The Five Rules of Risk

Get your custom domain or email for 10% off at
Listen to Extremities at
Buy a Wendover Productions t-shirt:
Subscribe to Half as Interesting (The other channel from Wendover Productions):
Instagram: sam.from.wendover
Sponsorship Enquiries:
Other emails:
Animation by Josh Sherrington
Sound by Graham Haerther (
Thumbnail by Simon Buckmaster
Music by
Select footage courtesy the AP Archive

Tải video

lượt xem


  • We got some math wrong. The annual fatality odds for licensed drivers in the US is actually 1 in 6,000 which translates to lifetime odds of about 1 in 75.

    Wendover ProductionsWendover Productions9 ngày trước
    • That is probably big enough, that you should fix the video or at least do an overlay and note in the description.

      Peter VPeter V11 giờ trước
    • Yeah you can't leave this video up with that massive of an error bro. You have to change it. Great video though. Really important rn.

      SamtemberSamtemberNgày trước
    • So are we overestimating or underestimating the coronavirus risk?

      Jeff TimmerbergJeff Timmerberg2 ngày trước
    • @panzerveps not true...

      CptKirk FPVCptKirk FPV2 ngày trước
    • I do have to wonder what happens if one omits behaviour like driving drunk, which would increase risk dramatically. Do the odds of death/injury then fall, and by how much?

      Stephan BrunStephan Brun4 ngày trước
  • why does he sound and talk like Half As Interesting?

    homeihomei12 giờ trước
  • "one death is a tragedy, a million deaths is a statistic" Joseph Stalin

    James HarmonJames Harmon13 giờ trước
  • <a href="#" class="seekto" data-time="610">10:10</a> reminds me of a quote allegedly by stalin: "One death is a tragedy; a million deaths a statistic."

    DasNotizPapierDasNotizPapier18 giờ trước
  • I walk outside because life wouldn't be life if I didn't as to say I wouldn't call it life if there was no walking outside

    Iqbal HassanIqbal Hassan21 giờ trước
  • PLEASE LEARN ABOUT DE-ESSING and improve your videos

    Emil PEmil P22 giờ trước
  • Isn’t it great that he uses a cyber truck as his example of a car🤣🤣

    Six Dawg 18Six Dawg 1822 giờ trước
  • <a href="#" class="seekto" data-time="71">1:11</a> But there's a VERY low chance of being hit by a Cybertruck, considering that it hasn't been released yet.

    bibasik7bibasik722 giờ trước
  • if i were to mountain bike id be much more likely to die than if i were to drive. you frame the statistic as if driving was more dangerous but: the average driver drives many many more miles than the average mountain biker. and the average mountain biker is much more skilled at mountain biking than the average driver. and if everyone was to start mountain biking the fatality rate of mountain biking would increase much more sharply than for driving if everyone started driving

    Collin yanCollin yanNgày trước
  • <a href="#" class="seekto" data-time="170">2:50</a> you forgot to consider the fact that in any given area at a given moment, the factor for the car being in a more dangerous scenario is higher simply because there are more cars in a given area moving at high speeds than there are bikes. When bikes match the number of car population, then the numbers would relatively equal. For ever 10 car drivers, it's reasonable to assume that at least 7 of them drive with no fuss, just going from point A to point B. The 3 of those 10 however, could have possibly been the ones who contributed to car crashes as they perhaps were those to either drive wrecklessly or drunk drive. The point is, we also have to evaluate these two vehicle's usage and the nature in which they are more likely to be used. Its simply evident in your video examples wherein cars were simply driving about while the mountain bikers will tend to flip around and do "dangerous" leaps onto rocks. I can assure you, of a comparison between 1000 regular cars, going about doing their city driving, and a group of 50 kids intentionally trying crazy stunts on a mountain bike, there will be more injuries with that group of kids. Again, it doesn't have much to do with the "vehicle" perse, it's merely a factor of sheer number of them and the ways in which they are being used, that increases the chance of unwanted combinations of possibilities (crashing, falling off into a cliff, etc.). Now take that previous example of 50 stunt-driven kids and push them to 1000 to equal those of cars. Now you tell me if 1000 stunt-daring kids are less likely to be injured than 1000 regular-driving parents or perhaps elderly people simply going to the grocery.

    Fallen PastabeanFallen PastabeanNgày trước
  • My takeaway from this video was about masks - in the end he shows people walking with masks and a few without. The people who choose not to wear them think they are only choosing a risk for themselves when in fact, because covid-19 can be without symptoms for days, they are choosing the risk of others too.

