This post has been de-listed
It is no longer included in search results and normal feeds (front page, hot posts, subreddit posts, etc). It remains visible only via the author's post history.
#
After a while away attenging to personal matters I'm excited to be back and writing again! It was tough being away for so long! Still, I'm doing much better now and I've gotten a *lot* done so all is well!
## Perception of Risk
It's well understood in the world of probability and statistics that people in general have a very poor ability to assess and understand risk levels and how dangersou anything actually is. We also have a lot of instinctual biases that cause us to misjudge the real danger levels, and our ability to handle things. Ultimately this tends to manifest in ways that exaggerate risks or exaggerate the perceived amount of harm that is likely to come from different activities. Everyone in the world is susceptable to this; *even the most well studied actuaries and mathematicians in the field of probability admit to instincts and reactions that go against the statistical truth.*
As a result it's even harder to compare risks between activities, or events, and understand that comparison on a rational intellectual level, and on an emotional level. **This can be a good thing!** It's where the sense of terror that makes roller coasters and most fear play activities thrilling comes from. It's also where we get our capacity to empathise with each other from; we perceive outcomes as far more likely than they are & so relate to the people that they happen to on a deeper level. We always thinki "That could be me" - even when that's almost certainly untrue.
On the other hand, sometimes we also have a bias towards drastically *underestimating* risks. This generally happens once taking a risk becomes routine. We mostly remember the anxiety and nerves that come with our first time operating a car or riding with someone who's a bit less cautious of a driver than we would prefer. At the same time, after a year of not dying we generally become a bit blase towards these activites; even getting to a point where we don't perceive them as dangerous at all. This can be **very deadly** when it applies to things with a high level of danger. We lose the ability to be cognizant of how on the edge we actually are and thus ignore hazards and take extra risks that we never would had our experiences not made us complacent.
## Comparing Risks
The good news is that scientests and actuaries have created methods of giving us a reasonable way to compare the risk level & danger level of doing certain things objectively. Additionally groups such as the FAA, NTSB, CDC, and many international militaries have created incredible systems of risk analysis to determine whether something is an acceptable or unacceptable risk. Using these we can help ourselves see the reality of our potential harms.
Let's take **Driving** as our first baseline and walk through the process. The first step is to list as many of the unacceptable outcomes we can have from driving:
- dying
- killing a pedestrian
- killing a passenger
- killing another driver
- Severe injury
- injuring another driver
- injuring a passenger
- injuring a pedestrian
Is the list I've been able to come up with now, I'm sure everyone will have their own that they could add that would be unacceptable to themselves as well.
The second step is to look up the liklihood of each of those events happening in a given number of times we drive (so the units are {INCIDENT TYPE}/[TIME DRIVING], {INCIDENT TYPE}/{TRIP}, or {INCIDENT TYPE}/{DISTANCE}. They may be slightly different, but each gives us a useful baseline to work from.
The odds of dying in a car crash are about 1 in 103, so we can say 1%, the odds of injury are closer to 45%; the odds of dying as a pedestrian are 1 in 405 people. In pedestrian collisions the odds of killing the pedestrian are about 7%, 35% receive serious injury. Ultimately this isn't terribly useful until we find out the number of car crashes per mile driven: 1/366,000 or 0.0000028 crashes per mile driven. Unfortunately we also have to account for the other people in these crashes too. We know that about 11% of these are pedestrian collisions so we can split this out:
Odds of Injuring or Killing Ourselves: 0.0000028 x (1% 45%)
Odds of Injuring or Killing Another Driver or Passenger: 0.0000028 x (1% 45%) x 77%
Odds of Injuring or Killing Ourselves: 0.0000028 x 42% x 11%
(There are a few other collision types such as animal only that can't injure an outside person.
This gives us ta total of: 0.0000028 x (0.01 0.45 0.077 0.35 0.05) = 0.0000028 x 163%
Where the final number of unacceptable incidents per mile driven is: 0.0000046
Since the average person drives about 13,500 miles a year the risk level of driving is thusly:
0.0000046 INCIDENTS/MILE x 13,500 MILES/YEAR = 0.0621 Incident/Year.
As a result every year that we drive we accept that there is about a 1/20 chance of us having an unacceptable outcome from that activity. Our lifetime of driving (18 years old to 75 years old on average) is about 57 years. That means our expectation of an unacceptable outcome driving in our lifetime is: 57 {YEARS} x 0.0621 {INCIDENTS}/{YEAR} = 3.54. That is to say the average person can expect to experience between 3 and 4 traffic collisions where they or another person is seriously injured or killed during their lifetime. One every 26 years. (Note, this includes situations where multiple people are injured or killed in a single incident as separate incidents of fatality or injury so it's imperfect but illustrative of the number of people who will be injured or killed by a person driving during their life).
We can convert INCIDENTS/MILE (0.0000046) to INCIDENTS/HOUR (0.00005336) on average by using the average speed travelled (11.6 MPH including time stopped but operating the vehicle), and INCIDENTS/TRIP (0.0000506) by using the average trip distance (about 11 miles).
It should also be noted that **we have socially and individually accepted this level of risk as entirely acceptable to the point of triviality.** This is thus what I will consider a "reasonable risk tolerance baseline." That is: It's an activity that holds significant risk, and danger but is socially acceptable and ubiquitus in nature.
## A bit about units
Ultimately the best units for us to compare to are going to be INCIDENTS/TRIP as that can be generalized to INCIDENTS/EVENT which is what we can consider "doing the thing in play" for our needs. It's also super awkward to work with units that are so small in nature (1/10,000th of a unit) all the time. It's also awkward to type "Incidents Per Event" each time so I'm going to switch to "morbidity-events" or Nightengales "Ng" eual to 1 morbidity per 10,000 events (Coining this unit here). I selected 1/10,000 as that seems to be the threshold for where most people see a risk as "High" or "Unacceptable" in a routine activity. So 1Ng is an activity with questionable risk. 100Ng would be an activity with very high risk, and exceeding 1kNg indicates an extreme risk activity.
