0795: "Conditional Risk"

This forum is for the individual discussion thread that goes with each new comic.

Moderators: Moderators General, Prelates, Magistrates

hthall
Posts: 95
Joined: Wed Oct 10, 2007 3:40 am UTC

0795: "Conditional Risk"

Postby hthall » Mon Sep 20, 2010 4:02 am UTC

Image

Title text: " 'Dude, wait -- I'm not American! So my risk is basically zero!' "

You can never discount observation effect when trying to assess statistics on humans. They're too clever for their own good as subjects.
Look at me, still talking when there's Science to do.

User avatar
CorruptUser
Posts: 10495
Joined: Fri Nov 06, 2009 10:12 pm UTC

Re: "Conditional Risk" Discussion (#795)

Postby CorruptUser » Mon Sep 20, 2010 4:04 am UTC

So at most 270 people know that statistic.

User avatar
glasnt
Posts: 539
Joined: Fri Jan 25, 2008 5:18 am UTC
Location: SQUEE!

Re: "Conditional Risk" Discussion (#795)

Postby glasnt » Mon Sep 20, 2010 4:06 am UTC

CRACK HI JOEE BOOM!
Image

User avatar
phlip
Restorer of Worlds
Posts: 7572
Joined: Sat Sep 23, 2006 3:56 am UTC
Location: Australia
Contact:

Re: "Conditional Risk" Discussion (#795)

Postby phlip » Mon Sep 20, 2010 4:07 am UTC

CorruptUser wrote:So at most 270 people know that statistic.

Not necessarily... it doesn't say those 1 in 6 deaths are caused by lightning...

Code: Select all

enum ಠ_ಠ {°□°╰=1, °Д°╰, ಠ益ಠ╰};
void ┻━┻︵​╰(ಠ_ಠ ⚠) {exit((int)⚠);}
[he/him/his]

roflcopter
Posts: 26
Joined: Thu Nov 13, 2008 2:14 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby roflcopter » Mon Sep 20, 2010 4:10 am UTC

I race sailboats and the weather doesn't always cooperate and we tend to forge on anyways, so naturally this subject comes up a bit, the chances of getting struck by lightning are about 1:280,000. Not to mention being on a boat that has gotten struck doesn't normally do anything to the passengers(personal experience, multiple times). Now the one statistic that makes me feel even safer when it comes to lightning is the fact that only about 10% of lightning struck people die from it, most actually have no adverse effects.

Take that Zeus!

User avatar
joee
Posts: 227
Joined: Mon Dec 03, 2007 5:53 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby joee » Mon Sep 20, 2010 4:13 am UTC

glasnt wrote:CRACK HI JOEE BOOM!
Image


Ahahahaha <3

Also, it seems dumb to quote that statistic. Surely you would need to know how many people are in a position to be struck by lightning: people who never leave their basements shouldn't count, for example
Hi glasnt.

jman077
Posts: 6
Joined: Sat Sep 26, 2009 3:52 pm UTC

Re: "Conditional Risk" Discussion (#795)

Postby jman077 » Mon Sep 20, 2010 4:14 am UTC

Imagine for a moment, in 2011, there is a significant increase in lightning strike deaths. All of the sudden, one in six xkcd readers are dead.

User avatar
Nintendon't
Posts: 61
Joined: Fri Apr 17, 2009 1:55 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby Nintendon't » Mon Sep 20, 2010 4:20 am UTC

Darwin award recipients, of course.

akrolsmir
Posts: 15
Joined: Wed Mar 25, 2009 4:42 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby akrolsmir » Mon Sep 20, 2010 4:21 am UTC

So I guess the other five understand statistics?
Image

User avatar
arbivark
Posts: 531
Joined: Wed May 23, 2007 5:29 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby arbivark » Mon Sep 20, 2010 4:32 am UTC

This comic contains an inside joke.

Sarck
Posts: 2
Joined: Mon Sep 20, 2010 4:30 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby Sarck » Mon Sep 20, 2010 4:32 am UTC

Wouldn't it have been better worded as,

"Of course, a lesser known corollary is that every time someone quotes that statistic, the annual death rate rises."?

Or something along those lines?

