Morbid statistics

2»

Comments

  • slowbike
    slowbike Posts: 8,498
    Well - the OP just wanted some explanation to the relatively constant cycling mortality rate - ie explain the statistics ...

    As statistics do feed into probability it's reasonable to assume the mortality rate will be similar this year and next - which seems disproportionate to the number of journeys and/or miles and/or riders on the road given the explosion in cyclist numbers in recent years - which raises the inevitable question - WHY ...

    If the number of deaths are not linked to the number of journeys, miles or number of cyclists on the road then what is the link? Answering that is the start to reducing these fatalities.

    Of course, we could save time and simply answer that the problem is usually motorists not looking properly - coupled with a few cyclists not riding defensively - gives you the results we currently see.

    Statistically cyclists are unlikely to die whilst riding - unless they decide to ride like an arse on busy roads in rush hour.... then the probability of Darwin selection is much greater ... ;)
  • nathancom
    nathancom Posts: 1,567
    smidsy wrote:
    OK my very last word on this. I can see that my attempt to keep things simple has been confused.

    Your definitions agree with my simple summary.
    nathancom wrote:
    Statistics -
    The mathematics of the collection, organization, and interpretation of numerical data, especially the analysis of population characteristics by inference from sampling..
    History. You can not collect something which does not exist (has not occurred).
    nathancom wrote:
    Probability -
    The likelihood of the occurrence of an event. The probability of event A is written P(A). Probabilities are always numbers between 0 and 1, inclusive.
    Future. If something has already happened it has no likelhood - it is absolute/difinitive/already occurred.

    I have never said that statistics are not used as part of prediction, nor have I claimed that they are unrelated to probabilty.
    If 19th Century mill workers had a mortality rate of 100 per 10,000 in any given year we can work out that the P(dying) is equal to 0.01 or 1%. A set of data, namely death records of mill workers, has been collected and analysed to allow us, 150 years later to work out the probability of death experienced by a subset of the population. Probability maths has been used to arrive at a meaningful piece of information.

    This has nothing to do with a future event - there is no necessity to apply probability only to future events as it is simply mathematics. Again, your understanding of the terms is very vague but you are trying to tell Ai_1 (who has a clear understanding of the terms), what they mean.
  • Simon_Rhino
    Simon_Rhino Posts: 66
    apologies if someone has already mentioned it, but I'm sure I saw a Hans Rosling program where he looked at this (although IIRC it was pedestrians rather than cyclists)
  • hangeron
    hangeron Posts: 127
    OP

    have a look at this -

    http://understandinguncertainty.org/davidsblog
  • thegibdog
    thegibdog Posts: 2,106
    smidsy wrote:
    the outcome of me being wrong is not even possible. There is still a chance though.
    Brilliant! :D
    https://gibworth.mypixieset.com
  • ai_1
    ai_1 Posts: 3,060
    thegibdog wrote:
    smidsy wrote:
    the outcome of me being wrong is not even possible. There is still a chance though.
    Brilliant! :D
    Love it - why was I even trying to reason with this.......
  • vertigo16
    vertigo16 Posts: 91
    Statistically cyclists are unlikely to die whilst riding - unless they decide to ride like an ars* on busy roads in rush hour.... then the probability of Darwin selection is much greater ... ;)

    A lovely example of the diference between statistics and probability right there!
  • slowbike
    slowbike Posts: 8,498
    vertigo16 wrote:
    Statistically cyclists are unlikely to die whilst riding - unless they decide to ride like an ars* on busy roads in rush hour.... then the probability of Darwin selection is much greater ... ;)

    A lovely example of the diference between statistics and probability right there!

    There is a direct correlation between statistics and probability. But you have to choose the right statistics. ....

    If you're an average cyclist you're unlikely to be killed on the road.
    There are no statistics gathered for cyclists who ride like an arse, but if there were they'd show that they have a higher proportion of accidents and a greater proportion of the fatalities.

    This is how many probabilities are calculated - statistics that fit the situation.
    Where there aren't statistics then a certain amount of educated guesswork is involved.
  • smidsy
    smidsy Posts: 5,273
    Ai_1 wrote:
    thegibdog wrote:
    smidsy wrote:
    the outcome of me being wrong is not even possible. There is still a chance though.
    Brilliant! :D
    Love it - why was I even trying to reason with this.......

