus monkey. So it is a valid species that has been described in 1887. Around that time there have been two specimens that have been collected that are kept in the natural history museum in Paris. And there are a couple of more specimens in total. There should be eight all over the world. One in the states as well. And so from these specimens I tried to figure out where they came from. That's not always easy because the village names of about 100 years ago don't always correspond to the village names that are used today. But from these specimens I was able to pinpoint basically only two locations where this species was collected. And there is also one book from 1949 where two French people spent a couple of years in the former French colonies. And so they also provide information on the occurrence of different primate species and also of this red colobus monkey. So Leven had pretty good ideas where to look. That's a good start. So let's learn a bit more about this species. These red colobus weigh a bit more than 10 kilos, I think 12 or 14 kilos. They are among the largest monkeys in central Africa. So they are a bit smaller than the boons or mandrels if you know these semi-terrestrial monkeys. They are very curious. So they are very visual, they make a lot of noise, they jump through the trees. So for hunters they are very easy to shoot and that's what makes them so vulnerable. It's a critically endangered species. The main threats to this species is hunting. You have to make a distinction between commercial hunting and local hunting. Really I think that's an important distinction. For us it may seem like a very weird thing that people hunt monkeys and eat monkeys. But for these people that live there, it's part of their culture for several centuries or since they have arrived in the Congo basin, they have been hunting monkeys. That's just a ray of living and they really depend on the forest for animal protein intake. So for me that's not a problem, I'm not against the local scale hunting. But the main problem now is the commercial hunting that poses severe threats to a lot of monkeys and other wildlife in the Congo basin. Several people wrote before that this species might actually be extinct. But I didn't know really what to expect from the satellite images. I knew there was still quite a lot of forest left. Actually there is no forest expectation because it's very swampy and inaccessible area. So it's difficult for commercial exploitation but there is of course also a lot of hunting and it's a problem all over Central Africa. So sorry to break in here, I just had to make one quick point, I couldn't resist. We didn't get into it but I suspect leaving was sort of manually looking through those satellite images. So we have a very inaccessible area and a potential problem of hunting and lots of satellite images, lots probably. That sounds an awful lot like a good data science problem. So a hint hint, wink wink, nudge nudge. The traditional hunting used to be low scale but now people are doing commercial hunting. So it's to gain money and that means that they just shoot whatever they can get. Since these monkeys occur along several major rivers in the Congo basin close to the big cities. My fear was that they could have been hunted out but I didn't know what to expect. I thought if the local people know them then there is a good chance that we will find them. So it was in the Republic of Congo, that is Congo, Brazil, not the bigger Congo, not the Democratic Republic of Congo. So the Republic of Congo was a former French colony, it's on the western side of Central Africa, it's on the equator. So all the transport in the area is on the rivers. There are basically no roads or no paths through the forest because it's all inundated forests or at least seasonally inundated. So during some parts of the year the river rises and the forest is just flooded everywhere. We started in a major town and then we went to the first village and from the start of the expedition. The people in this first village they said they know about this monkey. It's the only red monkey in the area and it's a fairly big monkey. So they hunt different animals, they hunt monkeys as well so they know the monkeys very well. From the beginning people in this first village they knew about these monkeys. They said they were not present on that side of the river where we were but they said we had to go to the other side to a village called Ntoku. And that's also a village that was mentioned in the book from '49. From the beginning it was quite clear that we were on the right track and now we will start writing up a publication. And this information will give a better overview of the conservation status and the occurrence of this species in the Republic of Congo. But the positive thing is that we found them in a national park that was created only two years ago. So it's a new protected area. It still has to be further developed because at the moment it's mainly a paper park, what they call it. So a park that is just protected on paper in the law but in reality there is still a lot of work. But at least there is potential to have a protected area where this red colubus species occurs. And my hope is that our study and our future publication will draw attention to this national park. And it would be great if an international organization would get involved in a project there because I think it's a national park that really deserves our attention. And that deserves a more complete protection. I neglected to ask leave and about his companion or the other people that helped him on the expedition. So a quick shout out to Gail Gobolo, hope I'm saying that correct, who, if my research is correct, accompanied him. I'm sure there was a lot of support back home and locally as well. If you'd like to support the conservation efforts, I believe the best place is to go to the Wildlife Conservation Society at WCS.org. So what have we learned? It's possible for a species to be plausibly extinct but later confirmed to still exist. That's got to be important to the Bigfoot hypothesis somehow. But what cryptozoologists often do is they take credit for these findings in other countries and rather remote areas. I don't know, I've been thinking about this. What is the reasonable expectation of burden of proof? You know, to go into the Pacific Northwest and find a Bigfoot or evidence of one. I mean, it's not the same thing, but let's compare it to what Leavin did. In an area that's pretty inaccessible, he had whatever he paid out of his own pocket plus a whole $2,567 from his Indiegogo campaign. How long did it take him to find the red colobus monkey? We had a bit of delay because of the administrative problems or challenges in the Congo basin. Alright, here come the excuses. I bet he was done. There were a couple years working around the clock. But then as soon as we headed off in the forest, then it took us five days to find these monkeys. Oh, so if a relatively low-funded expedition can find a species in the Congo basin in five days, why can't cryptozoologists find Bigfoot? This is a case of hypothesis testing. The null hypothesis is that there is no Bigfoot. The alternative hypothesis is that Bigfoot exists and is very elusive. We'll talk more about that later, but for the moment, let's talk about updating our beliefs given available evidence. We don't eliminate the alternative. The absence of evidence is not evidence of absence. However, the absence of evidence in situations where it's reasonable to expect evidence increases our posterior belief in the null hypothesis. As I was studying the Bigfoot topic, a lot of people point to the coelacanth. The coelacanth is a fish that was long thought to be extinct. It evolved to its current form about 400 million years ago, and it's endangered and threatened today. But for a while, people assumed it had gone extinct. So if a species that used to exist was thought extinct and was found again can happen, could the same happen for Bigfoot? Is this a fair comparison? They will say, "Oh, look, there's this new species out there, so maybe Bigfoot is out there." It really does not make a connection to me. The coelacanth was living in an area where we really didn't have a lot of biologists around, and they weren't examining the fish that was being brought up. The fish wasn't a very popular one to eat. It wasn't very common. It was relatively rare. They said that there weren't any fossils for 60 million years. Well, there actually were a couple fossils from the coelacanth that would have suggested that it continued to live on past the Cretaceous extinction, but we didn't even really know about those fossils until relatively recently. There was a couple of things that had to fall into play post-discovery of the coelacanth. So I don't like when the cryptozoologists used those, "Oh, look, we found this new monkey. The olegrito is this little animal that they found in South America. Look, we found this fantastic new animal." Well, that's because it looks an awful lot like it's very similar relative, and we just didn't notice that it was different in its DNA and genetics. So there are those that will say, "Oh, science got it wrong. The coelacanth wasn't actually extinct." And I don't think that's really a fair representation of what the scientific community would say about the coelacanth. All evidence we had available led us to believe it was probably extinct. We got new evidence, and our beliefs were updated. The same would happen if we got credible evidence of Bigfoot. So I'd like