I'm "obsessing" over it, because its a bad example. If anything, this whole exchange is a good example of how people will ignore valid, relevant statistics because they find the implication of a given statistic inconvenient. Claiming to be an exception necessarily opens you up to greater scrutiny, so it behooves you to critically examine what, exactly, makes the affected group subject to the stated phenomenon, and honestly assess whether or not your circumstance is indeed an exception.
Here's an illustrative example from my realm of experience: between the fall of 2009 and spring of 2023, 363 people have died in avalanches in the US. Of those, 173 were skiers or snowboarders, and 17 of those were "inbounds" [0]. Somewhere on the order of 60 million people visit ski resorts each year[1]. A surface level understanding of those numbers might cause a ski-hill operator to conclude "the money I spend on avalanche mitigation is wasted, since deadly avalanches are significantly less likely at ski resorts".
Obviously this is a straw-man. But there are a number of ways to misinterpret the data to arrive at a conclusion you want (to whit, in that time period, 3 times as many skiers than snowboarders died in avalanches, and more than twice as many snowmobiler/snowbikers died in avalanches than snowboarders. But it takes really digging in to realize that a) there are also a lot fewer backcountry snowboarders vs backcountry skiers and b) snowmobiliers are consistently the least avalanche-educated of all backcountry users who actually play on avalanche prone slopes)
You're entirely correct that statistics don't dictate outcomes. But understanding what does dictate outcomes, without bias, can help you change those outcomes.
I just don't understand how we are talking past each other. We seem to agree while disagreeing (basically just cause guns are a contentious issue. I said it was a bad example at the outset...) right : Statistics do not dictate outcomes. I am identifying something that does definitely without question change outcomes for gun ownership (that is safety practices) and you seem to be making the argument that this does not change my chance of injuring myself with my own firearm.
In fact, I currently have no bullets in my house at all, and have no plans to purchase any. So right now my chance of injury is 0. It is not possible for me to be in the same cohort as the people with average risk of injuring themselves.
Its because, as you say, guns are contentious, and its a distractingly bad example.
I have a bit of a chip on my shoulder when somebody says "oh, this doesn't apply to me, I'm just that good". Most of the time, in my experience, that's somebody who can't be bothered to follow the rule, and not somebody who is actually a valid exception to the rule. I tend to throw water on such people whenever I can, to try to counter that effect.
I used to teach people how to rock climb, and the sketchiest people I knew were my fellow instructors, because they always had some complicated reasoning why the rules we taught (which, as the saying goes, are written in blood) didn't actually apply to themselves. It's some weird Dunning-Kruger thing going on, where they somehow convince themselves that all of this time spent avoiding death or serious injury by shear luck is somehow evidence of their greater skill and capability to manage the risks.
I watched several of these people take bad falls, walk away from them, and just go right back to what they were doing, with no behavioral change whatsoever.
When I called one of them out for not following a some minor rule around equipment checkout, he said "I don't need to follow that rule, I'll just be responsible". He went on to do a series of dangerous things that probably caused his students to quit climbing afterwards, and was ultimately fired because of gross misconduct surrounding (you guessed it) irresponsible behavior.
Reading climbing and mountaineering accident reports, the accounts from those involved are stacked with "it had always been ok in the past" or "it looked solid to me" or "I had done 'x' (which was completely unrelated to the outcome of the incident), so I thought we'd be ok".
There is less daylight between "there's no scenario where I won't flip out and turn my gun on somebody else" and "I was the last to use that firearm, so I know its unloaded, I don't need to check" than I think either of us are entirely comfortable.
I'll also say that the standard "good" practice is to just store your weapons and ammo in separate lockers, not entirely separate locations. That's a mitigation above and beyond (and punters need to understand why that makes a difference in the statistics).
Here's an illustrative example from my realm of experience: between the fall of 2009 and spring of 2023, 363 people have died in avalanches in the US. Of those, 173 were skiers or snowboarders, and 17 of those were "inbounds" [0]. Somewhere on the order of 60 million people visit ski resorts each year[1]. A surface level understanding of those numbers might cause a ski-hill operator to conclude "the money I spend on avalanche mitigation is wasted, since deadly avalanches are significantly less likely at ski resorts".
Obviously this is a straw-man. But there are a number of ways to misinterpret the data to arrive at a conclusion you want (to whit, in that time period, 3 times as many skiers than snowboarders died in avalanches, and more than twice as many snowmobiler/snowbikers died in avalanches than snowboarders. But it takes really digging in to realize that a) there are also a lot fewer backcountry snowboarders vs backcountry skiers and b) snowmobiliers are consistently the least avalanche-educated of all backcountry users who actually play on avalanche prone slopes)
You're entirely correct that statistics don't dictate outcomes. But understanding what does dictate outcomes, without bias, can help you change those outcomes.
0. https://avalanche.org/avalanche-accidents/
1. https://coloradosun.com/2023/05/29/national-ski-areas-record...