Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you: No adverts like this in the forums anymore. Times and dates in your local timezone. Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance. Email notifications for threads you want to watch closely. Being a part of the oldest wargaming community on the net. If you are already a member then feel free to login now.

First the Oregon school shooting, which just so happened to be the same school that Alek Skarlatos attended. Didnt really raise any alarm bells for me because a single isolated incident does not correlate or lead to causation.

But now this:

http://www.airforcetimes.com/story/breaking-news/2015/10/08/airman-1st-class-spencer-stone-stabbed-chest/73578902/

Airman 1st Class Spencer Stone, who helped take down a gunman on a train in Belgium, was stabbed four times in the chest in Sacramento early Thursday morning, Air Force Times has learned.

"A1C Spencer Stone has been transported to a local hospital, and is currently being treated for injury," Air Force spokesman Lt. Col. Christopher Karns said in an email in Air Force Times. "The incident is currently under investigation by local law enforcement. He is currently in stable condition."

Sacramento police tweeted: “The assault incident is not related to a terrorist act. Assault occurred near a bar, alcohol is believed to be a factor.”

A Sacramento TV station reported on the stabbing but did not include the victim's name. The stabbing happened about 12:45 a.m. on Thursday on a Sacramento street corner.

The victim suffered "multiple stab wounds to his torso," ABC10 Sacramento reported. Police initially said he was in critical condition but is now expected to survive.

Although the Air Force has identified Stone as the stabbing victim, Sacramento police have not officially named the person who was stabbed.

Sacramento police said they received a call from about the stabbing from a passerby, a news release says.

"It is believed that the victim was out with a group of friends when a physical altercation led to the victim being stabbed multiple times in his upper body," the news release says. "Detectives were called to the scene to assist with the investigation and the victim is currently being treated for what appears to be non-life threatening injuries."

The news release does not include the name of the man stabbed, who is only identified as "a man in his 20s."

"The Sacramento Police Department respects the identity of all crime victims and appreciates your patience as we work to provide you with further details," the news release says.

and I say... whats the probability of that?

Are you suggesting a conspiracy to attack the guys who got the Legion d'Honneur for their anti-terrorist work on the train recently?

It's easy to see a low probability event in the two cases however the true probability depends on how many people are involved in shooting or stabbing incidents every year, which is quite a lot in the USA.

I don't think the shooting was motivated by terrorism. Don't know about the stabbing.

I'd say the probability is low that these two attacks were concerted.

Stabbings at bars aren't exactly unheard of.

"The school that guy attended"...

Gonna hurt yourself if you keep stretching that hard.

This is Stretch Armstrong levels of reaching.

Quick and dirty probabilities, assuming 340 million people in the US, 10,000 gun-related murders per year, 1800 knife-related murders per year.

Likelihood of an average American being shot in any given year: .0000294%
Likelihood of being stabbed: .00000529%

Likelihood that specific individual A would be shot in any given year: 8.647058823529412e-14
Likelihood that specific individual B would be stabbed: 1.555882352941176e-14

Likelihood of both individual A and B being shot and stabbed in supposedly independent events: 1.34538062283737e-13

Obviously thats a super tiny number, and obviously theres a bit of inaccuracy there, as it doesn't statistically account for non-death shooting/stabbing incidents, nor does it account for statistical variations in probability based on area, situation, etc. Nor does it account for the lifetime probability of being involved in such an incident, but even then you're probably still looking at a % with a decimal point and 9 or 10 zeros in front of it.

If it was a terrorist, they would have made a big claim about it.

chaos0xomega wrote:
Quick and dirty probabilities, assuming 340 million people in the US, 10,000 gun-related murders per year, 1800 knife-related murders per year.

Might want to use the right numbers if you want to play a statistics game.

Except it is not "one person got stabbed and another person got shot". It's one person got stabbed, and there was a shooting at the school of another. It's fun to make big numbers, but keep the facts straight, please.

Plus, how exactly does this match any MO of a known terrorist group? They don't exactly do subtle, varied attacks. Why would they bother with attacking one guy's school, and a nonfatal stabbing on another?

I don't think the shooting had anything to do with it, but a stabbing on this particular guy? I spent time in Korea, got into my share of bar fights with armed GIs, spent time at bases here in the States with civilian party districts that were less than savory. I never got stabbed. Who pulls a knife in a bar fight? I think it was motivated by something other than just random bar fight.

 d-usa wrote:

chaos0xomega wrote:
Quick and dirty probabilities, assuming 340 million people in the US, 10,000 gun-related murders per year, 1800 knife-related murders per year.

Might want to use the right numbers if you want to play a statistics game.

http://www.economicpolicyjournal.com/2012/12/how-people-are-murdered-in-united-states.html

I rounded down because I know, from other sources, that the numbers have in fact been trending downwards.