User avatar
Eternal Density
Posts: 5580
Joined: Thu Oct 02, 2008 12:37 am UTC
Contact:

Re: "Conditional Risk" Discussion (#795)

Postby Eternal Density » Mon Sep 20, 2010 4:38 am UTC

Chances of an xkcd forum member who goes out in lightning dying: 1 in 5.
Number of Randall's friends to die in lightning: 0
Play the game of Time! castle.chirpingmustard.com Hotdog Vending Supplier But what is this?
In the Marvel vs. DC film-making war, we're all winners.

Techrocket9
Posts: 4
Joined: Mon Sep 20, 2010 4:40 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby Techrocket9 » Mon Sep 20, 2010 4:42 am UTC

So if the risk of death by lightning is higher for those who know that statistic, xkcd is now responsible for millions of deaths among its readers. (Because all xkcd readers now know it)

User avatar
CorruptUser
Posts: 10495
Joined: Fri Nov 06, 2009 10:12 pm UTC

Re: "Conditional Risk" Discussion (#795)

Postby CorruptUser » Mon Sep 20, 2010 4:47 am UTC

Only a 1 in 6 chance of death? So if you tell your 100-year old Great Grandmother this statistic, she has a 5 in 6 chance of living till next year? Or, a 1% chance of making it to age 125?
Last edited by CorruptUser on Mon Sep 20, 2010 5:07 am UTC, edited 3 times in total.

Edrees
Posts: 32
Joined: Wed Dec 23, 2009 8:16 pm UTC

Re: "Conditional Risk" Discussion (#795)

Postby Edrees » Mon Sep 20, 2010 4:48 am UTC

Alt text made the comic for me! Hilarious.

Oh, and I hate these kind of statistics. I can never find the data/survey where they came from, so they might be completely made up. Or they might be done the complete wrong way - some people live in areas where lightning never strikes, or they use lightning rods around their house, etc.

I'd be more interested in knowing, of people caught standing outside during a severe lightning storm, what's their chances of being struck by lightning? Its probably much more significant.

Does the statistic that one in some million people gets killed by a shark include the people that never swim, or only the people that swim in the ocean - for instance? Huge difference.

squareroot
Posts: 548
Joined: Tue Jan 12, 2010 1:04 am UTC
Contact:

Re: "Conditional Risk" Discussion (#795)

Postby squareroot » Mon Sep 20, 2010 4:56 am UTC

Good.
Gave me numerous lolz.
+1 internets for Randall, yet again.
The alt text did me. (Yes, that was correct.)
<signature content="" style="tag:html;" overused meta />
Good fucking job Will Yu, you found me - __ -

thkng
Posts: 9
Joined: Fri Apr 09, 2010 12:36 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby thkng » Mon Sep 20, 2010 4:58 am UTC

Did you just condemn 1 in 6 of you readers to death by the wrath of Zeus?

myrcutio
Posts: 44
Joined: Wed Dec 30, 2009 7:28 pm UTC

Re: "Conditional Risk" Discussion (#795)

Postby myrcutio » Mon Sep 20, 2010 5:04 am UTC

Nintendon't wrote:Darwin award recipients, of course.


Darwin awards tend to favor situations in which the participants died mid-copulation. They would have to be having sex on a tin roof in the middle of a thunderstorm to get a Darwin award. That said, I invoke Rule 34 of the internet, and will begin searching for meteorological snuff films.
"Lightning must have hit it, and now it won't work in anything but Windows 95."

Faustus runs afoul of Microsoft.

Radders
Posts: 1
Joined: Mon Sep 20, 2010 4:42 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby Radders » Mon Sep 20, 2010 5:07 am UTC

I am sure XKCD knows that only 1 in 10 die from being struck by lightning. Which means, for a death rate of 1 in 6, the people who know this statistic are hit by lightning 1.6667 times a year. Do these people get hit once, and then again as they try to run for shelter? Or do they get hit once and decide that must be very unlucky, and continue?

...Only to repeat it again the next year?


Not that I can talk, this is the sort of logic I live by...

(Its now pinned to my wall, with a comma added to the annotation... Had to reread it 4 times to stop trying to guess what "The annual death rate among people who know 'that statistic is one in six.' " was...)

User avatar
cjmcjmcjmcjm
Posts: 1158
Joined: Tue Jan 05, 2010 5:15 am UTC
Location: Anywhere the internet is strong

Re: "Conditional Risk" Discussion (#795)

Postby cjmcjmcjmcjm » Mon Sep 20, 2010 5:18 am UTC

thkng wrote:Did you just condemn 1 in 6 of you readers to death by the wrath of Zeus?