to put to bed the argument of, "But what about them? What about the coelacanth?" Just because something could be true doesn't have an impact on whether or not it is. Why does this argument persist? Let's get one theory from former guest Alex Bocklin, who appeared in episode 27. Yeah, I've been thinking about this a little bit since you mentioned this topic to me. And I think that in many ways, just like the secret to any other fringe stuff like the UFO, the feeling of mystery and the feeling of some hidden information. The information that is hidden from the general audience by the government, by the scientists, whatever. It's very appealing. Be that Bigfoot or aliens or the Loch Ness beast, things like that. So in this sense, I think that Bigfoot is not an exception. But it's an interesting case in a sense that it's connected to paleontology. It's not something radical like flying saucers, for example. It's more subtle and it's very interesting. Yeah, I agree. That's actually why I picked this topic. There's something intuitively plausible about Bigfoot. There's no reason why such a creature couldn't exist biologically, at least not to the best of my knowledge. Thankfully, it's connected to actual science. We have good protocols for how to evaluate the Bigfoot hypothesis. But what should we be looking for? What would be convincing evidence? Some people are doing some really interesting work. They're working on DNA collection or they're working on putting out gamecams to catch a trail, photographs of whatever comes by. And I think they're doing it the right way. Recently there was that show, the million dollar Bigfoot bounty that came out where it showed that real scientists were out there helping these people and showing them how to gather good DNA evidence and how to take good samples, environmental samples. I thought that was fantastic and I think that they really learned a lot from that. So from what I can tell, there's nothing harmful about investigating Bigfoot. On what's the harm.net, there's no Bigfoot or cryptozoology section. And I imagine that's because these beliefs don't directly damage anyone. Aside from maybe, you know, mosquito bites or other unpleasant trees related to stomping around in the forest or something. There's still a right and a wrong way to look into these things. There can be no special exception here. We should expect the same rigor and standard of evidence for claims about Bigfoot that we do for anything else. So what types of evidence are worth pursuing? Unfortunately, a lot of what I found in my research amounted to anecdotes. And as someone where he wise once said the plural of anecdote isn't data, it's anecdotes. We have been amassing Bigfoot footprints since the 50s. You know, they're still no Bigfoot. Those footprints, they have not carried us any further. You know, the same is true with vocalizations. Many Bigfooters believe they have an understanding of what Bigfoot sounds like, but is never being confirmed. And you, again, have this situation where you have people who have the hobby of going out in the woods and listening for Bigfoot. And you also have people who have the hobby of going out in the woods and sending out recordings of Bigfoot. Really loud over loud speakers. And they literally could be in a situation where Bigfooters are talking to each other in Bigfoot ease. Without ever having studied an actual Bigfoot to hear what they sound like. Yeah, for that reason we're going to have to cross audio off the list. What about hair samples, nests, footprint casts, photos? My answer to that I think would not be very encouraging. My sense is that all those forms of evidence, with the possible exception of DNA, which is an interesting area, are played out. We managed to get the first photograph of a living specimen. I also got fecal samples. With these fecal samples, we can look at DNA, because in the feces we have little cells from the intestines. So we can get DNA out of that. And with that we hope to answer the question about its position in the followed genetic tree to see which other species of red clovers it is most closely related. Okay, the fecal samples actually sound promising. I think Bigfooters should look for those actually. Maybe more than they look for the footprints that they later cast. They've talked believers into an understanding of the Sasquatch in the absence of any kind of type specimen to check it by. So they've conjured up a kind of fantastical beast of some sort, and that conjured version then informs further footprint discoveries of the manufacturer of footprints that you're hoaxing. But we have no way to lock it down. It's just a kind of folkloric understanding that emerges from the footprint evidence in the end. So let's get into footprint evidence. There was a researcher by the name of Dr. Grover Kranz. He was a professor of anthropology at Washington State University. Grover was a Bigfoot believer in his book Bigfoot Sasquatch Evidence, the first line of evidence he goes to is footprints, or rather casts of supposed impressions made by Bigfoot. Further, Grover claims to have been able to identify what he called "dermal ridges" the same characteristics as our fingerprints in several casts. I was unconvinced of this. Grover also said he knew of two properties that were definite telltale signs of true footprints that he could use to filter out the fix. As far as I've been able to determine, Grover never told anybody what those two features were. Despite these secret metrics, it has been documented that Grover endorsed a few casts that were later confirmed to be hoaxes. Yet overall, there's a great deal of Dr. Kranz's work that is quite good. He frequently applies aspects of science in a way I appreciate. He's not alone in this. I should also mention Dr. Jeff Meldrum. In addition to being a full professor in the Department of Biological Sciences at Idaho State University, he wrote a book called Sasquatch Legend Meats Science. It's an absolute tour de force of all the supposed Bigfoot evidence. If you want to find the best available presentation of pro Bigfoot material, this is a good starting point. I may not agree with everything in the book, but I respect the approach and methods Meldrum takes, in most cases. Meldrum's book also led me to the work of Dr. Henner Farrenbach. Farrenbach did an analysis of a legit Bigfoot impression cast and wrote up his findings in a paper titled Sasquatch, Size, Scaling and Statistics. Naturally, this piqued my interest. Farrenbach claims that in his data set, when he looked at certain features like the length, with, heal with, foot with index, step length, and things like that, that the measurements conform to standard statistical distributions, exactly as would be expected from a real population. In other words, that because the data appears to be distributed like, let's say, the Gaussian distribution, the data are trustworthy. Indeed, if the footprints are real, I would expect the types of distributions an analysis Henner showed. Listeners might remember episode 52 z-scores, in which I talked about a very related idea and actually chose human height for my example. However, Henner really further suggests that if these casts were the products of hoaxes, that they would not have had these statistical distributions, that hoaxers and misidentifications could not possibly have these statistical properties. I have quite a lot to say about this whole paper. Some of the things that I want to say can't possibly fit in the audio format. So I'm going to put out a video a little later on with a deeper response to this specific paper. This is going to take me a little time to do well. Expect it out later this year. But for now, my response is this. The claim that hoaxers would not produce believable sizes and shapes I find unproven. No evidence whatsoever was provided for this lemma. As such, I find these statistics to be equally consistent with the Bigfoot hypothesis and the null hypothesis. The data are unconvincing for updating my beliefs. I also wonder why we're looking at footprint casts in general. If someone is on trial for a robbery and you have a complete clear indisputable video evidence with a crime, just show the video. You don't need to look at the transcript blog for their metro car to establish that yes, they might have been at the crime scene at the right time. I mean, unless you're Angela Lansbury showing your suspect the flimsiest evidence to try and scare or trick them into confessing, go directly to the best evidence. Do not pass, go. Do not collect $200. Show me your one most convincing support so that we can agree. By definition, if the first best piece of evidence isn't good enough, neither will be the second best piece of evidence. No gish gallops, please. So I don't know what the best piece of evidence is. Maybe we can find out. Speaking of my video analogy though, isn't there a famous Bigfoot video? Patterson Gimlin film. I'm wondering what's your take on it and especially your thoughts on if that footage had never been shot. Do you think Bigfoot would still be capturing people's imaginations? It's a really interesting question and I'm not sure I know the answer to it. The Bigfoot world would be different. I know that. For many Bigfoot proponents, that film is really the kind of the rock bottom thing. It's the thing they can't get out of their mind, the thing that keeps them slogging on year after year and decade after decade. By coincidence, I just recorded a quote from Michael McCloud's book quoting Renee DeHinden earlier today. He