    M GrohM GrohNgày trước
  • Now I dont want to walk outside

    Westjet 003Westjet 003Ngày trước
  • mountain biking has a lesser risk than driving a car because although the number of people mountain biking is high. it is still significantly lower than the number of people driving a car.

    fin thehumanfin thehumanNgày trước
  • I knew this video would get to that conclusion.

    sor3999sor3999Ngày trước
    • Interesting topic. But very subjective as with anything psychological

      abbsnn coseabbsnn coseNgày trước
  • I.e. COVID-19 is NOT risky. The lockdown will eventually kill more people and break the lives of billions.

    Gerbert HellerGerbert HellerNgày trước
    • Covid scamdemic brought me here. Oh lets kill Fauci and Gates. Seriously.

      abbsnn coseabbsnn coseNgày trước
  • While I agree with the general concept and you later in the video do say that this list is not complete, I have one question regarding your first point that is decoupled from your argument: Do you think the low risk of mountainbiking could be correlated to the fact that those who do this activity are generally skilled at it, in turn reducing the average risk? Now, while I don't know if that is true or not, I assume that, also due to the perceived risk of mountainbiking, only a few people try it. Of these few people some might get injuries or simply dislike the activity and stop after a couple of times, elaving only those skilled at the activity for a vast number of data points that contribute to this statistic. Now thinking about it, maybe the statistic you used as a example accounts for this behavior, but my pessimistic mind kind of doubts that. Not to say that you misinform anyone with this, just an idea to think about. Really love this video btw!

    ShaldisShaldisNgày trước
  • I try to avoid driving due to risk. Yet I love mountain biking.

    nick parrynick parry2 ngày trước
  • Please don't become CGPGrey and start making philosophy videos... I just want planes, trains, logistics, and science.

    InfiNorthInfiNorth2 ngày trước
  • The graph shown at around 10 minutes in is wrong. It is the other way round. Also, human brains are extremely pragmatic, so telling us not to trust ourselves is not very helpful.

    BoraCM 39BoraCM 392 ngày trước
    • @nieooj gotoy What on Earth are you on about?

      BoraCM 39BoraCM 39Ngày trước
    • Ok but if you never walk outside you will die from no sunlight and no excersize so not doing that is also a risk

      nieooj gotoynieooj gotoy2 ngày trước
  • Greetings from the town of Wendover, in Buckinghamshire, England 🤚

    Will BanisterWill Banister2 ngày trước
  • Totally expecting never go for a war in Asia... but this will suffice :P

    Kevin NeilsonKevin Neilson2 ngày trước
  • Rule #6: Many risks follow the power law distribution or are chaotic but they are reported using the normal distribution. For example, how many people who died in a war (pure chaos) will be reported along side how many people died from coconuts falling on their heads (normal distribution). The former may increase 1000x in a single year, while the latter will rarely deviate more than 1-2 standard deviations between years. This is often used to report that chaotic things are as risky or less risky than common everyday sources of death, and is one of the most common forms of fraud.

    Bryan CroteauBryan Croteau2 ngày trước
  • "Why do you ever walk outside" Corona:

    YINong XuYINong Xu2 ngày trước
  • final assertion that evokes a call to consciousness in May 2020.

    afutla qianafutla qian2 ngày trước
  • Seems to be a mistake in numbers. "1 in 600" dies while driving is way off (277mil / 600 = 378k. Do you have 378k deaths on roads annually?)

    Kromster80Kromster802 ngày trước
    • This assumes risk is perceived as absolute. I don’t mountain bike because I think I will likely die, but because I don’t want to injure myself.

      afutla qianafutla qian2 ngày trước
  • a lot of people seem to still be irrationally against nuclear power, though

    ThoperSoughtThoperSought2 ngày trước
  • Fun fact: the military is one of the safest environments to be in the developed world.

    Tal SheynkmanTal Sheynkman2 ngày trước
  • Skip to <a href="#" class="seekto" data-time="222">3:42</a> to start the video.

    A GamerA Gamer2 ngày trước
  • Human perception and opinion is more than purely comparing risk vs risk. For instance, driving a car brings loads of value to the family owning the car in terms of convenience and time saved. Granted people don't normally think of the risk on a daily basis, but I still think the comparison to mountain biking in school is flawed. Mountain biking doesn't add (in parents' view) much more value than other PE activities, so it's not worth the added risk. Compare driving vs taking the bus though and you have loads of added value. So it's not just about comparing risk vs risk, it HAS to be concidered together with the VALUE of taking the risk. Any thoughts?

