Would you pay for transparently useless advice? Nick Powdthavee at TEDxZaragoza

Hello! ¡Gracias! ¡Hola! ¡Buenos días! My name is Nick and I have a confession! And my confession is that's practically all the Spanish I know! (Laughter) So...For the non-English speaker I guess that was my TED talk! For everybody else, thank you very much for having me here today. What I would like to talk to you about today, might sound a little crazy at first. But please do bear with me. What I'd like to tell you today is that last year, my colleague and I had come up with a systematic way of predicting with 100% accuracy, future winning lottery numbers! (Laughter) Now, this might sound, like I said, a bit crazy, but as I guess for everybody here today, for this early morning session, here are our predictions for the 27th September, next Friday's lottery numbers. Now, you may wish to write this down! (Laughter) But I know what are you thinking! I know what are you thinking! That's crazy! It's really crazy! Only fools would believe that somebody has that predictable power of predicting the future! Well, I'm standing here, right here today, to tell you that it can be done! Future can be predicted! It is not impossible to predict the future, it is just highly improbable! (Laughter) I'll show you how, how we do it. But in order to understand how my colleague and I do it, we must first understand this man right here. Now, I don't know if that's any English person here in the audience today, probably not, but if you don't know this is Derren Brown. He's a famous British mentalist, magician, illusionist. And one of his shows is called The System. So, this is a TV show that he put on, I think about three years ago, and you can Google it, it's on YouTube. In this show he said he has a system for winning at the horse races. More specifically, the whole show follows this young lady right here and the whole show is built around the idea that a system can be developed for picking the right horses that will win; at each different horses around the world. So the camera followed this young lady right here for many days as Mr. Brown anonymously sent tips, correct predictions of 5 consecutive horse races in a raw. As the problem goes on, this lady will get a tip, she'd go to the betting place, and she'd bet on a horse. And that horse would win! So she starts betting more money, as the next tip comes in, and the tip was correct again. This happened 5 times in a row! And in the final race, she was encouraged to put her entire life savings on this one final horse. And that's when she met Darren Brown. And he explained to her how he did it. This is how he did it! She wasn't the only person in the whole show. He started off with 7,000...700 people and he divided this 7,000 people into six columns and then he would anonymously send tips of horse number one to the first column, horse number two to the second column, horse number three, fours and so on. Lady just happen to be in the second column, where the tip was correct. So, the horse number two came in the first race. What he did next, was he eliminated everybody else, leaving only that column. And he further divided that column into six more columns for the second race; then he sent tips for horse number one to column one, horse number two to column two, and so on and so forth. And he redo that five times in a row. It's just that she happens to be lucky to be in that particular column. So the tip had no predictability whatsoever. She was just lucky. You can imagine that! This lucky string of her! She was willing to put her entire life savings on this final horse race! Now, come back to the winning lottery draw. The odds of winning is one in 3.2 million. There are 47.4 people in Spain. I can send everybody in Spain different combinations and I guarantee that each week I'll have 14 people winning the lottery. Now, this is all nice and well, but imagine, if I was saying to you that "Look, I'm an expert on this. About lottery predictions. I'm gonna send you tips, ok? And, if you win, and only if you win, you're gonna pay me some money." (Laughter) "10% of your winning. Only if you win!" Fair! That sounds fair to anyone! That's not even a scam, I'm not even asking you for money! Now, if you think about this, it does resemble a little bit of people who work in financial sector, doesn't it? The Mutual Fund Managers, right? They are trying to predict the future! Now, obviously, with Mutual Fund Managers and with horse races, people can still believe in it, because people tend to mistake luck for skills. There's some kind of fussiness in it. Because, you know, people working in the Financial Industry, politicians, they must know something that we don't know. That's why they make all these predictions. We're not going to fall for this scam of lotteries or these things because we know they are completely random. Well, that's what my colleague and I wanted to test! I have to say I am a Professor in Economics from the University of Melbourne and London School of Economics and I am a Behavioural Economist and we do experiments. So, along with my colleague here, we wanted to test whether people are willing to pay some money for predictions of future coin tosses. So, me and my colleague here, holding the award for the most useless advice ever given to students, we carried on our experiment. And the experiment was "Imagine you are one of the top students in Singapore in the entire school", 'cuz this is where we ran our experiment... Imagine you are one day randomly selected to come to the experimental lab, and as you walked in, you've seen many cubicles, and you're given some numbers randomly to assign you to a random cubicle and there you go and sit down. You see this five numbered envelopes, stuck there on your table: one, two, three, four, five. And then you're told that we can give you some money and you can use that money to play a game. And this game would be to bet on the outcomes of coin tosses. And we play five rounds of them, and it's up to you how much money you want to put into the bet. So, if you bet correctly, you doubled the amount that you bet. if you bet wrongly, you'd