Scientific Analysis of Football Strategies

This BBC podcast is supported by ads outside the UK Queen Carvania stood haled by the morning sun An army hung on her every word. My champions, I have sold my chariot on Carvana. 'twas a lovely SUV, an inexplicably queenly offer. They're even coming to the castle to collect it. Tonight, we feast An offer you can feast on, seell your car today on Carvana Pick up these mailify Evening B' remorse Buy a new car I'll moving in Let's get started. U Sorry, I think there's been a mistake. I bought it from Carvana. You what? Yeah, great price. I even have seven days to love it or return it. So there's no No, no buyer's remorse. More like buyers rejoice? I guuess I'll let myself out. Congratulations. I mean it. Buyers rejoice. Buy your car today on Carbana. Limitations exclusions may apply See here Sundayorurn Picy, Carona d com Hello, I'm Tom Whipple and welcome to Inside Science from the BBC World Service The thing about football said the author Terry Pranches The important thing about football is that it is not just about football Here at Inside Science, we agree Don't worry, this is still a safe space, a World Cup exclusion zone, just about For the next half hour, we have no interest here in Harry Kaine's left foot, Ronald Dino's right foot Zidan's forehead. whatever body part is pertinent, whatever other footballer is famously good at doing ball related things with it Do doesnn't mean we aren't into the footie When the world gets together to watch those feet and balls and foreheads, to cheer, commiserate, celebrate That means only one thing statistical power Ecology studies at scale, science from the epidemiology of population scale sleeplessness to the optimal game theoretical strategies in shootouts and beyond What can we learn from football That new segment comes later And with it is Kit Yates, maths prof and football fan Who will be our boot boy as we choose a World Cup squad of science L it Hi Tom, thanks for that First though, this will be the last World Cup where teens in Britain can follow the games on social media This week, the UK goovernment announced under sixteens would soon be banned And the kids, well, they're not al right with that. and they have questions I would like to know why they chose under sixteenens and not like under fourteenens or under seventeenens Are you sure the ban will really work? Britain follows Australia in banning teens on social media, it is said one politician, a defining moment It is said one teteen. A waste of time They look forward, they said to a future, staring at a wall How do we decide who is right? We spoke to Professor Amy Orbin, program lead of the Digital Mental Health Group at the University of Cambridge and Dr. Katherine Sebastian Head of evidence at Wlcome. But first, we asked a group of year nine students at Clifton High School in Bristol what questions they have about the band How long did it take for the scientist to come to the conclusion that these apps needed to be banned? What's the thought process behind banning like communication Because a lot of people like talk to people using Snapchat and stuff like that. I'm wondering if once people reached the age of sixteen They might go on even more because it's like the forbidden thing they're now allowed to have access to. How did the scientists come to a conclusion that Pacific app is like not safe for the young people. Is it purely based on mental health and addiction, or is there a different factor? Does social media have a different impact on our brain to adults My parents say I'm addicted, Will I go through withdrawal Doid the scientists think that children's brains are developing differently to how their parents' brains did when they were kids? Do you think the way that will interact with one another changed Amy and Kat, welcome. Amy, there's lots of fascinating questions asked there by those students, but I think the biggest question of all is what's the evidence so far I think there's different ways to look at evidence in this space. Naturally, technology is really fast moving and companies don't provide a all of access to data or resources to test the impacts and the safety. So I think evidence is evolving, We have pretty good evidence that it has harmed individual young people, right? Coroners have found that some young people have died partially because of what content they've been exposed to on social media. Individual harm of severe harm is enough cause major policy change in those studies with young adults, we do see that there is much better evidence of reduction in social media use leading to improvements in wellbeing Wh abstinence from social media completely, the evidence is still much more mixed and difficult to interpret. What about wider population correlations? How much of this is I know scientists are incredibly wary about things like causality, but how much of this is simply looking at two graphs? One of which is social media use and the other is teen mental health? Yes, you're right, right? We've seen teen mental health decline from the time that social media or smartphones were released. and actually there was a similar time there was an economic crash, there's been a lot of other changes going on and there's been a lot of debate what they say or whatnot. We wouldn't expect evidence to be great just given the lack of the raw materials for proper