Except it is not "one person got stabbed and another person got shot". It's one person got stabbed, and there was a shooting at the school of another. It's fun to make big numbers, but keep the facts straight, please.

The probability of being involved in a school shooting is actually even smaller than the probability of being murdered by a gun.

Plus, how exactly does this match any MO of a known terrorist group? They don't exactly do subtle, varied attacks. Why would they bother with attacking one guy's school, and a nonfatal stabbing on another?

Its not necessarily terrorists, there are any number of state and non-state actors that could be involved. I mean, I'm putting on my tinfoil hat, might as well go "full gaktard". Russian FSB was conducting a false flag operation in France, these guys fethed up their plans by accident, now Russian FSB is trying to get revenge.

chaos0xomega wrote:
Its not necessarily terrorists, there are any number of state and non-state actors that could be involved. I mean, I'm putting on my tinfoil hat, might as well go "full gaktard". Russian FSB was conducting a false flag operation in France, these guys fethed up their plans by accident, now Russian FSB is trying to get revenge.

Don't see the point of a russian false flag operation aiming to murder random people. It wouldn't get us more involved in the fight against ISIS.
And the shooting/stabbing would a really weak "revenge".

And by the way there were more than 3 heroes, but they're not american so nobody talks about them

I wasnt necessarily being serious about the Russian false flag thing, it was more of an example of a situation in which someone other than a non-state actor would be involved in something like this.

chaos0xomega wrote:

 d-usa wrote:

chaos0xomega wrote:
Quick and dirty probabilities, assuming 340 million people in the US, 10,000 gun-related murders per year, 1800 knife-related murders per year.

Might want to use the right numbers if you want to play a statistics game.

http://www.economicpolicyjournal.com/2012/12/how-people-are-murdered-in-united-states.html

I rounded down because I know, from other sources, that the numbers have in fact been trending downwards.

Keep in mind that I might be a government agent send to DakkaDakka to spread misinformation, but let me tell you why your numbers are crap to begin with:

1) Number of people in the US. Right there the rest of your calculations are pointless because you are not even looking at the right population to begin with. If you are trying to figure out what the statistical risk of a women being raped is you don't start out with "there are 340 million people in the US..." and if you are trying to figure out what the statistical risk of a child being killed by a firearm is you don't start out with "there are 340 million people in the US...". The population you need to start looking at is "Number of people in the US that go to bars". Even if we want to junk it up and just assume that people under 21 don't go to bars (23% of the total population) and that people over 65% don't go to bars (15% of the population) that brings your population down to less than 200 million. And that's just the roughest of cuts to narrow it down to the population. Doing a quick Google it looks like on average ~20 million people a month go to a bar or nightclub. Even if you want to go with that number you need to figure out how many of those are repeat customers to narrow it down to the actual number of individuals that go to a bar. That would be your population. If you want to make it really specific you can figure out the number of people going to bars where the guy was stabbed. Now you got really useful statistics. But "there are x people in the US" is useless.

So just for fun lets find out the population of Sacramento that is between 18 and 65: ~315,000. If we knew how many people went to bars the number would be a lot smaller.

2) Number of people killed by being stabbed. Useless mostly for two reasons. First of all he wasn't killed, so it's a bad statistic to use for that. Second of course is that you then end up using "number of people killed" to come up with "risk of being stabbed". The vast majority of people who are stabbed are not, in fact, killed. So looking at deaths doesn't really help us come up with "risk of being stabbed and surviving", which is of course what happened to our friend. So if you want to get a better idea of what happened you might look at "number of people stabbed in the US", "number of people being stabbed in alcohol related incidents", "number of people stabbed in bars", or "number of people stabbed in Sacramento".

As to the school:

1) Number of people in the US. Again, completely useless. Especially since a person wasn't even targeted to begin with. The school was the target. So by looking at number of schools we go from 340 million people to 4,140 colleges, which I probably don't have to point out is la bit of a statistical difference. So now you can look at the odds that a random person just happens to go to that school and look at the odds that a random person shoots up that school. The odds of one of 4,140 colleges being shot at is a lot higher than the odds of one specific person in the US being shot.

2) Number of people killed by guns in the US. Again, completely useless. Again especially since he wasn't shot and killed. So the number to look at would be school shootings.

Both of those combined:

Lots of complicated maths.

Summary:

Using random statistics completely unrelated to what you are trying to look at is useless and gives you useless numbers. Start by looking at the right numbers to begin with. You still end up with numbers that are unlikely, but my gut feeling is that is is probably still more likely that this happened to those two people than any given person winning the lottery. And we all know that there are still lots of lottery winners out there.

