Killed by a falling tree: what are the chances?

The Ranger worked as a Tree Officer for some months not too long ago, and got quite an insight into the arcane world of TPOs and irate householders. A bit different from being a real Ranger, where pretty much everyone is glad to see you. A tree officer often has the thankless task of trying to preserve a tree against the wishes of the tree’s owner, who wants to fell the thing to get a sea view, or build a house. For some reason they rarely come out with the truth and say they want to fell it. They usually proffer some excuse, prefaced with “I like trees but…”. Perhaps the most common of these is the constant refrain heard by tree officers “But it’s a dangerous tree!“. Often the complainant then over-eggs the pudding with references to their little, blond grandchildren, innocently gamboling underneath the looming tree, which has regularly been heard to creak and groan ominously, and even, believe it or not, sway in the wind! It’s hard not to get cynical about some of these requests. It would be a lot better if they just came out and said “I want to sell off half my garden to build a block of eight flats on it.” Then at least we’d know where we all stood. But, can they always be wrong? Just how dangerous are trees, generically?

Tree fallen onto bicycle

Indeed, some individual trees are dangerous and need work. But that does not mean they all are. Often the dangers of trees are considerably overestimated. Just what are the chances of a tree falling on you and killing you? Chris Hastie, arboriculturalist and webmaster of the The UK Tree Care Mailing List recently got fed up with the assumptions that are made about such things. He writes:

After the storms the other month I was phoned by a journalist who questioned me about various things, mostly to do with a very large horse chestnut by the side of a busy road which managed to blow over and do no harm to anything except a lamp post. Trying to explain the nuances of risk management to her and knowing everything I said was going to be massively dumbed down, I started to wish I had a few soundbites at my fingertips.

So Chris took the question at face value and worked out some statistics. He started by pointing out that the chances of being killed by a tree in a public space in the UK is about 1 in 20,000,000 (according to the UK Health & Safety Executive, “Management of the risk from falling trees – Internal guidance”). So, what about winning the lottery jackpot? Actually, Chris also demonstrated that rather than the 14 million to one which is usually quoted, the chances of winning it are actually better expressed as 1 in 268,920. This is because although the chances of winning with one ticket are indeed 1 in 13,983,816, accidental deaths are usually expressed as the chances of any incident happening to any person in one year. So, assuming a lottery player buys one ticket per week every week for a year, the odds are reduced to 1 in 268,920. Thus a regular lottery player is 75 times more likely to win the lottery jackpot than be killed by a tree in a public space. He goes on with some other sobering illustrations. The total number of accidental deaths in the UK number is over 12,000 per year. About 6 of these are due to trees. So you are 2000 times more likely to die from some other type of accident than by being hit by a falling tree. More specifically, 3,501 people were killed in road traffic accidents in the UK in 2005. So you are around 600 times more likely to be killed in a road accident than by a falling tree. The Ranger adds one of his own – the annual risk of being struck (and not necessarily killed) by lightning is 1 in 10,000,000. So you are more likely to be hit by a bolt of lightning than killed by a falling tree. That seems to put things into perspective. Anyone else want to have a try? Cite your sources if you do!

19 thoughts on “Killed by a falling tree: what are the chances?”

  1. I actually found the Ranger’s approach to human life quite disturbing, very disturbing in fact. Equating a person’s life who you don’t know, with family and friends that you do not know into the statistical aboration of chance whether that be 1 in 10 or 1 in 100 million is a clear example of just how out of touch you are with reality and how little value you place on human life as against a stupid tree. Does it really matter what the chances are? Does it really matter the chances of something else happening is less? If I am fearful thast a tree in my backyard or my neighbour’s backyard will fall on me, or my neighbour, isnt that the issue that should be addressed and not the well being of a tree? Trees can be replanted to replace a removed tree but me, my family, my freinds, my neighbours or even strangers can not. I am not a statistic that you can merely write off, I am real, I am a human being.This nonsense of tree preservation is exactly that nonsense. Remove the source of the problem and plant something else in its place. Its pretty simple but only if you value human life more highly than trees. I clearly do, you clearly don’t and you should be ashamed to justify tree preservation over and above that of human life. Trees are not part of human endeavour. It always amazes me that councild and governments will remove thousands of trees for a new road or housing development but when Mr and Mrs Joe Average want to remove a tree from their backyard all hell breaks loose….its the environment…its the this…its the that…..If its my property and its my tree it should be my decision to remove it, no matter why I want it removed. If you want balance then simply plant another tree in the local park or some other open space….but please dont tell me I must accept the risk of death simply on the basis that the statistical chance of me being killed is greater than winning the lottery…..