science to be done. And so I think it's critical for us not to just talk about is there evidence or is there not? But what are the risks of different options? What's the risk of saying there potentially isn't an impact when in five years, we find out that there has been And I think that helps policymakers make those judgments. I think in this space, they'll always have to be gambled. pololicy gambles based on the best evidence available. And you're currently involved in efforts to get evidence. Talk us through what you're doing. We are running a world first trial up in Bradford where thousands of school children will be randomized in whole school year groups to either just use social media as they did before or to be given a curfew on their social media use and a reduction. And that will happen in the fall and the plan was that in early twenty twenty seven, we could then give some of that really important causal evidence or leading onto that causal evidence around whether social media might be impacting young people and how that intervention works. So a randomized controlled trial, that's rather than making a big population difference where you don't know whether team mental health has changed because of the financial crash in two thousand eight or the iPhone in two thousand eight you can say it is a thing, willill you be able to get an answer before they're all banned from using social media? So we will do it all in the autumn and be done by December. Kat, you're involved in you're welcome and you're involved in trying to get evidence. What is your take on this? What is Welcome hoping to now get from the UK ban and from what's going on now Yeah, so welcom' one of the largest global funders of mental health research. And the digital world is a massive part of everyday life for young people So we actually got involved in this space, funding the trial that Amy talked about because we really wanted to move the evidence beyond the associational evidence that you've discussed it's very hard to know whether social media is actually causal for mental health problems and really try and unpick whether social media restriction is something that could actually help young people. And how do you do that? Because we're once again in a situation where something's happening in the whole population all at once, along with a load of other things that are unrelated that are happening in the population all at once. So how do you possibly get data where you can tell the rest of the world, Yes, this works or No, it turns out it makes people's lives even worse. Yeah, it's a really good question. and I think that's why there's still tremendous value in studies like the IRL trial. which are the gold standard randomized control trial and have a control condition, so you can have that comparison. Once a van comes in for the UK as a whole You won't be able to have that nice controlled design. but there are other methods that you can use to try and infer what the effects of the ban might be so for example We might fund cohort studies which follow up thousands of young people over time. Many of them have got decades worth of data from young people often from birth And we would fund them to collect baseline mental health measures before the ban is implemented and then follow up young people over time And then it's not ideal for causality because there isn't a control condition, but you can use modelling techniques to control for variables and look at pathways to understand not only if the ban is effective, but If so why that might be So you can look at variables that might be of interest to mental health like sleep. So for example, is the ban on social media meaning that young people are sleeping more? Are they interacting with their friends more? So for us, it's really important that we understand is the ban effective? Is it acting as it's intended? Are there in fact any unintended consequences that are harmful to young people. but also if it is effective, what are the mechanisms or pathways by which it might be having an effect What are the sort of outcome variables that we'd like to look at? Because mental health isn't one thing and it might not be the same thing for different groups of people. Well I think the first thing is to think about what outcome variables really matter And I think here the ban is really marketed in a way as a silver bullet for a lot of different issues. So we're talking about mental health and wellbeing, but there's arguments around educational outcomes, bullying health, eye health, sleep. And so the first bit is actually thinking about what does success look like? I don't know Kat if you want to discuss different populations as well. Yeah, it's really important to understand not just the average effect, but as you say Tom, there might be minorities where there's a particular impact. So we're interested in studying For example, young people who are already experiencing mental health problems, is the effect different for those young people? Young people living in urban versus rural environments, For example, those in isolated communities might feel particularly affected by the ban. We're interested in minoritized groups such as those from LGBTQ plus populations because we know that young people from those groups might use social media for social support in particular. And so I think as well as the quantitative measures