There's all kinds of coincidences in reality.

Guy my dad knew in 'Nam spent five years in-country, LRRP, went through some truly hair-raising, butt-clenching gak, 2 Purple Hearts, bunch of other medals, gets out, gets killed 3 weeks later in a construction accident.

Did the Viet Cong or the NVA set up that accident?!

Of course fething not. Caca occurs.

 Psienesis wrote:
There's all kinds of coincidences in reality.

Guy my dad knew in 'Nam spent five years in-country, LRRP, went through some truly hair-raising, butt-clenching gak, 2 Purple Hearts, bunch of other medals, gets out, gets killed 3 weeks later in a construction accident.

Did the Viet Cong or the NVA set up that accident?!

Of course fething not. Caca occurs.

I grew up in a small town in Germany before moving to BFE rural western Oklahoma.

One day I was pulled over for speeding by a deputy who was stationed in my old home town.

The odds of two completely unrelated people from opposite ends of the world both spending time in the same small town in Germany and then ending up driving past each other on a small country road in Oklahoma is pretty dang low.

I doubt it was anything deliberate, but you never know. It should at least be investigated.

 d-usa wrote:

Spoiler:

Keep in mind that I might be a government agent send to DakkaDakka to spread misinformation, but let me tell you why your numbers are crap to begin with:

1) Number of people in the US. Right there the rest of your calculations are pointless because you are not even looking at the right population to begin with. If you are trying to figure out what the statistical risk of a women being raped is you don't start out with "there are 340 million people in the US..." and if you are trying to figure out what the statistical risk of a child being killed by a firearm is you don't start out with "there are 340 million people in the US...". The population you need to start looking at is "Number of people in the US that go to bars". Even if we want to junk it up and just assume that people under 21 don't go to bars (23% of the total population) and that people over 65% don't go to bars (15% of the population) that brings your population down to less than 200 million. And that's just the roughest of cuts to narrow it down to the population. Doing a quick Google it looks like on average ~20 million people a month go to a bar or nightclub. Even if you want to go with that number you need to figure out how many of those are repeat customers to narrow it down to the actual number of individuals that go to a bar. That would be your population. If you want to make it really specific you can figure out the number of people going to bars where the guy was stabbed. Now you got really useful statistics. But "there are x people in the US" is useless.

So just for fun lets find out the population of Sacramento that is between 18 and 65: ~315,000. If we knew how many people went to bars the number would be a lot smaller.

2) Number of people killed by being stabbed. Useless mostly for two reasons. First of all he wasn't killed, so it's a bad statistic to use for that. Second of course is that you then end up using "number of people killed" to come up with "risk of being stabbed". The vast majority of people who are stabbed are not, in fact, killed. So looking at deaths doesn't really help us come up with "risk of being stabbed and surviving", which is of course what happened to our friend. So if you want to get a better idea of what happened you might look at "number of people stabbed in the US", "number of people being stabbed in alcohol related incidents", "number of people stabbed in bars", or "number of people stabbed in Sacramento".

As to the school:

1) Number of people in the US. Again, completely useless. Especially since a person wasn't even targeted to begin with. The school was the target. So by looking at number of schools we go from 340 million people to 4,140 colleges, which I probably don't have to point out is la bit of a statistical difference. So now you can look at the odds that a random person just happens to go to that school and look at the odds that a random person shoots up that school. The odds of one of 4,140 colleges being shot at is a lot higher than the odds of one specific person in the US being shot.

2) Number of people killed by guns in the US. Again, completely useless. Again especially since he wasn't shot and killed. So the number to look at would be school shootings.

Both of those combined:

Lots of complicated maths.

Summary:

Using random statistics completely unrelated to what you are trying to look at is useless and gives you useless numbers. Start by looking at the right numbers to begin with. You still end up with numbers that are unlikely, but my gut feeling is that is is probably still more likely that this happened to those two people than any given person winning the lottery. And we all know that there are still lots of lottery winners out there.

I was thinking the same thing, the 'killed by knives = stabbings' is just way too obvious.

We could also go into full-on "blame the victim" mode:

The guy was probably all "I'm a hero, look at my medals" and hitting on every girl on the bar telling them to drop their loser boyfriends and drop their panties. No wonder some guy stabbed him!

"A1C Spencer Stone has been transported to a local hospital, and is currently being treated for injury," Air Force spokesman Lt. Col. Christopher Karns said in an email in Air Force Times. "The incident is currently under investigation by local law enforcement. He is currently in stable condition."

Sacramento police tweeted: “The assault incident is not related to a terrorist act. Assault occurred near a bar, alcohol is believed to be a factor.”