got up at a convention at one point and he said, "All our doubts, we argue, we doubt and really get blue about it and say, "Jesus Christ, what the hell are we doing? We're wasting our time." But we always come back to this damned film and say, "What about the film?" And I think that really goes to the heart of the Bigfoot subculture. It really has just an unparalleled role. Yeah, that's incredibly true, as I look through those who really believe in Bigfoot, almost everyone hinges a large amount of that on the belief that the Patterson Gimlin film shows an actual Bigfoot and not a man named Bob Hieronymus in a monkey suit. We're going to come back and talk about the Patterson Gimlin film a little bit later, but there's got to be something better than a 50 year old video. I mean, what about DNA? Is anyone doing DNA test on it? Right, Brian Sykes is a researcher out of the UK, a geneticist or genetic expert. He did a special through British television station where they did this three or four-part series where he looked into claims of the Yeti in Nepal and Tibet, the Amasti in Russia, and then he came over and talked to the Bigfooters here in the Pacific Northwest. He looked at their data and he took back some of their data samples and he did this big study out of his lab, which was cutting edge genetic technology to identify DNA strands. And he was able to find that nothing interesting came out of the United States. Everything was bare, wolf, dog, human, whatever. Unfortunately, even though he looked at what they were called the best samples, they had nothing came out in Russia. They did not find the Amasti. What they did find was a interesting idea about a woman who was said to be an Amasti long, long ago, when they took her DNA samples from her remains. It turned out that she was from a completely different part of the world that they weren't expecting. She must have been imported possibly as a slave. So I thought that was really interesting. I think Moore is going to be coming out about that eventually. But he also found that some of the samples in Tibet and Nepal, I'm not sure which ones exactly where he found them, turned out to be a strange species of bear. What looked like a hybrid bear between the local bear and a polar bear. The polar bear was clear out of Svalbard in Norway, which was very far away. So they were trying to make the connection of how this descendant of this ancient polar bear got his DNA down into the Himalayas. And he thinks that he may have discovered a new species that may account for some of the Yeti sightings, which I thought was extremely interesting. Yeah, that's actually really cool that, okay, maybe they didn't find the Bigfoot evidence, but maybe they uncovered a new species, or if nothing else, certainly an interesting genetic link. So it seems that some of academia and science are actually genuinely looking into this question. Yeah, but I mean, the whole idea that, you know, sort of academia just, you know, wants to freeze out new ideas and stuff like that is just it. Well, hold on. Hold on. Carl, I actually forgot to introduce you earlier. That's Carl Mamer, host of the conspiracy skeptic podcast on which I was a guest and also guest of this show on episode 23 data myths. If you are a member in good standing of your appropriate scientific society, you can turn out to conferences, you can put up a poster. If you've got something, you can get it in front of people's eyes. People go to these conferences to kind of have discussions and debate and stuff like that. It's a lot more open than the laymen thinks it is. There's this whole idea. It's like science is about having, you know, we were right all along parties or something. This comes up a lot in Bigfoot and other sort of conspiracy oriented things. People say the academic community is trying to shut them out or there's a cover up to silence them. I just don't see this. If you had evidence for a species like this existing in the modern day, like credible good evidence, the academic community would embrace that. They'd want to see that. Now, if you have a clugey half piece of evidence, you're going to hit a lot of resistance. That's just part of the process. So has there really been no interesting DNA evidence found? Well, a couple of years ago, a Texas veterinarian by the name of Melba Ketchum put forward the claim that she'd found DNA evidence that established the existence of a North American Bigfoot. And she and her team were working through the traditional academic process to get that published. I spent a lot of time investigating her history and what she claimed and followed it for the two years that she was saying that more exciting news is coming. I tell you what, I was hopeful and I was on her side. I was hoping that she had something really interesting. I want her to have a chance. She's got some credentials. You know, she was a veterinarian. I don't think she's crazy. I want to hear what she has to say. Maybe she found something different. And then it all came out and it was just worse than I imagined that her study, she had trouble getting it published. The people who reviewed it at nature were extremely tolerant of her, in my opinion. Academic conspiracy, you know, the ivory tower. Academics don't want to look at my evidence. They reject this stuff out of hand. So peer review is the correct process to get the community to validate something. Getting rejected from a journal is also sort of part of that process, I suppose. That doesn't invalidate. Maybe she has the evidence and it just needs to be written up better. But more and more really odd and bizarre stuff started coming out. She was working with the Russians. The Russians have crazy ideas about the Yeti. They're trying to promote it as tourism. She started talking about horses mains being braided by Sasquatch, where anybody who has a horse out in a pasture has seen their horses mane and tail come out looking like a ball of, you know, just knots. It just happens naturally. There's no big foot doing it. It's just absurd. The ideas of stick structures being made by Sasquatch is that she communicated with Sasquatch. I mean, then we started getting into things like Sasquatch were the nephilim of the Bible, the giants of the Bible. And I was like, I'm actually going to cut you off right there, Sharon, because we're almost in the point of ridiculousness now. I mean, in reality, even the most hardened of Bigfoot believers, they think she's nuts. So I include a mention of Ketchum, not as a means to ridicule the Bigfoot research community, but actually to make a point about evidence. Bad evidence shouldn't reduce our belief in something. It's just irrelevant. I'm an advocate of Bayesian methods of entertaining how we update our beliefs given new evidence. That's really the method every good data scientist uses. Long-time listeners of the show might remember way back to episode six in which I discussed Bayesian updating with my wife in the context of a convoluted example involving lemons and pomegranates. Part I didn't really cover very much in that episode is the importance of what we call the prior probability that every model, model being, you know, hypotheses in this case, Bigfoot or not Bigfoot, must have some a priori likelihood. Well, to start with, we need to talk about models and what they really are and what they mean. Yeah, yeah, exactly. So I don't know. Is there a theory in data science about that? Well, I don't know that it's exclusively a data science theory that we can take credit for, but model is essentially synonymous with hypothesis. And in some sense, any complete description of an idea is a reasonable model. Many kind of paranormal beliefs, homeopathy or telepathy or precognition or these kinds of things, in order for them to be true, the universe would have to be different in a really fundamental way than science currently understands. Bigfoot is interesting and most cryptids because they're 100% plausible or very close to it on the face of it. You know, there have been creatures very much like Bigfoot, the two people on our planet, you and I are creatures very much like Bigfoot. There are billions of us. Bigfoot is not a weird idea. So this is a good time to finally touch on Occam's razor. I think everyone's probably heard of that, or at least some version of it. In the colloquial sense, it's the idea that the simplest explanation is probably the most correct one. And this is a heuristic in science. It's not a rule, it's not a law, it doesn't derive from first principles or anything like that. It's just an observation many people have made about reality. By and large, the simpler solutions tend to best describe the actual fundamental phenomenon. So yeah, if someone tells me, hey, you know, I have a friend or relative and they took a magic water treatment and it cured them of some ailment, that's not credible evidence of homeopathy. And this is an important and sometimes subtly overlooked point. There's a tendency to maybe give equal footing to all hypotheses. So for example, either the moon landing happened or it didn't. There are people that actually deny we landed on the moon in 1969. That doesn't make it 50/50 at all. One thing I found with Bigfooters is I'm not sure they understand the concept of the way skeptics approach things. Well, at least the way I approach things. I don't discount somebody's experience. They had an experience. I wasn't there. I can't say that their perception of it or their description of it is right or wrong, but it could be a ton of different explanations. It