that we might look at, those questionnaire measures of mental health or measures of what people are doing on their screens at any given moment, we also want to look at the qualitative data from young people and find out how the ban is affecting them in their own words A beginning of this, we heard a whole bunch of questions from teenagers who are about to be banned from social media. I think we've answered a few of them, but by no means all of them or come even close. Do you want the same answers? Do you have the same answers? What did you make of them? I mean, I think a theme that came through really clearly if social media is being taken away, how will that affect them? There was a really interesting question about Is it really addictive? Will I experience withdrawal symptoms? I think there's also a really interesting piece to be done around what young people will be doing with their time instead. I think from what I've heard, the perception that with a social media ban, something might be being taken away, what is going to be put in its place. Amy talked earlier about a cultural shift, what does that actually mean? So yeah, a lot of questions. alsoso, why the age of sixteen has been chosen. I think that's a really interesting question Any age that would be chosen would be to some extent arbitrary In the UK it's sixteen, I believe in France, it's fifteen. So to some extent, the choice is a social one, as much as a scientific one I think to the extent that we do have evidence, we know that the adolescent brain is still developing Th these are formative years young people are particularly susceptible to mental health problems with half of mental health problems emerging by the age of eighteen. And so I imagine that this underpins some of the decision to choose sixteen, but I would by no means say that this is one hundred percent a scientific decision. It's a social one as well. Thanks, both. Thank you Kit, you know about stats, you also, I think, know about having children. Where do you stand on the stats of mental health and social media and everything else? I mean, personally, I struggle with social media a bit myself. I find seeing the best bits of other people's lives, you know as if that it was always their life is a bit difficult. I have a thirteen year old and an eleven year old and we've made a conscious choice Not to let them use social media. And I think maybe the social media ban will make that choice easier for other parents. You're listening to Inside Science from the BBC World Service Tell us what science you think we should be investigating email address is inside science at bbc. co Close your eyes. Summer smells like sunshine, fresh citrus, and saltty air. What if your living room could feel just like that? With Pura's new summer collection, it can restestore your sense of wellbe with fragrances designed to move with your day. From bright, energizing mornings to soft, relaxing evenings, make the invisible unforgettable this season Visit purea. com to find your new favorite summer scent When it comes to looking your best, Beachbum Tanning does it better. Beachbum delivers advanced sun and spray tanning, luxury skincare, and an elevated salon experience designed around you. It's why so many guests trust Beachbum for flawless color and real confidence. And now Beachbum is expanding wellness services to many locations, with red light therapy and infrared sauna, with more on the way. Rcharge your body, refresh your skin, reset your day. Beeachbum isn't just tanning. It's full spectrum wellness. Visit beachbum dot com to find a location near you UK Now over to Roland Pees, who has been investigating the latest international science B brief leap In time Today's story concerns what could be the first evidence of a disease outbreak yet found in prehistory. It concerns plague Yurinia pestes made famous by the medieval Black Death, but yet spreading five and hal thousand years ago in mobile huns and gatherer groups, not in settled urban or farming populations The story starts, Archaeog geneneticist Rory McLleod told me puzzling grazves near Lake Baial in easastern Siberia. There was one particular site which had very high excess of dead children And as well as that, the radiocarbon datating showed that These deaths took place over an extremely short time period. Pretty much everybody seems to have died at the same time But there was no clear explanation for what could have caused this mass mortality. There was no evidence for violence, there was no evidence for skeletal trauma. So for a long time they really couldn't explain what had happened we're very interested in the possibility that ancient DNA might be able to shed some light on this. And so the idea there being that by looking at the DNA, you would find not just who they were. But you'd find if they'd been suffering any infection. Yeah, I don't think infectious disease was really something that immediately occurred to us. So ancient DNA would potentially provide more insight into the identity of these individuals in terms of their ancestries, if these were an external group from the rest of the Beal hunter gatherers, for instance. So we studied the remains of forty six individuals in these for these bicoppic extremes And we do see a lot of biological related this And so you see lots of cases of siblings being buried together in shared graves, parents