    1. This blog was brilliant! The negative poster above obviously has never had a neighbor who wanted to cut down seven 200 year old trees in no danger if dying just because they don’t like leaves or want sun for a mint patch. Buy, they claim they’re afraid that they will fall. Trees provide privacy, sound barriers, habitat, the shade underneath is specific to necessary bacteria and insects. You just can’t replace old-growth trees with saplings and call it even. I’m posting this information on my neighborhood blog today! Thank you.

  2. I was in a car over 30 year ago that was hit by a falling tree 4 people were killed and I survied I never buy a lottery ticket as I think it’s a waste of money I still love trees and all things nature theses things and not statices it’s just life enjoy it no matter what happens

  3. Paul Jenkins, chief executive of the mental health charity Rethink, claimed “you are more likely to be struck by lightning than killed by someone with schizophrenia”. (therefore more likely to be killed by a tree falling on you 🙂 (actually that is someone you don’t know)

    The number of people killed by lightning strikes is around three a year, according to the Royal Society for the Prevention of Accidents, though there are more strikes which cause only injury.

    This compares with an average of 32 homicides a year by schizophrenics, according to the Manchester University project.

  4. Last friday on my way to work, stopped at a traffic light. 12″ caliper Black cherry fell on my car. Missed hitting me on my body by about 5 feet. had 4″ of rain the previous night, and tree was on a steep hillside… I guess an check off my list, one way I’m NOT going to die. The tree had its chance and failed…

  5. I Lived On 100 Acres In Kentucky 4 Years Ago. While Hiking I heard A Creaking Noise And Started To Run. Not Fast Enough, An 8″ Oak Tree Fell On Me. Broke My Back, My Ribs, Lacerated My Liver, Broke My Left Foot And Crushed My Right Ankle. I Am Still Having Surgeries. So To Me The Odds Caught Up With Me.

  6. You state the chance of being killed by a tree.. This is for all the people all over the earth, who may or may not be standing, sitting, or living, under a tree for most of their life. What about people who live under the branches of a massive tree? The risk goes up surely? And what about if that tree is very old? Even higher risk. I’ve read that the risk of an old tree dropping a branch is 10% per year. And they’re a lot heavier than they look. You try sleeping under one in a storm and see how you feel!

  7. Dear Ranger,

    Is there any evidence that risks of personal or property injury from falling trees or branches have changed over the past few decades? Of course you would have to allow for population growth and development.

    But my observation is that trees are dying at a very rapidly accelerating rate, with a sharp increase in falling branches and entire trees. Trees of all ages, in various habitats.


    Gail in NJ, USA

  8. Yeah.. some creepy stuff huh? I can’t believe that the annual risk of death by lightning is 1 in 18,700,000. So you are more likely to be killed by a bolt of lightning than a falling tree. So that falling huge branch that almost crushed us is was 1 in 18,700,000 likely to happen?!!!! I should play the lottery!

  9. Listen to this shit, i went to a local state park in the US, Rhode Island and we were having a fire and just chillen there and we heard a small branch fall but we said fuck it and enyoed the fire for a few then we heard a loud ass crack as if it was lightning or something so we booked it and one of us got hit with a small branch that fell off about a 28ft branch from a huge tree the branch was huge that almost crushed us and our car the small branch that hit someone he just fell and got knocked out for a few seconds.. yup check the odds on that shit

    The Ranger responds: This was the tree you just lit a fire under, presumably? Smooth move. Goodness knows what made it drop a branch just at that moment.