doesn't have to be either you're a liar or you saw Bigfoot. That's a false dichotomy. It could be a whole host of other things. You could have seen a person that was dressed up in camouflage. You could have seen a bear that's walking upright, which happens quite frequently. There could be so many things that could explain their experience, including misperceptions. I really don't want to discount people's experiences because they're emotional to them, but I am not going to conclude that that was Bigfoot sighting. It could be when everything else hasn't been excluded. This is an excellent point, Sharon. This is what I like to refer to as the principle of indifference. It's generally attributed to the Greek philosopher Epicurus. It puts bluntly, it's the idea that one should keep all hypotheses that are consistent with the facts. In other words, don't rule things out unless you can absolutely contradict them. This really has to be balanced and balanced well with Occam's razor. I mean, simpler solutions are probably more true of priority, but you need to keep around all the other hypotheses because new evidence might promote them. Maybe more complex ideas will demonstrate themselves to be true later on. But I would argue complex models are preferable to a simple model that invokes an unknown entity. So what is more likely to explain to UFO reports, a very complex model where it's a combination of hoaxes and misidentifications and people being drunk or UFO proponents will go. But no Occam's razor is much more simple explanation. It could just be space alien. But it's a very simple model. Space aliens. But that simple model requires the addition of an unknown entity. It seems to me that sometimes complex models are preferable to simple models that do require the invention of an unknown entity. Yeah, this is a good point as well. I guess my take on it is that these are incomplete models. If one says, "Oh, space aliens is the answer," that also implicitly presupposes the existence of those space aliens that they have the ability to travel vast distances. It's presumably greater than the speed of light that they've arrived here at Earth and been undetected and so on and so forth, whereas the Chinese lantern hypothesis doesn't require so many assumptions. So it's the complete description of a hypothesis that we really need to be after. The Bigfoot hypothesis requires it be bipedal for one thing. It also requires that whatever the creature's diet, it's able to find and consume it without an ecologist being able to notice the consequences of that in the food chain or vegetation of the area. Wow, there's a new feature we have to add to Bigfoot. You've got this creature that peaks and cabins and lifts up trailer homes and eats bagels left out for it and all sorts of things. And yet there's not one single master hunter out there that has ever been able to shoot one or capture one or get dung or something like that. Now there's an important part of actually leaving out here because it's sort of out of scope for this, but I'd ask some listeners to look at Commodore of Complexity as a possible way of evaluating how we find the a priori probability of complex models. Minimum description length is a proxy for that as well. So that's potentially the formalization of Occam's razor. We need to keep in mind what Sharon told us about the principle of indifference. Let's keep around all potential hypotheses appropriately weight. And let's keep basis rule around as well and update our probabilities given new evidence. On the other hand, there are great many enthusiastic believers for these creatures. And they are by and large stuck in a very uncomfortable position where they are proposing creatures which should be fairly easy to confirm, but which for decade after decade, never are confirmed. And so they've had to kind of insulate themselves from that fact with a series of arguments for why we should expect to see no evidence. And if you're in, you know, once you've argued yourself into that position where evidence is just not expected, then, you know, what evidence could change your mind. You know, Bigfoot has to apparently run at 40 miles per hour to sort of explain different kinds of sightings that when you talk to sort of Bigfoot people and you talk about, you know, the fossil evidence, woodsmen can't find Bigfoots or, you know, why we can't, you know, find bodies and all those sorts of things that add a new trait to Bigfoot. You know, Bigfoot has really good hearing and knows when you know hunters are around and they hide. And, you know, every time you sort of come up with a reason why we're not seeing the evidence we should see, it's explained away by a new trait of Bigfoot. It seems to me that, you know, one or two, yes, but when, you know, you have 10 reasons why the expected evidence we should see for Bigfoot is absent and then you have 10 kind of excuses. To me, it sort of begins to go, well, maybe the more likely answer is not these 10 really improbable features are in this one creature, but maybe the creature just doesn't exist. Yeah, this is what professionally I might have called scope creep, but here I'll call hypothesis creep. Every time you explain away something, that's adding complexity to your hypothesis. So, you know, why don't we find a Bigfoot body? Maybe they're really good at hiding from us. Yes, exactly. That is apparently one of the traits. Yes, it would be a necessary condition that Bigfoot had several behaviors where it actively tries to obscure itself. Now, in the case of the red colobus monkey, they are very curious. So, they are very visual. They make a lot of noise. They jump through the trees. So, for hunters, they are very easy to shoot, and that's what makes them so vulnerable. Aww, but yeah, for the Bigfoot hypothesis to hold, it would be required that they bury their feces, that they generally wander alone, that they actively try to avoid human beings. They collect their dead and they bury their dead, you know, which explains why you don't find bodies. I mean, people also say, "Well, you never find a body of a bear in the forest," which then other people who are, you know, hunters go, you know, you find bear and deer bodies all the time, you know. But I actually find this sort of logical fallacy in reading stuff kind of fun. That's partially why I took this section to go over Occam Epicurus. I didn't even talk about Solomonov. I mentioned Kamagorov and these important people in this area. But probabilistic reasoning is one of my favorite aspects of the statistical side of data science. Yet, I see it misused a lot too, and I was really surprised when I got into this whole Bigfoot topic to find that creationist people that express a legitimate belief that the universe, or that the earth specifically is only 6,000 years old, are really interested in Bigfoot. They got this strange idea that if they prove that there's a lot less monster or a Bigfoot and it turns out to be some sort of evolutionary throwback that's still existing, that that invalidates evolutionary theory. It's ridiculous. You can tell that they really don't understand the basic concepts of evolution. They haven't thought about it. They haven't studied it. They haven't come to grips with what it even means. Often, they go to Africa. They have these expeditions still fun to go to Africa to find what might be a living dinosaur or some giant spider. I don't know what their idea is that they're trying to find, but they have this strange idea that if they could find some strange anomaly, it will shake science to the core. Yeah, but let's get down to now. Where should we be looking? I want to treat this subject with as much scientific scrutiny as I can. Let's go back to definitions. Bigfoot is a large, hairy, bipedal, North American. Wait, is Bigfoot only supposed to be in one specific area? Oh, yeah, probably should have mentioned that earlier. According to Wikipedia, Bigfoot is said to inhabit forests mainly in the Pacific Northwest region of North America. If you look at where do primates live besides humans, it's pretty much like along the equator. There has not been a primate in North America since like about the last 50 million years. There's not a fossil record of any kind of primates existing in North America in the last several million years, that kind of thing. The Bigfoot proponents, they've kind of got excuses. It's like, well, fossils are hard to find. And it's like you've got hundreds of Gigantapithecus teeth in Asia, which again, Bigfoot people sort of think, well, maybe it's Gigantapithecus or something like that. But it's like we've got hundreds of these teeth and we've got a couple of jaws and stuff like that, and teeth are pretty robust. Teeth lasts a long time. If we can find hundreds of these in Asia or just the faintest traces of early human civilization in North America 20,000 years ago, it sort of defies belief that this continent ranging primate that's apparently still alive somehow has not left any kind of fossil record. Okay, for a moment, let's talk a little generally about the best place to look in the world. We can all agree there's no Bigfoot anywhere in Manhattan. By definition, the Bigfoot hypothesis claims the species exists in the Pacific Northwest of North America, but for a moment, let's entertain other similar claims like the Yeti. Where would theoretically be the best place that a creature like that might exist today undetected? Maybe not a Bigfoot like primate, but what about an unknown primate in general? Yeah, I'm a primateologist, so