being buried close by their offspring at the large site of Ishida one where we have this single dramatic catastrophic Mass deeath event You see quite a lot of shared graves in particular. But in the event you found out a lot more than that. Yes, so what was kind of extraordinary was that we found that really plague seems to have been the cause of death for everybody here. And the reason why we didn't expect this result at all is because this flies against the expectations in our understanding of disease epidemiology for really devastating impacts of disease outbreaks upon communities I have spoken to people in the past that Pague has been known mayaybe not this far back But you know, thousands of years ago, so Plague has been recorded from around the world Yes, so we are well aware of plague strains having occurred in late Neolithic Europe. Until around about five thousand two hundred, five thousand three hundred years ago are the oldest examples from Europe. There's one individual that's ded between five thousand three hundred and five thousand years ago. This is the first instance that we have of really prehistoric plague from outside of Europe, and it's also by far the oldest And what's kind of really exciting about these outbreaks are that they show definitively that these plague strains were deadly. And where do you think the plague came from? I have heard other work that says that marmoths may have been the original source. Yeah, so marmots have kind of long been guessed to be potentially the natural reservoir of plague. And marmots are the kind of primary risk animal for plague today if you're in Central Asia. You see lots of news reports, unfortunately of horse hererites, for instance, eating a marmot that's undercooked, for instance, unfortunately infected and dying of plague. And wouldould it just be repeated infection or do you think that this was human to human transmission? So we think that the evidence that we have is more consistent with human to human transmission. We see multiple instances of the same strains of plague turning up at contemporaneous cemety sites quite far apart. So the first outbreak really concentrates on these two cemetery sites of Uustida and Shumelia And those are thirty seven kilometers apart along the River Angora But we see some distant biological relationships on the level of cousins or slightly more than cousins between the individuals there Given that they're at such different cemetery sites, we interpret archaeologically that these are different groups of people, but people that would very very likely have come in contact at different times. I mean to me, this is very interesting know because we know of the Back death and the great outbreak, shall we say of the medieval times Do you know how if this is very different as a bacterium five thousand years ago than it was a thousand years ago. This is really, really different. So what's quite exciting is that The plague strains that we have from Lake Bial are quite close to the kind of root of the evolutionary tree of all the plague So these are really close to being an ancestor to the entire diversity of plague Ysinia pestis, including those black death plague strains You can kind of think of plague strains as somewhat like cocktails, cocktails of jeans. And some of these genes also interact to make the cocktail significantly more potent So in the case of the plague ext drains that led to the first pandemic and the Back death A gene emerges around about three thousand eight hundred years ago allows the plague infection to take on this form of bubonic plague, being transmitted by flea bites and resulting in these horrible bubes have led to so much suffering and are classically depicted with the black death. So that's a later adaptation. That's a much later adaptation, exactly. So what's happening with these early plague strains in prehistory, well over a thousand years before that emerges, you have a different set of ingredients resulting in potency. And unfortunately, this different set of ingredients includes a gene that we don't see later on PM which is the super antigen, which is unfortunately especially deadly for children. The one furious about is that you seem to have been so lucky that these individuals were preserved to record this unique insight into the early evolution of play Yeah, no, it's really extraordinary that they buried the dead in this particular way and it kind of really makes you wonder What would have happened when there weren't people who were survivors who were able to bury the dead like this I mean, are we missing a huge amount of other evidence of outbreaks of plague amongst Hunter gathr is simply because everybody died or there wasn't anybody fit or strong enough or in that position to be able to bury the dead. And it's also just so close in the evolution of plague to when we think plague first emerges as a separate species. Rory McLeod of Oxford University on the Beginning of epidemics, described this week in nature Thanks for Roland Now C He's up and running in. It's been brilliantly. It's been absolutely sensational in the second half. What an unbelievable moment. B is shiny. G is Hatrick. What a story. What a story Ah, the beautiful game. The game of two halves The game with the ball and the people and the one man who's allowed to use his hands. feear not Over the next month, it is our firm pledge