  10. The odds of dying falling down stairs are about 1 in 1,000,000. The odds of dying from drowning in Australia are about 1 in 300,000. The odds of dying under a general anaesthetic are 1 in 10,000 (source Australian College of Anaesthetists and people choose to undergo elective surgery).

    The odds of dying from food poisoning and influenza are each far greater than the odds of dying from trees. In Australia influenza kills more than 1,500 people a year (about 1 in 14,000). In the Australia about 120 people a year die from food poisoning (Australian Academy of Science) which equates to about 1 in 175,000 … perhaps we should also stop eating and breathing when we cut down those dangerous trees

  11. A couple more points:
    (1) The chances of winning by buying 52 (differently-numbered) tickets in a single lottery are not the same as the chances of winning at least once in 52 successive attempts with a single ticket. The former’s value is 52/14,000,000 = 0.00000371429 (6sf), which is ever-so-slightly greater than the latter’s, 0.00000371428 (6sf). Hence you are 1.000002 (7sf) times as likely to win by buying 52 tickets in a single week as to win by buying one ticket a week for 52 weeks.

    (2) How long would it take to be virtually certain of winning the lottery at least once by buying a single ticket every week? A very long time indeed! If you did this for 1,000 years (i.e. ~52,000 weeks), your chances of winning would be 0.0037 (2sf), i.e. just under 4 in a thousand. If you did this for 100,000 years (~5,200,000 weeks), your chances of winning would be 0.31, i.e. just under one in three. Finally, if you did the lottery for one million years (~52,000,000 weeks), your chances of winning at least once would be 0.9756 (4sf), i.e. nearly 98%, which is getting close to certain. Of course, you’d have spent

  12. I remember very little of my 1st year probability studies (nearly 30 years ago), but I do remember this … and the probability of winning *at least* once in 52 successive lottery attempts is not the same as the probability of winning once in a single attempt. The best way of thinking about the former probability is to consider the opposite, i.e. the probability of winning precisely no times in 52 attempts.

    If the probability of winning on each occasion is 1 in 14 million, then the probability of not winning on each occasion is 13,999,999 in 14,000,000. Therefore the probability of failing to win on every occasion (out of 52) is (13,999,999/14,000,000)^52, i.e. 13,999,999 divided by 14,000,000 and raised to the power of 52. Therefore the probability of winning at least once in these 52 attempts is

    1 – ((13,999,999/14,000,000)^52)

    A quick calculation in my head (only kidding) gives 0.00000371428 (to six sig. figs.), whereas 1 in 14,000,000 equals 0.0000000714286 (6sf).

    Therefore you are 51.9999 (6sf) times as likely to win the lottery entering it 52 times as to win it by entering it only once. This should come as a relief to Camelot, since if it were equally likely to win the lottery once from one attempt as it were to win it from x attempts, there would be no logical reason to enter more than once.

  13. You’re kinda right, and kinda wrong. The chances of winning the lottery are, of course, exactly the same for any given ticket. But the chances of winning it over a given period of time are not the same, if you play regularly. In this case it’s a year, so the chances of a weekly player winning during that year are 52 times the chances of a single ticket winning. Or think of it another way – if all the tickets were bought on the same day, who has the greater chance of winning, someone who buys one ticket, or someone who buys 52 tickets?

  14. I haven’t done statistics since high school, over 20 years ago, so I’m a bit hazy. But I thought the probability of winning a lottery ticket if you played every week for a year was the same as if you played only once. This is because the probability isn’t cumulative – you have just the same probability of winning next week whether or not you played this week. If this were not the case then if you played long enough the probability of winning would be 1: an absolute certainty.

    This is in contrast to some probabilities that are cumulative, eg If I’m waiting for a bus, the longer I wait the more likely it is that the bus will appear (except in London where they operate using probability from another dimension).

Leave a reply