I can only tell about primates, but especially from the near tropics, I think there are still much to learn about biodiversity, and that's also about showing several discoveries from just last years. There are primate species out there that are not described by scientists. Most of the time, these primates are known by the local people and sometimes even scientists have seen them, but often people don't realize that it is a different species. Because sometimes the difference are only very slight, or people don't pay attention if you're not really into taxonomy and into species, then you might not realize that it is a different species. In Africa, we know there is just less biodiversity of primates, but also in the Congo basin, there has been a new species of Quenen that was described in 2012. And now it's 10 years ago when the Kipuchi, a monkey from Tanzania. So in 2005, this species was described and that was really a big discovery. So even in a country like Tanzania, where a lot of tourists go, even in such a country, there was a monkey that was not known by scientists. Well, I've done a fly fishing expedition to Alaska, out on the sort of in the southwestern sort of peninsula area there. I'm going in a month or two here to Russia to go wilderness fly fishing in the Kamchatka peninsula. Maybe there is a Bigfoot out there. They actually do have a legend of a Bigfoot out there. I will keep your eyes open. So I will collect information about their mythology. Maybe it's not your mythology, but I'll collect some information about it. That would be cool. Yeah. If there was a chance such a creature existed, it must be hiding in these remote areas. So I feel that if anyone's seen one of my guests, I would guess it would be a world traveler like you. But sounds like no personal incidences? No, none at all. I would welcome any kind of personal incident of any type that was unusual. I haven't had any, unfortunately, but I'll keep my eye out. And as far as the Kamchatka thing, it's really interesting. I will follow up because it's one of the reasons I'm going. It's supposed to be one of the last really truly wild places on Earth. It's supposed to be kind of like what Alaska was 150 years ago. So I think it would be one of the last places where a Bigfoot would be if we had one on Earth. Do you know any rough statistics about like number of people per square mile or anything like that? Kamchatka is about the size of California a little smaller. It has about 180 miles of road. It has tens of thousands of rivers, many of which and the ones I'll visit have never been fished by humans or seen by humans. The Russian government used this as a military base on the off limits for a while until the 90s. So it kind of opened up and they're actually going to start pushing, probably you'll see more tourism there. Interesting. Almost too bad though, if it's this kind of bastion of wilderness. That's true. In a way. I mean, there's a balance there. You know, when I started this project, I thought maybe I'd be coming up with some sort of two-dimensional probability distribution over the whole planet that estimates, you know, the likelihood a Bigfoot-like primate could be hidden there. And I actually think that's an incredibly complicated problem that I'm simply not qualified to answer. But given that the Bigfoot hypothesis, as I've said, places him in the Pacific Northwest, that's where we're going to look. Well, I'm from British Columbia, Canada. You know, it's a forested region. It's a coastal province. It's very large. You know, the population is not that high. And we have a lot of monsters here. So, you know, here in the provincial capital, Victoria, we have a very large sea serpent. We have a crazy fishman, just like monster. Over on the mainland, they have Ogopogo, a large lake monster that predates NASA. Bigfoot was born in British Columbia as well. So let's talk about reported sightings. There's an organization that accepts reported sightings in North America called the BFRO, or Bigfoot Research Organization. Their website, BFRO.net, has a public-facing catalog of reported Bigfoot sightings, along with a form for people to submit their own. I assume it goes without saying that I crawled their site, did some data mungings and made a cleanup, and produced a useful data set for myself to see what I might learn from it. Now, there's some questions here that data scientists sometimes fail to ask. Where did my data come from? Was it collected in an independent and identically distributed manner? If not, is there a heterogeneous data set, or does it reflect different data collection processes? What do I know about the bias of the data? How do I measure the data set's reliability, and so on and so forth? These are questions of data provenance. Now, any reasonable person will agree that this data set is composed of some percentage of hoaxes, of misidentifications, and some percentage of actual Bigfoot sightings. Even, I presume, the BFRO itself would agree with that, though they might assume a lower percentage of hoaxes and misidentifications that a skeptic might. Yeah, I mean, I love that site because it's like, when you look at it, there are literally Bigfoot sightings in, I think, every single state in America, except for Hawaii, and I think probably about 7 out of 10 Canadian provinces. You have this large primate that somehow ranges all across North America. You have to start to wonder, maybe it's not Bigfoot, maybe it's just human perception being fallible. Do you see things that are not really there? I think, ultimately, Carl's on the right track here. I did a fair amount of analysis on the BFRO data set I prepared. If you want to look into some of that, you can see some of the highlights at datasciptic.com/bf. Of note, is the fact that when you compare the reports to population density, you find a high volume of reports in low-density areas. The Yukon, British Columbia, Manitoba, and Alaska being the top four in terms of sightings given their density. But at the same time, these are the places that are likely to generate more sightings, given that this is where we expect to find Bigfoot, if it turns out to be a real species. Perhaps it's like having a Mickey Mouse sighting at Disney World to supplement what Carl was saying, we have 49 sightings in the state of New Jersey. So, ultimately, I think these efforts are on the right track for an organization like the BFRO to be doing. But since I don't know how to classify these sightings into hoaxes, honest misidentifications, and unknown, I do not know how to proceed with any useful analysis. There are 10,000 reported sightings, and if only 5% are accurate, then, you know, there must be something there. So, what's the fallacy there? I suppose my flippant answer is, "Two half-truths do not sum to a whole truth." You know, there's a ufologist named Stanton Friedman who makes a similar argument. He looked at the government's investigation into UFOs where they kind of explained that, you know, some were natural phenomenon, some were test planes, that there's some small percentage that remained unidentified. And he likes to latch onto that and say, "Look at these." But just because something is unknown doesn't mean filling the blank with your favorite hypothesis. Yeah, yeah. Well, I think it tells us described as a big pile of very bad evidence does not equal good evidence. You know, maybe you should come up with the Politch equation, all right? What it does is it describes the probability that this thing exists, and it's a series you just multiply a bunch of constants together that you pull out of thin air. It's fun, right? You know, I may do exactly that. This reminds me a lot of the Drake equation. Long-time listeners of the show should already be familiar with that. I was reading up Carl Sagan books and came across it. That's a clip from episode 33 where I interviewed Peter Bacchus and his unique twist on the Drake equation. He wrote a sort of tongue-in-cheek paper about the likelihood he would find a girlfriend. It just seemed like an obvious application of the Drake equation where you just apply in a series of increasingly restrictive criteria to a population. The Drake equation was defined by Frank Drake to estimate the number of intelligent civilizations in the universe with whom we might be able to communicate. One thing that's nice about the Drake equation is that it's a statistical construct, really. It separates the logic from the actual values of the parameters used. People can debate over the number of stars in the universe or the percentage of those stars that have planets one can plug in their own values, perhaps more liberal or more conservative than those generally accepted and come up with your own estimate of the number of such civilizations in the universe. I'd like to develop a similar set of scalar probabilities that predict the likelihood, however small, that Bigfoot could be alive today. If I make mistakes in my values, you can pick your own and argue about why they should be set one way or the other, but the logic should be consistent. So where do I begin? I'm going to start from some known species for which we have evidence in the fossil record. I'm not going to entertain anything science fiction like Bigfoot being a shape shifter or having the power of invisibility. I need to find the most credible historic species from which Bigfoot could either be a direct or an evolutionary relative. My favorite, this is totally off topic, but my favorite, not my favorite primates are Tarsiers. They're the ones that