Yes.
frezik wrote:Anti-photons move at the speed of dark

DemonDeluxe wrote:Paying to have laws written that allow you to do what you want, is a lot cheaper than paying off the judge every time you want to get away with something shady.

cptjeff
Posts: 44
Joined: Thu Sep 03, 2009 4:42 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby cptjeff » Mon Sep 20, 2010 5:35 am UTC

So cool story time: Not long ago I was on a hike with family, and it started to thunder. We went under a tree where we started putting a poncho up for a shelter. As we're doing this, my brother notices that the tree we're under has been struck by lightning in the past, and makes the "we're safe here since lighting never strikes twice, right?" joke. I then start glancing around. There were three or four trees within 50 feet or so that had been struck by lightning.

We decided to move away from that area.

User avatar
transatlantic
Posts: 6
Joined: Sat Jun 26, 2010 6:17 pm UTC

Re: "Conditional Risk" Discussion (#795)

Postby transatlantic » Mon Sep 20, 2010 5:39 am UTC

Let's see. The forum has nearly 60,000 members. Therefore there are at least that many reasonably regular readers. I suspect there are at least that many regular readers in the US. We should therefore expect a more than twenty thousand percent increase in deaths from lightning in the coming year.

cptjeff
Posts: 44
Joined: Thu Sep 03, 2009 4:42 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby cptjeff » Mon Sep 20, 2010 5:43 am UTC

transatlantic wrote:Let's see. The forum has nearly 60,000 members. Therefore there are at least that many reasonably regular readers. I suspect there are at least that many regular readers in the US. We should therefore expect a more than twenty thousand percent increase in deaths from lightning in the coming year.


This may help: http://www.alexa.com/siteinfo/http%3A%2F%2Fxkcd.com#

User avatar
skeptical scientist
closed-minded spiritualist
Posts: 6142
Joined: Tue Nov 28, 2006 6:09 am UTC
Location: San Francisco

Re: "Conditional Risk" Discussion (#795)

Postby skeptical scientist » Mon Sep 20, 2010 5:49 am UTC

Techrocket9 wrote:So if the risk of death by lightning is higher for those who know that statistic, xkcd is now responsible for millions of deaths among its readers. (Because all xkcd readers now know it)

thkng wrote:Did you just condemn 1 in 6 of you readers to death by the wrath of Zeus?

cjmcjmcjmcjm wrote:Yes.

transatlantic wrote:Let's see. The forum has nearly 60,000 members. Therefore there are at least that many reasonably regular readers. I suspect there are at least that many regular readers in the US. We should therefore expect a more than twenty thousand percent increase in deaths from lightning in the coming year.


Way to confuse correlation with causation.
I'm looking forward to the day when the SNES emulator on my computer works by emulating the elementary particles in an actual, physical box with Nintendo stamped on the side.

"With math, all things are possible." —Rebecca Watson

Sossen
Posts: 6
Joined: Tue Mar 23, 2010 11:21 pm UTC

Re: "Conditional Risk" Discussion (#795)

Postby Sossen » Mon Sep 20, 2010 6:24 am UTC

thkng wrote:Did you just condemn 1 in 6 of you readers to death by the wrath of Zeus?


Indeed you must be referring to the wrath of THOR!!!

SolkaTruesilver
Posts: 105
Joined: Mon Aug 20, 2007 11:30 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby SolkaTruesilver » Mon Sep 20, 2010 6:39 am UTC

The dudes who do the lightning clearly understand who you have to kill to become Master of the Earth: Americans.

As Canadian, I welcome our new Lightning Overlords.

Tova
Posts: 100
Joined: Wed Mar 24, 2010 3:44 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby Tova » Mon Sep 20, 2010 6:53 am UTC

So, wait... the joke is poking fun at people who don't understand "conditional risk", and then the punch line not only makes up a statistic but also demonstrates a lack of understanding of the very same thing? Huh.

"The annual rate of death by thunder strike among people who know that statistic and get caught in a thunder storm is one in six."

... guess it doesn't really pop. Oh well. Just ignore me.

NotAllThere
Posts: 145
Joined: Fri Aug 06, 2010 12:54 pm UTC

Re: "Conditional Risk" Discussion (#795)

Postby NotAllThere » Mon Sep 20, 2010 6:58 am UTC

myrcutio wrote: They would have to be having sex on a tin roof in the middle of a thunderstorm shouting all gods are bastards to get a Darwin award...


ftfy
yangosplat wrote:So many amazing quotes, so little room in 300 characters!