It always is
I do not miss those days of dealing with my children who had bouts Liquid Motivation

Edit: This has already been addressed...

In terms of the OR shooting, if the goal were really to target the serviceman the shooter would have simply targeted him. Why walk into a school shooting people based on their religion if your goal is to shoot a particular person? As others have said...definitely reaching.

So the relevant stats may be 'number of drunken troopers who get into bar fights' compared to 'number of drunken troopers who get into bar fights where a weapon is involved'.

1) Number of people in the US.

We can't make the assumption that we have to look at the portion of the population that frequents bars, because we dont actually know he was at a bar. What has been stated was that there was an altercation *outside* of a bar (we know this because it was called in by an employee of a liquor store either across the street or next door depending on the source). Even if he was, for all we know the 'bar' was an Applebees or a generic restaurant that serves alcohol, in which case the number goes back up because you're looking at a more common event.

Re: schools, thats also debatable whether or not the school was the 'target'. It appears his target was a specific subset of the general population (Christians), the school was merely a location of convenience to find a grouping of said targets, unlike say Columbine, where the targets were specifically fellow students.

Number of people killed by being stabbed

Number of people killed by guns in the US

The stats for stabbing incidents is a lot harder to find than the stats for stabbing deaths, Im using what i got.

In any case, I said 'quick and dirty' for a reason, and caveated that it doesnt account for statistical variation/ was only a top-level probability calculation.

 d-usa wrote:

 Psienesis wrote:
There's all kinds of coincidences in reality.

Guy my dad knew in 'Nam spent five years in-country, LRRP, went through some truly hair-raising, butt-clenching gak, 2 Purple Hearts, bunch of other medals, gets out, gets killed 3 weeks later in a construction accident.

Did the Viet Cong or the NVA set up that accident?!

Of course fething not. Caca occurs.

I grew up in a small town in Germany before moving to BFE rural western Oklahoma.

One day I was pulled over for speeding by a deputy who was stationed in my old home town.

The odds of two completely unrelated people from opposite ends of the world both spending time in the same small town in Germany and then ending up driving past each other on a small country road in Oklahoma is pretty dang low.

I was born and raised in Florida. Joined the Army at 19, got stationed at Ft. Lewis, WA. Ran into one of my friends from high school randomly on the streets of Seattle one day. Coincidence? Most definitely.

chaos0xomega wrote:
Quick and dirty probabilities, assuming 340 million people in the US, 10,000 gun-related murders per year, 1800 knife-related murders per year.

Likelihood of an average American being shot in any given year: .0000294%
Likelihood of being stabbed: .00000529%

Likelihood that specific individual A would be shot in any given year: 8.647058823529412e-14
Likelihood that specific individual B would be stabbed: 1.555882352941176e-14

Likelihood of both individual A and B being shot and stabbed in supposedly independent events: 1.34538062283737e-13

Obviously thats a super tiny number, and obviously theres a bit of inaccuracy there, as it doesn't statistically account for non-death shooting/stabbing incidents, nor does it account for statistical variations in probability based on area, situation, etc. Nor does it account for the lifetime probability of being involved in such an incident, but even then you're probably still looking at a % with a decimal point and 9 or 10 zeros in front of it.

As far as I can tell your maths is right as a fairly rough estimate, but the point is it is the probability of any two individuals in the USA being shot and stabbed in one year.

It looks like targeting because you know with hindsight these two people have a connection between them apart from their being involved in violence recently. However, wild coincidences happen all the time if you look for connections between events.

Lincoln was shot in Ford Theater.
JKF once shot a load into Marilyn Monroe.

Or something. Did I do that right?

Its not just that violent events occurred, its the timing of them. Thats why I, again, did a top level quick and dirty stats calculation on it, you can add in all sorts of factors and variables that will inflate and deflate the number (for example, I could include the probability that specific person A and B were both involved in separate incidents within the same week, or 7 day period, depending on how you want to measure that timeframe, and arrive at an even smaller number).

chaos0xomega wrote:
Its not just that violent events occurred, its the timing of them. Thats why I, again, did a top level quick and dirty stats calculation on it, you can add in all sorts of factors and variables that will inflate and deflate the number (for example, I could include the probability that specific person A and B were both involved in separate incidents within the same week, or 7 day period, depending on how you want to measure that timeframe, and arrive at an even smaller number).

You could. And it would be as useful as the stats you used and the analysis you provided. As in, not very.

"Quick and dirty" generally does not mean "Used totally irrelevant stats". Neither does 'Top Level'.

Fair, but it was good enough to illustrate that the % chance of an occurrence like this gets you a smaller number than most people can comprehend. Even if you add in some factors to inflate it up, you're still looking at a fraction of a percentage.