not to tell you anything useful or insightful about the World Cup. Instead, we are interested in the useful and insightful things the World Cup can tell us about us. When football fans see crowds gathering to watch the game, they see a sublime gladiatorial contest of human excellence or some such Scientists, they see an opportunity to study crowds and psychology at scale. So over the next month, we're going to use that to make our own World Cup squad, a World Cup squad of science I am to be it's Thomas Toukell. The science mananager. Joining me is Kit Yates you can think of him as like the guy who cuts up the oranges, but nevertheless, I've allowed you to do your own talent spotting and I think you've found a promising striker in the big leagues of the American Economic Review Yeah, so I want to tell you about penalty shootouts and the paper is called Testing Mixed Strategy Equilibria when players are heterogeneous, the case of penalty kicks in soccer So as a mathematician What I like about penalties is that all the complexities of the game are stripped away. So penalty taking is really pure, simple sort of game between the taker and the keeper. And so from a scientific point of view, it makes it an ideal part of the game to analyze Unfortunately, there is this branch of mathematics known as game theory, which is dedicated to understanding these simple sorts of competitions. So game theorists would call taking a penalty of zero sum game for the taker to score the cper has to concede and for the cbe to come out on top They have to either save the shot or allow the takeer missing. But either way, there's only one winner so it's a zero sum game So the game theorist in this paper, they use a simple model of penalty taking where they just think about three options for the placement of the penalty to the left, to the right or down the middle and the same three options that the keeper can use to attempt to save. And they also assume that both the taker and the keeper make up their minds about what their strategy is going to be well in advance, whichich isn't a terrible assumption, it's not always true. But then they can say something about penalty taking strategy. And what they suggest is what game theory predicts is that no single strategy like always aim to the right is going to be optimal because it's predictable We see keepers nowadays with notes on their water bottles about which way the takeaker is likely to hit their penalty. So to avoid that predictability, the optimal strategy to use game theoretically at least is what they would call a mix strategy where the player deliberately introduces unpredictability or randomness into their decision making to prevent the opponents from exploiting patterns. So this would mean Theyre taking at different sides, left, right, middle each time. And so the study that I'm presenting to you is one that actually looked at the data in two of Europe's top leagues in France and Italy and found that rather than favouring one side of the goal, penalty takers did indeed choose randomly between kicking to the left to the right or down the middle consistent with this game theoretic mixed strategy. And I think this is in contrast to less good sort of more amateur players where actually they are predictable. So it's like they sort of learnt this Rndomness strategy. Yeah, exactly. When you hear the top penalty takers talk about take their penalties, they sometimes say look, I don't even know which way I'm going to hit it before I walk up to the ball. They are genuinely using a mixed strategy, whereas I don't think that's the case that l way down the leagues. Okay, so we promise listeners who might not be that into football that this is going to be about you know more than footballers are jumping off point. this has wider societal implications, doesn't it Yeah so these mixed strategies come up all over the place, but in particularly in the context of international diplomacy, where sticking to a pure strategy, having a preordained response to any given situation that might come up might reduce the ability of a negotiator to bluff or to bluster or to manipulate their opponent And conversely on the flip side, when you're negotiating with someone who might have their finger on the nuclear button one minute or advocate for total disarmament the next An opponent might find themselves making more concessions and famously, President Nixon used what was called Madman Theory in the nineteen sixties and seventies. and the aim, as the name of Madman theory suggests was to convince your opponent that you were a little bit unhinged. So Nixon, he reasoned that if his opponents judged him to be an irrational actor They wouldn't be able to predict his plays and they would then have to make more concessions to avoid the risk of accidentally triggering him into some sort of retaliatory action So in this sense, the penalty shootout is a sort of crucible kind of Paradigm for how everything else is going on in. It's a tightly controlled model of where these negotiations, this game theory can come up with lots of good data. I've been talent spotting as well. The study I want to look at this particular episode, the first study I brings to squad, published in the BMJ in twenty twenty and it's titled Association of High Profile foootball matches in Europe with and this is why I love it, with traffic accidents in Asia.