Will I die while cycling to work?

Based on a figure of 0.42 deaths per million hours of cycling, the length of my commute and some assumptions about when I'll stop working, I worked out the probability that my death will be on-bike.

It came out as 1 in 200, or 0.5%. Huh. I expected it to be a bit lower than that.

Did this need its own thread?

Think your maths is probably right. I fared a bit better at 1 in 375.

Earthloop, if your assumptions didn't factor in your X years of cycle experience, benefits of high viz clothing & lighting, & defensive riding style (as seen on the London Spiral) then I am sure your odds are far longer.

Yeah, but it's easy to convince yourself that you're above average, most drivers mange this.

Podcast:...do the health benefits of cycling outweigh the risk of injury? Professor David Spiegelhalter goes through the numbers. More or less, Radio 4

Bit of discussion in this episode of more or less.

Probably not but you can maximize your chances by not wearing headphones or performing risky manouvers

and carrying automatic weapons. Don't forget the automatic weapons.

Look at it another way. As the probability of your death is 100%, then the probability of dying not whilst commuting to work is 99.5%

Or another way, that 0.5% is a lifetime risk, so given 40 years of commuting to work, that a risk of 0.0125% per year.

Or yet another way, population statistics are meaningless at the individual level.

Carrying automatic weapons would likely increase your risk of injury. Those things are hard and heavy, landing on top of one will likely exacerbate any injury.

Not sure what headphones have to do with anything, unless you're wearing noise cancelling megacans blasting out Sunn O))) at full volume

and carrying automatic weapons. Don't forget the automatic weapons.

Join LFGSS automatically become a weapon?

Look at it another way. As the probability of your death is 100%, then the probability of dying not whilst commuting to work is 99.5%

Or another way, that 0.5% is a lifetime risk, so given 40 years of commuting to work, that a risk of 0.0125% per year.

Or yet another way, population statistics are meaningless at the individual level.

What's "meaningless" about these examples? Seem to me perfectly understandable and equally surprising as the original result.

I worked out the probability that my death will be on-bike.

It came out as 1 in 200, or 0.5%. Huh. I expected it to be a bit lower than that.

What's the probability of dying early from obesity, diabetes or heart disease should you not cycle to work but sit in a car/bus/train?

Bloody high. I'd make something up off the top of my head and say around 35%

Carrying automatic weapons would likely increase your risk of injury. Those things are hard and heavy, landing on top of one will likely exacerbate any injury.

Not sure what headphones have to do with anything, unless you're wearing noise cancelling megacans blasting out Sunn O))) at full volume

True about packing a piece.

I think even stupid earbuds cut people off from their surroundings. Always amazes me some of the riders I see with phones in seeming to never bother looking over their shoulders before changing direction. And it makes it hard for them to hear me muttering 'you fucking dick' when they nearly take me out doing this.

Carrying automatic weapons would likely increase your risk of injury. Those things are hard and heavy, landing on top of one will likely exacerbate any injury.

Doing it wrong.

What's the probability of dying early from obesity, diabetes or heart disease should you not cycle to work but sit in a car/bus/train?

Bloody high. I'd make something up off the top of my head and say around 35%

Agreed, no doubt that I'm much better off cycling to work than not. Although 0.5% is low, I'd expected much lower. I don't really know why.

How did you work it out?

What's "meaningless" about these examples? Seem to me perfectly understandable and equally surprising as the original result.

No, that's not what I mean. When you see a statistic that says the risk of x happening is y%, that's applied to a population as a whole and not to one particular individual's risk

Getting back to the original question...

Probably not.

Look at it another way. As the probability of your death is 100%, then the probability of dying not whilst commuting to work is 99.5%

This is great.

Stats (in general) say what the person using them wants them to say.

I may be wrong but I think as usual the statistics are misleading and mathmatically incorrect though I may be wrong. The statistic that for every 1 million hours of cycling there are 0.42 deaths concerns cyclists as a whole. I don't know who came up with the statistic or how it was calculated but I can only assume it relates to the number of deaths of cyclist on the road in a certain area divided by the total hours cycled in that area (i.e. the UK).

So the 0.42 deaths per million hours of cycling is not per individual but as a group of cyclists in the area concerned. You will then have to divide the 0.42 by the number of cyclists cycling at any one time to get the real percentage.

Statistics can be funny like that. Just like dividing the number of people who win the lottery one day by the number of people who bought a lottery ticket does not give you the true likelihood of winning. (you have to work it out theoretically in the case of the lottery).

No, that's not what I mean. When you see a statistic that says the risk of x happening is y%, that's applied to a population as a whole and not to one particular individual's risk

^This is what I'm trying to say

This is great.

Stats (in general) say what the person using them wants them to say.

^This is also very true.

As Disraeli said (according to Mark Twain) "There are three kinds of lies: lies, damned lies, and statistics."

Came out about the same - 31 now, say another 30 years cycling 5 days a week 45 weeks a year for 2x50mins per day
30×5×45×(100÷60)÷1000000×0.42=0.004725

If you're working out the chances of dying on a bike you'd also need to factor in that x% of your life is also spent on a bike so if you assume that you may have an aneurism, heart attack etc at random then it's a shade more likely than the above calculation.

To check my figures, I looked at strava. Currently we're on week 46 of the year, so 52×(46÷52)×(100÷60)×5=383 hours I reckon I must've done. Strava has counted 341 so far this year, so it's not too far off. 8 weeks weren't logged, so factoring that in gives 341×(52÷44) = 403 hours

380-400 seems reasonable, given that the summer brought with it all those wonderful extra hours with TNRC

How did you work it out?

Probably:

1. I will cycle a total of X hours before I stop cycling to work

2. 1000 000/X = Y

3. Y x 0.42 = 200