The unit is named for Florence Nightengale for her work in medical and wartime trauma statistical analysis and contributions to using that knowledge to improve trauma medical care and nursing care.
This then results in **Driving having a Risk of *0.506Ng* = 506mNg (506 milinightengales) ** This is our baseline by which all other risks will be compared for the rest of this series.
Two Calibrating activites with around 1Ng of risk are:
- Skydiving: 4Ng (So 8 x riskier than driving)
- Motorcycle Riding: 0.49Ng (So about the same as car driving - but MUCH greater fatality %)
## Kinky Activities
We don't get a lot of statistical data on kinky activities which lead to injury. In fact due to shaming and a desire to never admit fault out of fear of ostracization it's basically impossible to complie this data. Even the most trustworthy people are hesitant to admit when things have gone wrong to people outside the scene where it happened; on both sides of the Top/Bottom spectrum.
Still, there are a few inferences we can make about a few activities since their most heinous results will cause hospitalization or death. We can also use some Fermi Estimation to guess at the order of magnetude of people who participate in those activities and the number of times each year that they happen. Using this with the National Product Safety Commission website's catalogue of ER visits by type of trauma and what products were used it's possible to gleen some basic information then about where (within an order of magnetude) different activities actually fall. If there's no incidents in a given year where medical intervention is necessary then it's probable that nobody is getting seriously injured participating in that activity; after all.
This part of the analysis will be my next article as it will take some significant effort to compile the data and get some estimates. That said I do have a few preliminary estimates here:
Impact Play on Buttox: 1 Case Report Found from BDSM activities. Approximately 7,200,000 practitioners in the US, estimating an average of 1 activity per week 375,000,000 incidents per year. 1 case report per 5 years => 1 incident per 2 Billion events or 5 micro-Ng. This is to say **impact play *on the buttox* is roughly 100,000 times safer than driving.**
Breath Play with a partner: 16 Incidents Found, likely very under-reported. I'm going to guess about 5% of injuries are reported but that 80% of fatalities are. As a result this would lead to a number of about 40, so I'll round up to 100 to account for things I likely missed.
Far fewer people engage in breath play than spanking. 5% of surveyed groups say they have tried some kind of breath play at least once, so it's not totally negligable either. I'm going to estimate the population of participants is likely to be about the same as that of spanking, but that they participate only 10% as frequently. This would give about 37,000,000 incidents of people engaging in breath play each year. That makes breath play as a whole have a risk level of 2.7mNg. Note **this includes even "mild" breath play like a hand lightly pinching the nose, or breath holding games.** The incident rate for equipment assisted breath play is likely *much higher* (based on the idea that 10% of BDSM players do heavy breath play and that the incidents were all from heavy breath play) would put it right at 270mNg - **about the same as the risk of driving.**
Note: **Autoerotic play was not included in this calculation!!!!!**
Subreddit
Post Details
- Posted
- 1 year ago
- Reddit URL
- View post on reddit.com
- External URL
- reddit.com/r/BDSMnot4new...
No worries about tonal issues I got where you're coming from.
You're a bit ahead of me on this, as I've groups those additional variables into "hazards" outside of raw risk; it calls back to my first article but the final sections are going to be addressing the fact that identifying hazards is critical to responsible practice in any activity.
That said, the principle isn't as distorted based on pareto effects as you would expect for many of these because of the fact that we're all particles working in the same system. In other words while only 1% of drivers may present an extreme hazard at any given time we are all still on the same roads with all of them and as a result our risk is elevated since we also need to be on the other side of their risk equation. That ends up pulling the distribution into a much more symmetric shape. The inclusion of both getting hurt and hurting another as unacceptable is key to removing that bias.
Also if you're talking about rivers that's more of a Poisson thing since it's a fishy subject. ððĪŠ (okay, I'll stop with the math & language puns)
The same is somewhat true for kink risks, but it winds up looking a lot more like SCUBA diving and Aviation than something so simple as "higher skill means lower risk." Ultimately it's well understood in both fields that experience has a rapidly diminishing return for reducing risk level because complacency, poissonian distributed hazards, and many more factors yield a curve with multiple peaks with respect to skill level, experience time, and preparedness. The likelihood of injury is actually VERY well modeled with a Poison distribution as a result of this signal to noise issue.
As for the "Turkey Problem," it's not really applicable here. That's more about social structures and ignoring signs of danger and not looking for patterns in data broadly. Ultimately the issue isn't that we don't have a ton of trials (we have millions of trials each year creating data points) we just don't record any negatives so all we ever see are the positive results (in this case "yes, someone got hurt or died") leaving us to estimate exactly how big that negative space is. A better analogy is bow fishing in a lake. A boat of bow fishers goes out and comes back with 20 fish and says "THAT LAKE IS FULL OF FISH!" So you go with a net and try to scoop up fish at the end of a dock. You can't catch anything because the lake wasn't full, they just sampled the exact spots where the fish were and didn't even notice the empty water was there because it wasn't their focus.
This happens a LOT with medical case studies and doctors' in-action research. It's a really protracted and insidious version of a sampling bias.
Just remember: lots of light aircraft crash, and lots of people die in light aircraft crashes. At the same time every pilot alive today has dozens if not hundreds of hours flying light aircraft because it's how you learn to fly in the first place. When trying to learn how to manage hazards you need to focus exclusively on accident analysis. When you're looking to see if some broad ranging activity is too dangerous to participate in you need to keep your focus on the millions of hours that were boring; not the few hours that were exciting, or the even fewer of those that were tragic.