    Eric JohnssonEric Johnsson2 ngày trước
  • Covid scamdemic brought me here. Oh lets kill Fauci and Gates. Seriously.

    Tomasz KliszTomasz Klisz2 ngày trước
  • Interesting topic. But very subjective as with anything psychological

    Ignacio IruritaIgnacio Irurita3 ngày trước
  • There are a few issues I had with the assumptions made in this video, but the one I disagreed with the most is this: "The value of your first dollar is the same as your millionth" Well, the first dollar I earn goes towards lunch. The millionth gets me unnecessary luxury. I don't think those two are equivalent in any way.

    GyroCannonGyroCannon3 ngày trước
  • <a href="#" class="seekto" data-time="166">2:46</a> that's a Tesla Cybertruck, I see what you did there 😏

    Abhiyan BhandariAbhiyan Bhandari3 ngày trước
  • I think the intro to the video took a weird perspective. For the sidewalk analogy there was no alternative. The risk that came to mind for me was not whether or not to use the side walk, it was should I use the road or sidewalk to walk on. Obviously the risk of walking on the road is significantly higher than the sidewalk so I'll choose the sidewalk. For the mountain biking one you weighed the risk of death from driving or biking but not injury. I don't know for sure but I'd be willing to bet that the statistics for biking injuries is much higher. Although it would be a tough stat to track cause not everyone is going to report their biking mishaps.

    RNG-esusRNG-esus3 ngày trước
    • Love how he used a Tesla truck as the vehicle representation👍🏼👍🏼 ironically those vehicles will probably change the death rate statistics

      nieooj gotoynieooj gotoy3 ngày trước
  • I think this might be a bad video?! There certainly is fact, value and knowledge... So the overall content should be good, but somehow the video is bad. Like it's not articulate enough to it's point and it brushes over deeper reasoning without consideration. Point: video ought be good, objective and nuanced, but it isn't. It's choppy and open to misinterpretation, which is also risky. P. S. Long time fan, love your work. Some of your work is truly great and informative but this vid falls pretty short. Thus my (hopefully constructive) criticism.

    Fizzy ZailaniFizzy Zailani3 ngày trước
    • This video is riddled with logical fallacies and I've only been watching for 3 minutes

      nieooj gotoynieooj gotoy3 ngày trước
  • Ok but if you never walk outside you will die from no sunlight and no excersize so not doing that is also a risk

    Chongo OkChongo Ok3 ngày trước
  • Okay the fatality rate of driving is higher than mountain biking, but what about injury rate? Possibly parents were more worried about injuries rather than straight up death

    Andrea FogliaAndrea Foglia3 ngày trước
  • Your maths on vehicle deaths - 1 in 600 per year - is laughably far off. It's not even close, your LIFETIME risk of dying in a car is still orders of magnitude less than that.

    Chris DomsChris Doms3 ngày trước
  • When comparing life and risk, there is a catch 22 - because when I step out of the door I risk my life, but if I don't, I don't live it.

    jayjay.24jayjay.243 ngày trước
  • <a href="#" class="seekto" data-time="161">2:41</a> Now we know where Elon Musk got his concept for his Tesla rover.

    Rage KnightRage Knight3 ngày trước
  • This assumes risk is perceived as absolute. I don’t mountain bike because I think I will likely die, but because I don’t want to injure myself.

    Casey BernerCasey Berner3 ngày trước
  • came for board game, got existentialism.

    spaghetti is deadispaghetti is deadi4 ngày trước
  • I hope i don't' die after watching this video this dude made pandemic video before

    Ronak NikamRonak Nikam4 ngày trước
  • <a href="#" class="seekto" data-time="480">8:00</a> railroads is misspelled. "Rairoads"

    Brandon WatersBrandon Waters4 ngày trước
  • "If you kill one, it is a tragedy. If you kill ten million, it is a statistic."

    LexieAssassinLexieAssassin4 ngày trước
  • "Why do you walk outside?" I don't. I'm in social quarantine. Checkmate buddy.

    kolim jonekolim jone4 ngày trước
  • You cant compare driving to mountain biking The number of drivers are much higher than bikers

    قيس الجدياني | QGamer_Arقيس الجدياني | QGamer_Ar4 ngày trước
    • more exposure to cows than sharks for a fatality to even occur; I don't think you'd say based on that alone that it's riskier to work with cows than it would be to swim with shark

      kolim jonekolim jone4 ngày trước
  • Eu fiz uma legenda em português-brasileiro... se alguém puder me ajudar a deixar disponível já está na área de legendas do vídeo...

    caquitowscaquitows4 ngày trước
  • I think some risk numbers are just "correlation not causation" figures.