User avatar
SadinaSaphrite
Posts: 14
Joined: Mon May 24, 2010 7:34 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby SadinaSaphrite » Mon Sep 20, 2010 7:20 am UTC

One in six? Hey! That's the chance of a random object being a scone!

What are the chances of finding a scone if you get hit by lighting?
I put the fun back in dysfunctional and the romance back in necromancer!

User avatar
snowyowl
Posts: 464
Joined: Tue Jun 23, 2009 7:36 pm UTC

Re: "Conditional Risk" Discussion (#795)

Postby snowyowl » Mon Sep 20, 2010 7:32 am UTC

skeptical scientist wrote:Way to confuse correlation with causation.


Obviously they learned that statistic after being killed by lightning.
The preceding comment is an automated response.

User avatar
Time Kitten
Posts: 28
Joined: Mon Jul 19, 2010 4:38 am UTC
Location: Behind you!
Contact:

Re: "Conditional Risk" Discussion (#795)

Postby Time Kitten » Mon Sep 20, 2010 7:51 am UTC

Edrees wrote:Alt text made the comic for me! Hilarious.

Oh, and I hate these kind of statistics. I can never find the data/survey where they came from, so they might be completely made up. Or they might be done the complete wrong way - some people live in areas where lightning never strikes, or they use lightning rods around their house, etc.

I'd be more interested in knowing, of people caught standing outside during a severe lightning storm, what's their chances of being struck by lightning? Its probably much more significant.

Does the statistic that one in some million people gets killed by a shark include the people that never swim, or only the people that swim in the ocean - for instance? Huge difference.


Also, does it only count people getting killed via direct electrocution from it, or perhaps does it include associated fires, exploding trees, a fatality via extreme pissing of ones pants, or even long term health problems?

roflcopter wrote:I race sailboats and the weather doesn't always cooperate and we tend to forge on anyways, so naturally this subject comes up a bit, the chances of getting struck by lightning are about 1:280,000. Not to mention being on a boat that has gotten struck doesn't normally do anything to the passengers(personal experience, multiple times). Now the one statistic that makes me feel even safer when it comes to lightning is the fact that only about 10% of lightning struck people die from it, most actually have no adverse effects.

Take that Zeus!

Remember "no adverse effects" implies immediate. A lightning strike can fry your cartilage. You wont notice for a few years, but then your joints and back just start giving out as it doesn't replenish, as happened to my dad.

keithc
Posts: 124
Joined: Thu Mar 08, 2007 11:12 pm UTC
Contact:

Re: "Conditional Risk" Discussion (#795)

Postby keithc » Mon Sep 20, 2010 8:02 am UTC

myrcutio wrote:Darwin awards tend to favor situations in which the participants died mid-copulation.

No they don't. I got bored after going through the first three years' awards and only one sex-related incident occurred. They do tend to favour situations where the person who died is still of an age to have children.

dookiecheese
Posts: 22
Joined: Thu Feb 04, 2010 7:50 pm UTC

Re: "Conditional Risk" Discussion (#795)

Postby dookiecheese » Mon Sep 20, 2010 8:05 am UTC

Why are people so inept at understanding statistical probability? Also I'm getting tired of hearing the improbability of life arising naturally being some sort of 'proof' that life was designed--especially when the probability they give is about the same as getting any 5 card hand from a standard deck of cards.

User avatar
jacog
Posts: 56
Joined: Tue Dec 15, 2009 9:55 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby jacog » Mon Sep 20, 2010 8:11 am UTC

But what if the 1 in 7000000 that gets struck happens to be the fourth out of those five dentists - does that mean that only three out of five dentists will now recommend Trident? Statistics is confusing, man.

MasterDinadan
Posts: 11
Joined: Thu Apr 10, 2008 3:04 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby MasterDinadan » Mon Sep 20, 2010 8:28 am UTC

dookiecheese wrote:Why are people so inept at understanding statistical probability? Also I'm getting tired of hearing the improbability of life arising naturally being some sort of 'proof' that life was designed--especially when the probability they give is about the same as getting any 5 card hand from a standard deck of cards.