are like about the size of your finger with really, really large eyes, most of them are in the Philippines. And the interesting thing about Tarsiers is they haven't really figured out where they necessarily go. There's a little bit of contention, I believe, about like what they're most closely related to because they're just this kind of offshoot. If you ever look at a plate or gram of where Tarsiers are, there's kind of this weird offshoot because they're not lemurs and they're not loruses. They're not apes. So, you know, what are they? So there's an obvious candidate to point to, and it's one that Carl mentioned earlier, a genus of ape known as Gigantopithecus. Ah, Gigantopithecus. The species is believed to have gone extinct somewhere between 9 million and 100,000 years ago. There is fossil record evidence for its existence in Asia, specifically around China, Vietnam, and India. Gigantopithecus blackie is the largest known ape and stood at an estimated 3 meters or 10 feet tall. Let's see if I can, let's see where he ends up in the, um, plate or gram. You know, that's entirely possible. So what's possible is that a creature like or descended from Gigantopithecus blackie could, in theory, look like what we call Bigfoot today. So I have my candidate species for the first term in my equation, but of course, the existence of Gigantopithecus blackie 9 million years ago is a long way off from having a modern day Bigfoot. I need to figure out what the other criteria will be. Right. No, I like that. So if Gigantopithecus, Gigantopithecus apparently did have a large range of existence, right? So you're right. And so if it is the some family pungidae, so it is a hominid. So it explains why it's bipedal. So if that's from 9 million years ago to as recently as 100,000 years ago, and if it's decided to pawn that depending on where it ends up in the spectrum, possible, maybe, but if the origins of that particular genus was in, as you said, in Asia, maybe we're looking in the wrong place, right? So not that I know much about the migration patterns of some of the other apes, but if it originated somewhere like China, India, Vietnam, that sort of area of Asia, it makes you wonder one, is everyone looking in the wrong place? To what sort of living conditions were they was Gigantopithecus in? And then how do you, what would that say about the evolution that it's too Bigfoot, if that's, if that is a thing, is what they were, what it actually ended up as? I'll eventually plug in some numbers here, but for now, I'm going to follow up on some of the Newell's suggestions, and let's get a basic equation put together. We start with the probability that Gigantopithecus blackie was a real historic species. We multiply that by the probability that the species was bipedal. Multiply that by the probability that the species could have evolved into Bigfoot, multiplied by the probability that it exists today in North America, which also includes the assumption that it migrated to North America from where we knew it existed in Asia. So, I'm counting four important values so far, but we're not done yet, but let's just check in with Daniel to get a rough sense check of where we are. I would say that the odds are very close to zero at this point, because, you've heard the old saying that absence of evidence is not evidence of absence, that's only true if there are no kind of predictions entailed in the claim. And what Bigfooters say about Bigfoot actually involves a lot of implicit kind of predictions. So, for example, Bigfoot leaves a lot of footprints pressed deeply into the ground. Well, an animal that does that can be tracked. Bigfoot is large and has large bones. Well, a creature that does that is easy to see and their bones can be found. Bigfoot is often seen. That's why we believe in Bigfoot at all. But an animal that is often seen is often photographed, is captured, is available for study. There's a very difficult to resolve contradiction between the evidence that Bigfooters cite and what that evidence really implies. So, the fossil record does confirm the existence of Gigantopithecus in Asia, but could that species have come to North America? Unfortunately, it doesn't seem like the timing lines up. The species is presumed extinct between 9,000,000 and 100,000 years ago. The Beringian land bridge is only believed to have been exposed in a few eras more recent than that, the oldest being about 35,000 years before the present day. As an interesting aside, I read a 2007 paper titled, Genetic Variation and Population Structure in Native Americans, which suggests that human migration to North America actually occurred in a single wave about 12,000 years ago. Okay, but what if the species came in some tiny, tiny population that was just small enough to make no mark on the fossil record, but big enough to survive? Realistically, that's incompatible with the extinction timeline. But, let's call it 5%. I really don't want to say zero here. 5% was good enough for Pearson, right? So, I'm going to add one final term to the equation, but it's one that we'll have to unpack a little bit. Taking the results of the first four terms mentioned previously, we now multiplied by the probability that a minimum viable population of bigfoots could exist today in North America undetected by human beings. What's the size of that population? The general rule that biologists use is that you need about 500 individuals for any species. Yeah, this is what I found in my research as well. The rule of thumb is between 500 and 1,000 candidates to ensure between 90 and 95% probability of survival for between 100 and 1,000 years into the future. Okay, that's a lot of betweens. So, let's take the median value on likelihood of survival and number of years, giving us 92.5% chance of extinction every 550 years for that minimum population. All I need to do is figure out the probability that 500 bigfoots, bigfoots, bigfoots, could exist in North America. I've never been in North America, but I would think that a lot of people out there, a lot of people that also can report sightings. And if there is no proof for the occurrence of a primate out there, I would say that chances are pretty low. All right, so my goal, obviously, is to formalize that a little bit. Now, in truth, a proper calculation like this requires complex simulations of dynamic population systems and should probably include historical weather system data and a host of other features that add precision to justify their complexity. I was hoping to find a guest to bring on to this episode who specializes in this area, but my search did not turn up anyone doing this sort of work in a robust, stochastic sort of way. I'd love to learn more about anyone's work in this area that I miss. Please feel free to reach out if you were someone you know works on systems like this. I think these sorts of statistical models is great fodder for future episodes, and I'd love to revisit this, albeit for species in general, not necessarily just for this mythical species. But for a back-of-the-envelope calculation, I'll treat this as a binomial extinction threat event with a survival probability of 92.5% happening every 550 years on average, meaning 181 crisis events over the last 100,000 years. This is my probability of survival model, and I'll reveal its output in just a few moments. But first, let's discuss the last term in the equation, the probability that Bigfoot could exist unseen. Now remember what I said earlier about the best evidence. Now let's take a moment and ask how the famous Patterson Gimlin film fits in here. After all, if the film is good evidence of Bigfoot, then the probability the creature could exist, but be hidden should be high, since it would mean both that they exist, and that they evade being filmed before and since. Since the PG film has been deeply investigated, some people believe it's the most widely recognized bit of film ever shot, although I heard that fact before Gangnam Style came out, so who knows. In my evaluation of the evidence, mostly about the people involved in the circumstances of the footage being shot, I do not find the PG film convincing. But my non-expert opinion really isn't the point actually. This video is consistent with both the Bigfoot hypothesis and the man in the monkey suit hypothesis. So you rely on our prior probabilities, which is much higher that Bigfoot doesn't exist. If I had to put a number to it, I actually think the PG film would be lower than the parameter I'm going to use, somewhere in the single digits maybe even sub 1% likelihood of being real. But I'm feeling generous here, and I want to acknowledge that in my previous calculation dealing with the minimum viable population, my model really suggests a species teetering on the brink of extinction over the last 200,000 years. That being true, it's plausible for there to be a lack of fossil evidence. But how plausible? I looked at rates of fossil discovery for homo sapiens over similar time spans. By the way, I do realize that saying a bunch of numbers out loud and talking through the arithmetic of them is really kind of boring the most. I get it. You can see my work for some of those calculations online at datascaptic.com/bf. That's bf is in Bigfoot. But jumping to the conclusion, if in fact, a steady state of minimum population of 500 individual Bigfoot creatures has survived in North America and not ever shown up in the fossil record, then based on the rate at which we find human fossils, of which we have 41 to date, according to Wikipedia, given the historical