    Sam GarciaSam Garcia4 ngày trước
  • well consdering humans will always have some degree of error in our risk management system, our leaders will always over shoot or undershoot the risks of some things.

    Seth ApexSeth Apex4 ngày trước
  • This video is riddled with logical fallacies and I've only been watching for 3 minutes

    S EdwardsS Edwards4 ngày trước
  • Love how he used a Tesla truck as the vehicle representation👍🏼👍🏼 ironically those vehicles will probably change the death rate statistics

    Michael HarrisonMichael Harrison4 ngày trước
  • It's unbelievable most people can't understand such simple things.

    ХОРОШОХОРОШО4 ngày trước
  • Waited for risks of climate change as an example

    CaptiveCaptive4 ngày trước
  • <a href="#" class="seekto" data-time="134">2:14</a> wasatch crest!!

    stinksmccstinksmcc4 ngày trước
  • Flawed? Really? How about the fact that mountain biking has no useful utility in our lives; while driving is necessary to accomplish nearly everything.... Most of us have jobs away from home. Simple as...

    Joseph WaddellJoseph Waddell4 ngày trước
  • Walking outside vs staying at home isn't zero-sum - a huge number of accidents happen in the home. The risk of not getting exercise is very high, etc.

    psammiadpsammiad4 ngày trước
  • I went driving along a motorway once. It was more scary to me than a roller coaster as it was clear I was taking my life into my hands. I wish the government would make proper public transport. I don't intend to ever drive again.

    JackamomoJackamomo4 ngày trước
  • Did you just draw a image of tesla cyber truck

    Sanajit DasSanajit Das4 ngày trước
  • I think how much I value my life is fairly irrelevant for a risk calculation. When you weigh the value of the reward against the value of your life, how do you get the value of that reward? I think typically it could be understood in terms of mitigating some other risk. So it's not a series of isolated risk vs reward calculations, it's a single infinite web of risk vs risk vs risk. Driving is fine, because not getting to work or not getting to the grocery store is also pretty risky. Mountain biking after school isn't, because the risk from lack of exercise and entertainment isn't as severe. Or because there are less risky alternatives that solve the problem just as well. I also wonder if the thing about the 100th life having less value than the first could just be the effect of the other principles. I think your example was a terrorist attack which kills a few hundred being about as significant as a natural disaster that kills thousands...but while natural disasters aren't often predictable, they are known. They tend to be regional too which can make then seem more voluntary. You know what disasters might happen in your area, you know how to prepare for them, you move if you really don't want to take that risk. So that's a semi-known, voluntary risk against a generally unknown and "mandatory" risk. Not quite the same thing. Could even be as simple as more people dying means more people discussing it, which means more explanations for it are given making it seem more known or predictable too. Finally...humans wouldn't have evolved if we didn't also care about the survival of our family/tribe/species...which explains the "illogical" behavior around mandatory/unknown/high exposure risks. We're more willing to accept that we might die as long as it doesn't wipe out all of us at once.

    Urza9814Urza98144 ngày trước
  • I don't understand the statistics claims here. <a href="#" class="seekto" data-time="115">1:55</a>,000 change of dying if walking outside? reference 1 doesn't have that value. what it says is that there is a pedestrian death once every 88 mins in the US. So how many people walk outside every 88 mins? how many of those seconds of those walks were people not killed at? clearly dividing the number of pedestrian deaths per minute divided by the amount of persons minutes walked in the US every year, will show that the odds of dying by walking outside is extremely low. so maybe some other calculation? maybe the amount of pedestrian deaths in a given year divided by total amount of deaths in the US in a given year? that too doesn't reveal the correct risk, since it doesn't consider all the people in the US that DIDN'T DIE in that year. So the correct calculation should be the probability of dying in a given year * the probability of dying as a pedestrian hit by a car. Clearly this is much much lower than <a href="#" class="seekto" data-time="115">1:55</a>,000 claimed in the video this is the same for all other odds/risks presented. i don't think anyone would walk outside if the odds were so high. they simply aren't and that's why everyone does it. and in regards to mountain biking - the magnitude of the 2 death rates is simply nowhere near comparable. without taking into consider the trillions of driving hours of all americans a year is a warping of reality. what we all know intuitively is that if only 10 people do something and 1 person die due to a crash (for example), this activity is extremely more riskier than a different activity that has trillions of people doing it, but of all deaths during that activity 100% were caused by a crash (for example). the shear rarity of the death event itself is being ignored in the considerations presented in this video and skew what we all intuitively consider. that said, once the math is corrected, then indeed we all consider the value of risk vs. reward and make decisions accordingly