This amuses me too. I was on some website earlier that was talking about how "lucky" we are that Earth has the perfect conditions to sustain life and that it can only be the work of God.
I mean really? Life appears on a planet capable of sustaining life, and I'm to believe that it's a miracle? Tell me when life appears on a planet that doesn't sustain life, and then I'll be intrigued.

MasterDinadan
Posts: 11
Joined: Thu Apr 10, 2008 3:04 am UTC

Re: "Conditional Risk" Discussion (#795)

Postby MasterDinadan » Mon Sep 20, 2010 8:34 am UTC

Edrees wrote:Does the statistic that one in some million people gets killed by a shark include the people that never swim, or only the people that swim in the ocean - for instance? Huge difference.


Good statistics will make it very clear what population we are talking about. The statistics quoted in the comic simply said "Americans." It didn't say "Americans who go outside during lightning storms." If it had said that, there wouldn't have been any joke.

The joke is that the statistic he's referencing (whether accurate or not) refers to Americans in general (most of whom don't stand around outdoors during a bad storm) and he's assuming that the the statistic implies a certain probability for his survival regardless of other factors.

In other words, he's mistaking probability for conditional probability, which is pretty common.

I recently worked for the US Census, and my friend pointed me to an article talking about how X number of Census workers had been killed on the job and that I should be careful.

Of course, if you took that number as a percentage of the total number of Census employees, and compared it to the nationwide average for fatal car accidents, you'd find that Census workers are just as likely as anyone else to be involved in a fatal car accident. It's not as if carrying around a bag or putting a Census sign in my window attracts other reckless drivers...


If you liked this comic, I recommend The Cartoon Guide to Statistics:
http://search.barnesandnoble.com/Cartoo ... 0062731029
You might even learn something ;-)
Last edited by MasterDinadan on Mon Sep 20, 2010 8:36 am UTC, edited 1 time in total.

User avatar
creaothceann
Posts: 34
Joined: Fri Jan 26, 2007 9:44 am UTC
Location: Germany

Re: "Conditional Risk" Discussion (#795)

Postby creaothceann » Mon Sep 20, 2010 8:35 am UTC

Lightning in slow motion

Be afraid. Be very afraid.

User avatar
syko_lozz
Posts: 53
Joined: Fri Jan 11, 2008 5:30 am UTC
Location: Oz

Re: "Conditional Risk" Discussion (#795)

Postby syko_lozz » Mon Sep 20, 2010 8:42 am UTC

its called hyperbole people....

and yeah ^^ what he (MasterDinadan) said

Conditional risk is a huge issue in medicine. A lot of research is devoted to trying to determine more accurate risk factors for various groups of people.

One (atypical, but close-to-home) example of conditional risk- Its often said theres a 1 in 4 chance of losing your baby before the 12th week (which is ~10 weeks after conception). But most of those losses happen within a few days of implantation (around week 5). When you loose a baby at 9 weeks, people quote that statistic to try and make you feel better (as if trivialising someone's pain is useful). But the rate of loss is dramatically lower (and the psychological effects much more severe) after 8 weeks, we just don't have good enough stats for that.
Debate politics with a fern. If you lose, refuse to water it.

EthErealist
Posts: 14
Joined: Fri Mar 27, 2009 4:45 am UTC
Location: Los Angeles, CA
Contact:

Re: "Conditional Risk" Discussion (#795)

Postby EthErealist » Mon Sep 20, 2010 9:23 am UTC

LMAO, I love this comic

User avatar
Diadem
Posts: 5654
Joined: Wed Jun 11, 2008 11:03 am UTC
Location: The Netherlands

Re: "Conditional Risk" Discussion (#795)

Postby Diadem » Mon Sep 20, 2010 9:23 am UTC

roflcopter wrote:I race sailboats and the weather doesn't always cooperate and we tend to forge on anyways, so naturally this subject comes up a bit, the chances of getting struck by lightning are about 1:280,000. Not to mention being on a boat that has gotten struck doesn't normally do anything to the passengers(personal experience, multiple times

Are you claiming you've had a 1:280,000 event happen to you several times?

The odds of getting hit by lightning during a thunderstorm on open water are significantly higher than 1 in 280,000. As you can attest from personal experience :)
It's one of those irregular verbs, isn't it? I have an independent mind, you are an eccentric, he is round the twist
- Bernard Woolley in Yes, Prime Minister


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: No registered users and 54 guests