human population. I estimate that there's a 34% chance that a tiny population would not have appeared in the fossil record. But keep in mind, this reflects the fact that I'm assuming they're at a functionally minimally sufficient population for a long period of time, almost a ludicrously unsustainable period of time. But that fact is captured in the previous step, and you'll see that reflected in the scalar value for that step in my next calculation. So before I really get down to tallying things up and getting to my final probability, there's a few things we have to revisit. My first term, the probability that Gigantopithecus blackie was a real species, I never really connected with the right person to explain to me how we can be sure a species existed from a collection of teeth and a few partial jaw bones. But I'll take it on good faith, and I'll call this a 5 sigma confidence, giving us 99.99% likelihood that Gigantopithecus blackie was a real historic species, basically certainty. Regarding the likelihood that Gigantopithecus stood upright, no one has ever found a pelvic bone or a leg bone, so we're not totally sure if it walked upright or not. Grover Krantz, who I mentioned earlier, argued that certain features of the jaw indicate it was bipedal. Grover appears to be in the minority on this one. If the rest of a Gigantopithecus skeleton resembles any other great ape, it would not have the physiology to walk upright. But given that there's some uncertainty, despite being in the minority, let's give Grover's theory a chance here, and let's say there's a 20% probability that Gigantopithecus was a bipedal creature. Maybe I'm being generous, I don't know, but I definitely don't think that's unfair of anything I think that's high. So what about the probability that Gigantopithecus evolved into Bigfoot? Keep in mind, this is conditioned on my previous two steps. If I accept Gigantopithecus was bipedal, then it would match the description of modern day Bigfoot fairly well. Let's make this one a freebie, 100%. Okay, I now have all the pieces in place, sort of. I'd really hoped I could get precise in this part of the episode, but what I got for my panelists was directional. I would say that the odds are very close to zero. Chances are pretty low. I'd say that's really, really, really close to zero. I don't blame them, these are hard questions. It's difficult to think quantitatively about unlikely states of the world, but I put all the pieces in place, let's see this thing through. These are my estimates of these probability values. The probability that Gigantopithecus blacking was a real species. 99.9999. Multiplyed by, probability it was bipedum. 20% Multiplyed by, the probability that Gigantopithecus could have evolved into the modern day Bigfoot. 1.0 Multiplyed by, the probability that Bigfoot could have arrived in North America from Asia. 5% Multiplyed by, the probability that 500 individuals could survive over the last 100,000 years. 8.64 times 10 to the minus 6 Multiplyed by, the probability that that minimalist population could go undetected by human beings. 34% equals 2.976 times 10 to the minus 8. So yeah, I don't know how to react. Like, what does that mean? What do you mean? What's, what's more likely is more likely I'll be struck by lightning five times. More likely I'll fly and become a unicorn like. I think it's more likely I'm flying and becoming a unicorn. Yeah, I'll give you that. Let me look at lightning for you. Does it have to be five times or could we just do ones? I'm just asking like, what is the probability that we would get struck by lightning? I don't know, I'm googling it. The answer to the question, what are the odds of being struck by lightning is actually, it depends. So, okay, so here's how this site does it. The estimated US population as of 2011 is 310 million. The average annually reported US lightning deaths and injuries from 2001 to 2010 are 280. Thus, the odds of being struck by lightning are one in 1.107143 million or approximately one in a million. So, Bigfoot is less likely by my calculations than being struck by lightning. But even that's a tough way to look at it, right? Because struck by lightning is within your lifetime. Well, I feel like if I were to see Bigfoot and Bigfoot was that rare, it would be considered an event. Well, yeah, that would be your personal event if you're seeing something. That's a personal event if you get struck by lightning. It's a deep, but I feel like a doctor could confirm it. Oh, you mean you want it to be compared to something where someone else can confirm it? Yeah, I want evidence. How would someone provide evidence that they actually saw Bigfoot? Oh, well, we talked in this episode about DNA, about just a body in general. Mostly those two, I guess. Yeah, I mean, me personally, I'm having a hard time visualizing the difference between these probabilities. You make a really good point. Like, when you talk about these really tiny micromort kind of numbers, it's... Let's assume the average of being struck by lightning is once a year. Well, one... One... One person once a year. How many times a year does one person get struck? I'm confused. So according to this slightly maybe questionable side, I'm on 280 times a year. There are lightning deaths or injuries. So that's like basically one in a million chance of that happening to you. So that's more likely than seeing Bigfoot in my lifetime? That's more likely than Bigfoot existing. Oh, in general. Yeah. Yeah, I mean, that's really tough because number one, being struck by lightning is something that happens during your lifetime. Yeah. Whereas Bigfoot existing happens throughout the ages. Yeah, that's an interesting point. But like, the odds of seeing Bigfoot are tricky because you could see a man in a gorilla suit, you know, or you could see a misidentification. So I don't know if I've gotten anywhere at the end. Yeah, I mean, can you like, write it out? The two differences or be like, "This is the cup that shows how likely are visually to get struck by lightning." And here it is in relation to Bigfoot existing. Well, I guess you're kind of getting down to the fact that this is an apples to oranges comparison now because struck by lightning has this implicit on a per year basis. So we automatically extrapolate that into our lifetimes of, you know, between, hopefully at least 27 to a little over 100 years or so, whatever. But like, yeah, the odds that Bigfoot is a modern day species is a bit more open-ended. Well, I mean, you could make lightning, then lightning strikes statistic more like something that happened ever, which is let's assume we live forever. What are the odds of being struck by lightning ever? If you live forever, then the odds of being struck by lightning are 100%. There's an imminent amount of time within which that event can occur. But if you live forever, it doesn't mean it would occur. It kind of does, actually. Like, let's say there's 100 people and they decide to live. I don't know, forever. Just decide. I'm going to live forever. But on this side, there's 100. You're saying all of them would be struck by lightning. Yeah, actually, yeah. Well, that would be at some point, but like... Right, at some point. Well, I guess for Bigfoot, you're just saying for how many years you're saying forever, ever, it never existed. No, it doesn't exist right now. That's my question. There's this temporal element that's really tricky. So then how do we juggle that time with comparing to the flight? I was hoping I could just sidestep that. Yeah, I mean, I don't think you really have a comparison. Can you think of one that's better? Better, in what way? Well, I gave the struck by lightning one. Is there some other probability? No, I mean, I really like that. It's tough. So first of all, all my scalar values are sort of my approximations, because I was really interested in more of the logical process than the actual values, because I could not pin down experts. Like the people that know about Gigantopithecus, they actually didn't want to talk to me. The two that got back to me both said no, and other people didn't respond. Why'd they say no? One guy said, "I don't want to talk to you because I don't talk about Bigfoot because Bigfoot does not exist." So implicitly, that means zero probability. And the other person just said, "No, thank you," in a very polite way. Yeah, I mean, you can't blame them because, let's say you're like a serious... I don't know what... Researcher, yeah, they're both like legit, serious top level. Like, I don't want to say their names. Yeah, then like, let's say you devoted your life to researching something that actually existed, and then someone asked you to come on a show to talk about something that probably didn't exist. I mean, that's kind of insulting. I don't know that I insulted that. Like, wouldn't that like knock down their credibility? Like, "Hey, I talked about Bigfoot." Yeah, I guess so, I mean, I would hope that it would be more like I talked about statistics and logic, but I see your point. Maybe that's why I had trouble getting the Gigantopithecus people to talk to me. I just have a hard time, like, you just spit out a number and I don't really still know what it means, you know? Yeah, no, that's a fair point. Like, I don't know, do you think it was worth it? Is this number worth it? Well, there's a clip at the beginning of this show from about six months ago. I don't know if you remember, we were walking down Santa Monica Boulevard there and I was recording us. Do you recall