lose. In each round. And you get to keep whatever the money that's left at the end of the day. Sounds very simple, sounds very easy. So, for example, I have a coin right here, just to demonstrate. Raise your hand if you thing that's going to be head. I' gonna flip this. Raise your hand if you thing that's going to be a head! Ok, ok! Thank you! Now, raise your hand if you think is gonna be tail. Ok, fifty-fifty. Now raise your hand if you have no idea whatsoever what I'm talking about! (Laughter) Ok, so I suppose, everyone! Ok, so, let's see! Head and tail! Right! Ok! Oh, sorry! I thought it was a British pound coin. It's actually a Euro, so... (Laughter) So, I guess, everybody losses! Except Greece! (Laughter) So, anyway! You got the idea! You got the idea! So, the thing is there is something else to go into the experiment, because we said to these, you know, highly intelligent students, that, before you place a bet, these five envelopes, you have the opportunity to pay a small fee to open the envelope, the corresponding numbered envelope. And, containing within these envelopes are the predictions of the coin tosses that hasn't happened yet. Envelope one corresponding to round one, envelope two corresponding to round two, and so on. But you don't have to buy, you don't have to open it. But if you buy, you can open it before you place your bet. But you don't have to buy it, the buying is not compulsory. And you can see it for free. You can see what's inside the envelope for free, but only after the coin has already been tossed and the bet has already been made. In each round, ok? Very simple, very straightforward. Now, obviously, most people would say "Oh, that's ridiculous! Nobody's gonna pay to see this! Why would I pay some of my money to see what I can also predict myself?!" But we played five rounds of this. But imagine you are sitting down here... So, the order would go like this. You'd come in, the first thing is, you choose whether to pay or not. You pay, you'd get to see what's inside, and inside it's just nothing, just H or T. That's it. Nothing fancy. Just H or T. We just showed them to them as well. That's no value whatsoever. So, if you pay, you get to see it, then the next step, you make your bet, the coin is tossed, you see what the outcome is. Then, if you didn't buy to see the envelope, you can open it and see what's inside for free. That's end of one round, then we begin round two. What the students don't know, obviously, is that...Oh, before that! To ensure that the coin...--because the people will think, you know, these coins are not fair, we even get the coins from you. And you'd be the person who tosses it. We won't even touch it. And we've got to change the coins in every round as well. And you even change the people who tosses it. So, just to make sure that the coin is fair! Of course, what the students don't know is we did exactly what Darren Brown did. The first envelope, divided in half, so half would get head, half would get tail. We further divide them, and so on. So, you always make sure that, at least some people would get correct predictions of coin tosses all the way through. So, how did our students do? How did our future leaders do in this particular case? Well, as you can see, look at the bar graph, after only a few correct observations of the coin tosses people got confused. They started buying the next prediction. So, that's the first bar here, so, at round five, you see four correct predictions of coin tosses in a row, 50% of you started buying, paying, in round five. The funny thing is that the bar next to it are the people who keep getting wrong predictions in a row. These are the other group! They also bought in the last round! We've also found that they bought it and then bet the opposite! (Laughter) So, they are still quite clever! 30% randomly bought... So, did the participants who know more about probabilities make less mistakes? They got them to do some tests as well, probability tests. Well, we found out, no! The smarter one do fall for this! Nothing! No statistical differences between these individuals! So, why did they buy the advice? Is it because they wanted to please us, the Professors? They wanted to, you know, we can get a pay for publish, the scientific research? Oh, no! Not really! Because when they buy, they start betting more. They start betting a lot more money because they really believe in this predictability of the envelope. There are many reasons or many potential explanations to this. One of which is that you want to delegate your decision-making to something else. You don't want to be responsible for it. But one stood out the most. And this one, I particularly like. And it comes from, not Economists, but it comes from Psychologists. And the Psychologists call this 'Hyperactive Agency Detective Device'. I call it 'The Jesus on Toast Theory'. (Laughter) And this is a theory that says that humans are hard-wired to detect patterns in otherwise unrelated events, for surviving purposes. When we see something moving in the dark, it's much more safer for us to assume that's something there so we are not gonna be eaten in the evolutionary process. But these things doesn't really matter so much anymore, you know, in this age and age. And I'll come to the end. The most important question of all. 'So what?' Why should we care about this? What's the point of my talk today? (Laughter) Well, I got my paper published, but...anyway! The point of my talk today is that according to psychologist Philip Tetlock, experts--and we see it on TV, and people are making predictions all the time--are empirically, no better than the average population, at making predictions. But for these 'experts', when they get it wrong, they are never held accountable. They are never punished, nearly enough! But when they get the improbable, correct, when they get that little..."Financial crisis, coming in 2014!", they said "I've got that right!" and they become famous, they get paid so much money for this! But we all know that future is highly unpredictable. So, let me end with this note. If it seems too good to be true, it probably is! Thank you very much! (Applause)