    Ariel KassAriel Kass4 ngày trước
  • "infinity multiplied by anything is infinity" zero enters the chat

    bowen voowybowen voowy4 ngày trước
  • if 2/3 of people drive of course the numbers go up, the number of people that mountainbike likely leads to less risk because of selection bias, the people that do are less likely than the average person to suffer an accident in the situation; if 2/3 of people did mountain bike the numbers would go up, maybe not to the same number but yeah. In the US 1 person dies by a shark every 2 years, cows kill 20 people per year; Does that mean cows are more dangerous than sharks? No, it probably indicates that there's considerably more exposure to cows than sharks for a fatality to even occur; I don't think you'd say based on that alone that it's riskier to work with cows than it would be to swim with sharks.

    Vlad Pintea-GärtnerVlad Pintea-Gärtner4 ngày trước
    • Wealth is relative. The comparison with the dollar doesn’t hold up because in the case of a dollar, the perceived value loss at higher quantities is logically true as well. 10$ is

      bowen voowybowen voowy4 ngày trước
  • The animation in the video was really distracting

    kityew23kityew234 ngày trước
  • All of us Christians watching this: NBD

    LoungeSpecialistLoungeSpecialist5 ngày trước
  • Cough cough, cdc, cough, cough

    bob smithbob smith5 ngày trước
  • Is there not some Irony in this video ending with the words "or someone else will take it" re the domains?

    Julian Manieson JrJulian Manieson Jr5 ngày trước
  • We drive motorized vehicles because of the convenience, utility, necessity and pleasure they provide therefore we accept a higher risk (though we rarely think about it). We don't ride mountain bikes for the same reasons we drive therefore we accept a lower risk. If the very last image was some sort of 'dig' at the need for mass self-isolation, when was the last time so many did so much for so many? Thank you for not coviding.

    Marc GoodmanMarc Goodman5 ngày trước
  • not walk outside is guaranteed death by hunger.

    jun skijun ski5 ngày trước
  • Ha! I love it how you used a Cybertruck silhouette to represent a car :D

    Grzegorz DykGrzegorz Dyk5 ngày trước
  • <a href="#" class="seekto" data-time="163">2:43</a> this is true and i understand your point but comapred to other sports like soccer or basket ball, Mountain bikini is more dangeroise

    Axel PeterssonAxel Petersson5 ngày trước
  • @wendoverproductions The numbering on the 'Activity or Technology' is wrong @ <a href="#" class="seekto" data-time="497">8:17</a>, on the students side there's two #6's and #16 is missing

    Dean9419Dean94195 ngày trước
  • Comparing the risk of driving a car and the risk of mountain biking solely on the death toll is very skewed logic. Parents being against mountainbiking as a school activity is probably not because they are fearing their kids dying, but because of (severe) accidents, which especially untrained DO happen a lot.

    Jan PJan P5 ngày trước
  • This video: “Why do you walk outside?” Coronavirus: “Let me stop you right there.”

    WertsirWertsir5 ngày trước
  • The moving kaleidoscope background is really disorienting and makes it harder to read FYI. Good video otherwise

    Tom HammondTom Hammond5 ngày trước
  • "infinity multiplied by anything is infinity" zero enters the chat

    bodoti qwiubodoti qwiu5 ngày trước
  • I thought this was gonna be about a board game!

    Price ThomasPrice Thomas5 ngày trước
  • I have always been completely baffled by everyones blind acceptance of cars. Everybody just acts as though there is nothing that needs to be fixed and that it is a completely safe thing to do. Everyone deploys the them but not me defense. "Oh somebody else will get hurt, but I'm a good driver." When in reality, we are all just as likely to get hit by the drunk driver and killed, just as a few in my life have been.

    SJGSJG5 ngày trước
    • With this lockdown entering the third month, about 6,300 people who were supposed to die in car accidents during these two months, are not going to die. Maybe you’re one

      bodoti qwiubodoti qwiu5 ngày trước
  • Sadly this is why paintball isnt popular enough

    kelvink9999kelvink99995 ngày trước
  • Wealth is relative. The comparison with the dollar doesn’t hold up because in the case of a dollar, the perceived value loss at higher quantities is logically true as well. 10$ is as much money to someone who has 1000$, as 1000$ is to someone who 100,000$. It is us humans who assign value to the money, otherwise it would only be printed paper. Because that is the case, any perceived value equals the real value. If everybody was rich, nobody would be rich.

    Tristan MöllerTristan Möller5 ngày trước