this? Yeah. Yeah, and then you remember what you said? I'm not sure. Can you tell me? You were putting a lot of less defense and you said something like, "Oh, it's cute and it's about the journey." Well, I mean, if you're just going to say a number, sure, but you were talking about a number, maybe we should talk about your journey then. Yeah, we should. Well, I mean, everyone's just heard that. I think my logical process is pretty decent. My number is just, it is what it is, but I talk a lot about some really cool things about updating beliefs and I got a lot of good feedback from my guests. I hope that's the part of the message that carries through. That what people update their beliefs? No, about how we entertain evidence and how we look at the likelihoods of very unlikely things. So, I mean, I feel like maybe you're building up to this number and you should build up to more like, "Duh, duh, duh, what did we learn?" Yeah, I guess so. Or like, let's, you know, let's do a recap of all our journey. But I wrote all the music for the countdown part where I get to the number. Well, what if you counted down and then you said, "So..." I kind of took a step back and then I figured out it's not about the number. Okay. And you left them hanging. Ooh. Okay. And you actually more so talked about what you learned from it and was like, I mean, the summary is that the number is very, very small. Yeah, yeah. And actually like, and I tried, you know... And the number doesn't matter, I guess. Yeah, like, I tried to compare it to getting struck by lightning, but that was so hard because we live for a set amount of time and then big flow. We're just talking about, "Did it ever, ever exist?" So it just seemed like the number distracted from the message. Yeah, you might be right. So you could build up to it and like, "Dah, dah, dah." And I decided not to tell you. I can't rob them of it. It'd be like cereal. I think that's a great ending. Okay. Because I think people are going to be like on the edge of their seeds. And you could just give them the real killer, which is... I'm never going to tell you. The number is so small. It's like hard to imagine. In the end, what you're saying is how you think of the data matters so much more than the number. Yeah. And that's what you're trying to communicate in your show, right? Yeah, yeah. And in this case, this wasn't some data set I could just come up with. There were so many things to look at. And there wasn't just some big CSV. I downloaded off some portal and ran with it. This is more about logic and probability. Logic, I guess, more than probability. Yeah. I mean, if you're trying to teach people how to think about data, you should really emphasize that at the end, not your number. Yeah. That's true. The number is kind of boring, especially if it's basically zero. Well, let's call it epsilon. I don't know what that means. Oh, that'll be a good mini episode topic for the future. Epsilon. Great. Okay. So I would say to Bigfoot hunters, you have to go and get a Bigfoot. There's just no other way. And so even some very preposterous proposals for how to do that, hovering over the forest on a blimp, listening quietly for Bigfoot, are maybe more fruitful possibilities than casting more footprints. I tell you what, when I hear these stories about, we've got a Bigfoot body. I get excited and I'm thinking, oh, I finally got a body. And I can't help but be excited and hopeful that maybe they actually do have a body. It's always disappointing. It never turns out. Where do we go from here? When will a lack of evidence mean an end to the Bigfoot hypothesis? It did flare up quite a bit in the past couple of years. And with the Finding Bigfoot TV show, which is one of the most absurd shows on TV, hopefully that's fading away. I'm hoping that this is the last season for them. I just want them to go away because it's just so silly. And I think that Bigfoot also comes in waves. He gets popular and then he goes away and I'm glad I was able to take a break from Bigfoot for the past year since Ketchum's fiasco. Been a little bit quiet and I'm happy to see that. My intent with this episode was to do something fun and different for a special release and to talk about a few ideas related to models, statistics and beliefs. Perhaps this was a novel way to share what might otherwise have been sort of boring topics to some. I didn't do this with the intention of necessarily changing the minds of any true believers in Bigfoot. However, I thought just maybe I could challenge them with their approaches, being more rigorous and whatnot. Daniel Axton, how abominable science, which he co-authored with Donald Prothro, was received by the cryptozoological community. I'd say that by and large, it's fair to say that they hate, hate, hate it from the very depths of their soul. The Bigfoot times towards pieces has been widely flame, lots of one star reviews on Amazon, sometimes by people who never read the book. It seems to have really touched a nerve. There are many people who are kind of more serious in cryptozoology, less your kind of true believers and more on the scholarly end of things. I would be much, much more friendly about it. My co-author and I had sort of a running bad about whether or not cryptozoology had it in it to step up to the plate and just take a good faith critique in good faith and engage with it on that level. You know, it's been kind of a mixed bag, I think in the end, Don, who had very low expectations of cryptozoology as a subculture, he may have been right. But on the other hand, there have been some very kind and thoughtfulness as well. American philosopher Daniel Dennett, who studied many things, primarily consciousness, he had a wonderful thought on explaining consciousness to people and magic as well. So he said, if you explain stage magic and the workings of consciousness, likewise to people, they will treat you very badly because it's like telling people that Santa Claus does not exist. You're a party pooper, you spoil the fun, you reveal secrets and break the romantic atmosphere of mystery. Even though I myself agree with Richard Dawkins, who said that even if you perfectly describe physically how a rainbow works, how it's constructed and what causes it, it's still a miracle of nature and you can appreciate that beauty easily. But again, I think in this sense of believing in big foods and things like that, this mystery attracts us. Yeah, I love a good mystery and I have to say, as I've delved into the books and podcasts that the Bigfoot community puts out as part of my research, I really did develop a fondness for some of the folklore and some of the characters the same way I enjoy but do not believe in a good ghost story told around a campfire. It's a community, like any other community, with some reasonably scientific and skeptical people, a couple of wild and unbelievable characters and some fierce and fighting at times. It's a really mixed crowd, the cryptozoology crowd, I mean in some ways I'm part of it and now you're part of it. I actually got a bit of a chill when I was interviewing Daniel and he mentioned this. Bigfoot is a topic I thought I would touch on for this episode and then walk away from. If I'm now a part of this topic, there's a good chance people will want to engage me on it. Am I willing to be pulled in? Do I want your emails? Is it okay to send me your photos and other evidence? Sort of. Okay, so please do not send me your personal anecdotes. Personal experiences are only convincing to the experiencer. Don't send me your footcasts. I'm frankly not even qualified to look at them anyway. If you'd like to review the show, please do so in a public forum, maybe someplace more permanent than Twitter like a blog or a Tumblr page, let me know about it. If you make a noteworthy correction or have an interesting critique, I'll comment. But I think we're approaching the period at the end of the sentence of all that I have to say on this matter. However, if anyone has a legitimate data-driven claim or an interesting data set with good provenance and sampling about Bigfoot, I'm happy to engage. If you need advice on how to get your trailcam data uploaded in open-sourced, and if I can find the time, I may be able to give you some advice on this. In fact, that would be great, actually. If your trailcam has convincing evidence for Bigfoot, that would be really interesting. But I find it quite unlikely. However, your trailcam probably has a lot of great observations that would be interesting to biologists. I suspect the people over at ROpenSci.org would love to have access to that. And if not them, someone for sure. But in lieu of a body, or some other credible evidence, or a data-driven claim, this is all I have to say about Bigfoot. I want to thank all my panelists for their contributions, and I want to thank Linda for all her support. A lot of great conversations hit the editing room floor. For this reason, I'm assembling a torrent file with a complete, unabridged interviews for anyone that would like to hear a bit more. Check data-sceptic.com for more information on that. Data-sceptic resumes its regular format of alternating mini-episodes than interviews next week with a mini-episode on the Knearest Neighbors algorithm. If you enjoyed this or any other episode of the podcast, please go over to iTunes or Stitcher and leave us a review there. That's the best way you can support the show. Until then, this is Kyle Polich for Data-sceptic.com, reminding everyone to keep being skeptical of and with data. [MUSIC PLAYING] [BLANK_AUDIO]
