Right, so, you've hit on something that bothers me about a lot of the arguments from liberals these days, particularly from those that believe their arguments stem from "facts."
Many of these things are not facts so much as statistics.
So you see a group of statistics that are combined to paint a picture. And then that painting is presented as the only rational action a thinking person could do, and anyone choosing different is ignoring facts. They don't know as much about the issue as you?
What if I told you that 100% of people walking past a black person on the street who pretend, at a distance, to know the black person:
"Hey Jerome! Man what's up I haven't seen you in...oh, shit my bad, you're not Jerome! Keep it real man! *Does white person finger point, leg up knee bent* *biggrin* #whitefolks"
Aren't accosted in any way?
Lets say that was a true statistic. Would it change your behavior? No. Your behavior isn't actually based on those statistics you cited. You learned those statistics because of confirmation bias to support your racist actions. You don't feel safe around black people or men or whatever because of deeply ingrained racial prejudices within your culture, and then sought out information to rationalize that.
It isn't a big deal, it doesn't make you a bad person. It is, however, a blatant misuse of the statistics. For exactly the reasons you yourself identify. If you looked into the actual causes of the homicide within the black community, you of course find that it isn't a result of random violence. You are about as likely to be accosted walking down the street as a result of random crime regardless of the race of the perpetrator. In order to have the statistics show this though, you have to take out all sorts of other types of behavior you find among very poor, very urbanized communities in America.
The difficulty comes in just how hard it is to do this sort of analysis. You can read the liberal arguments for these things and really think they are well informed, but then, as a rational thinking person, you think about how they could possibly have some of this information, and the answer is they can't.
Some things we can find, others we can estimate, but the circumstances surrounding why this random black kid got shot today? Dropped off at the ER, died in surgery, we'll never fucking know. The doctor guessed, the nurse guessed, the cop guessed, it is noise.
Our data is not good data.
You throw out bad data.
If you look at the data we have that is good, or at least, reliable (understanding that throwing out bad data in and of itself forces you to doubt any remaining data as, at best, incomplete), it becomes clear that the following facts are actual facts:
1. The causes of poverty are complicated.
2. The causes of generational poverty are even more complicated.
3. The solutions to poverty are, consequently, also complicated.
4. The solutions to generational poverty, consequently, will be the most complicated.
5. Poverty correlates closely with crime (but correlation isn't necessarily causation, and also this is a bullshit statement because crime is far too general a term, more appropriately, the crime poor people do is closely correlated to being poor, no fucking durr).
And so when liberals put together 4 statistics that blame poverty on white people and say the solution is to tax the rich and give money to the poor I am skeptical, if only because that is entirely too simple of a presentation of what is a massively fucked up situation.
Likewise when people on the right show up with no statistics at all and argue from an ideological perspective that it all boils down to choices, I can agree that they aren't wrong! Because they aren't! But it isn't something that's particularly helpful.
Scott Addams has a great post on pizzagate ( I fucking love the name of that gate, and am relieved it's likely a batshit crazy conspiracy theory and not some revelation re:pizza causes cancer ) where he makes this point about confirmation bias:
A mountain of evidence is indistinguishable from confirmation bias. They look exactly the same. As in, all the evidence available to you could point to a thing being true and it still not be fucking true. If you don't think this is true, then you don't understand confirmation bias, and should stop talking.
Liberals are currently swamped with confirmation bias at every level. I can't say the Republicans have that problem because they tend not to look for evidence
at all. I'm not really sure which problem is worse, and I was certainly happier before I realized it in exactly those terms.
So here's what I will say.
No statistic in the world will tell you anything about whether that person walking towards you is dangerous. No statistic is useful in that situation. Relying on them or using them is not useful in any way for improving your outcomes. You hit on what your actual risk calculation is in this situation.
Do I have another option?
Nobody on the other side of the street? Cross the street.
I submit you're probably going to cross the street if the other person is larger than you or outnumbers you. I'm going to get mugged because I don't pay attention anymore.
You make that decision prior to any kind of rational thought process, that's something that has been proven scientifically. There isn't a soul in the world that doesn't make that decision based on all available stimuli instinctively, subconsciously, and then all your thoughts about statistics or trying to understand consciously why you made the decision is literally your brain rationalizing the decision. Your brain works this way, mine works this way, the few that don't are exceptions that prove the rule and likely appear so different to us we've incarcerated/committed them. This is a fact. Interestingly enough, given this absolute biological fact, the condemnation universally leveled against racists is an exercise in holier-than-thou moralizing that has heretofore been reserved for the "religious right." In quite simple terms, we literally are all racist.
Where statistics are useful is in combination with rational thought and good conversations trying to understand our world. They back up already solid arguments. But they are inherently exercises in masturbatory confirmation bias.
http://tylervigen.com/spurious-correlationsThink of statistics like things that can be graphically used to illustrate real-world applications of theoretical ideologies. The more humans are involved in the metric (per 100,000 people for instance) the less useful it is going to be. Failures per 100,000 vehicles is a useful number. Failures per 100,000 people is a fucking dangerous number to make decisions on.
"If my sample size is large enough, the individual variations drop out."
The thing about humans is, those variations simply do not drop out. Any policy implemented based on statistical certainty by definition is trampling on any human outside a standard deviation of the policy targets. So even if it's good policy, it's bad policy. The "standard deviation" when referencing a population of humans is an entire fucking human. Each of our stories is incredibly complicated and unique. It's a "special snowflake" problem. We don't have that with grains of sand or molecules of water. It is arrogance to not acknowledge this. Have some humility with the application of our feeble human tools.
So you look at all this, and you say, well you're just doubting all of everything and won't accept any argument.
And that's not what I'm saying at all. I'm saying that having some facts that support your position doesn't make you right. Science is not about certainty, it is about skepticism. "This is the best theory I have for what is going on, but I am seeking any additional information."
Science.
"White people are to blame for this."
Racism.
It's fairly easy to spot, essentially whenever anyone is dismissive of someone else as "arguing both sides" or "ignoring clear facts" then that's someone who has confused arriving at a conclusion based on facts with science.
This statistic asserts this one thing. Can I find a statistic that asserts the exact opposite? Lets reconcile the two. Lets talk about it. Lets do an experiment.
Left and right have both decided to just declare victory and stop talking, and it's really fucking irritating. It's fine to go forward with your "cross the street to avoid black people" policy. It's OK to do that even knowing everything I just said. What's important is that you maintain the awareness of why you did it, that you're doing it, and that you remain open to conversations about what you're doing and why. If it is defensible, defend it, if it is not, change it.