    Beats me, but you seemed insistant :mrgreen:
    Yellow is the new Black.
  • vertigo16
    vertigo16 Posts: 91
    In this analogy the assumption was that the coin flip was a 50/50. Under those circumstances the probability of 100 consecutive heads is so infinitesimally small that it is for all sensible purposes impossible.

    It's just as likely to get 100 consecutive heads as it is to get any other outcome though, isn't it?



    (Just messing!)
  • slowbike
    slowbike Posts: 8,498
    vertigo16 wrote:
    In this analogy the assumption was that the coin flip was a 50/50. Under those circumstances the probability of 100 consecutive heads is so infinitesimally small that it is for all sensible purposes impossible.

    It's just as likely to get 100 consecutive heads as it is to get any other outcome though, isn't it?



    (Just messing!)

    It's as likely to get 100 consecutive heads as it is to get 100 consecutive tails or alternate heads/tails or any other specific combination order.
  • smidsy
    smidsy Posts: 5,273
    In all probability you are equally likely to get a head or a tail on each flip.

    Statistically however it is very unlikley you will achieve 100% heads or 100% tails for a given number of flips.

    See its all perfectly easy.
    Yellow is the new Black.
  • vertigo16
    vertigo16 Posts: 91
    smidsy wrote:
    In all probability you are equally likely to get a head or a tail on each flip.

    Statistically however it is very unlikley you will achieve 100% heads or 100% tails for a given number of flips.

    See its all perfectly easy.

    Quite, but my facetious point (as Slowbike has pointed out and clarified) was that there is no sequence more likely than all heads - every sequence is just as unlikely as 100% heads.
  • vertigo16
    vertigo16 Posts: 91
    Slowbike wrote:
    vertigo16 wrote:
    Statistically cyclists are unlikely to die whilst riding - unless they decide to ride like an ars* on busy roads in rush hour.... then the probability of Darwin selection is much greater ... ;)

    A lovely example of the diference between statistics and probability right there!

    There is a direct correlation between statistics and probability. But you have to choose the right statistics. ....

    If you're an average cyclist you're unlikely to be killed on the road.
    There are no statistics gathered for cyclists who ride like an ars*, but if there were they'd show that they have a higher proportion of accidents and a greater proportion of the fatalities.

    This is how many probabilities are calculated - statistics that fit the situation.
    Where there aren't statistics then a certain amount of educated guesswork is involved.


    ...which was my point.

    In the post above the guy had begun by saying "Statistically.." and had then gone on to infer a probability. Conceding that this probability wasn't applicable to all cyclists (probably because by his own logic he believed that the majority of cyclists are actually less likely than average to be killed), he then judged the probability of death for a subset of the data, for which he didn't say "statistically", as though it was any less appropriate.

    Statistics is the collection and processing of evidence (usually large data sets). The aim of statistics is usually - though not always - to make an assessment of something's probability. That doesn't mean probability and statistics are the same thing (as has been debated earlier on in the thread), although they often go hand in hand. As his post demonstrates, one's conclusion about probability depends on the evidence one takes into account (demonstrating how you should be sceptical of many statistics).

    More fundamentally, I don't think there is any need to say "statistically" at the beginnig of his post as it doesn't change anything. Statistically, as opposed to what? The frequency with which people do use the word though is, in my opinion, the reason statistics have such a bad press these days. "Statistically" is used so much as a "you can't question me - my opinion is backed up by evidence, so there" approach, and it is this which I think has really caused the argument on this thread.
  • slowbike
    slowbike Posts: 8,498
    vertigo16 wrote:
    "Statistically" is used so much as a "you can't question me - my opinion is backed up by evidence, so there" approach.

    I know - funny isn't it ... a long time ago I studied statistics and probability so it's always quite fun when someone chucks a statistic into a conversation to "prove a point" ... the point being that they don't understand the parameters that made the statistics in the first place.
  • chris_bass
    chris_bass Posts: 4,913
    vertigo16 wrote:
    smidsy wrote:
    In all probability you are equally likely to get a head or a tail on each flip.

    Statistically however it is very unlikley you will achieve 100% heads or 100% tails for a given number of flips.

    See its all perfectly easy.

    Quite, but my facetious point (as Slowbike has pointed out and clarified) was that there is no sequence more likely than all heads - every sequence is just as unlikely as 100% heads.

    true, but I think the argument is generally the probability of getting 100 heads in a row (about 1/1.267651e30) vs not getting 100 in a row (about 1 - 1/1.267651e30) so t is much less likely
    www.conjunctivitis.com - a site for sore eyes

Categories