The dangers of being right

I’ve always been right. It’s a curse really. From a fairly young age, I’ve always been absolutely convinced that when I take up a position it’s because I thought through all the evidence and reached a balanced, logical conclusion. Sometimes that confidence serves me well. Other times…well, let’s just say I have some embarrassing moments.

Chris, you say, what the hell does this have to do with politics, food or wine? Some of you may be aware that I’ve taken up a career in data analytics. I quite fancy myself able to look at the numbers, understand the puzzle, and come up with the right answer. I hope to bring this sensibility to this blog. I see no reason to have another talking head spewing their opinion about whether John Boehner is sincere about equal pay for women. In the end, information wins the debate.  This blog will contain opinion, but if I am successful, it will be about data.

Lies. Damned Lies. And Statistics.

Anyone who has even visited a political debate forum is familiar with this sentiment. It’s easy to see why. When a person with an agenda doesn’t see what they like, they just torture the data with new assumption, or weighting, or biased time period, until they get their desired result.

As an open minded fellow, I of course would never partake in such bias. I am a pure snowflake of open-minded virtue.

Except I’m not.

I started to write a post on the police tactics in Ferguson MO. And…let me just get this out of the way.  Arresting journalists is…oooh…tear gas…grrr…arresting protesters protected by the first…%$@#!…shooting people should be….  Listen, Bashar Al Assad is calling up Obama and saying “not so easy is it, buddy?”  Seriously, this is embarrassing.

And angering. In my anger I thought I could prove the whole militarization of police was pointless anyway. “I’ll find some chart showing police shootings have risen since we started militarizing police.”  And then I thought about it and said, “well, maybe I should also show that police aren’t any safer either, since that’s why they supposedly use this military grade equipment anyway.” In the spirit of open-mindedness of course :-/  So I set out to make my chart. And it was supposed to look like this. 

hypotheticalchart1

Look at all the people dying, and for what?!!!

And then a funny thing happened. The data showed something else.

PoliceDeathsbyYear

That, my friends, is a clear downward trend, even removing the outlying years. Maybe the police are doing this for a reason other than wanting to play soldier. Maybe raiding your house at 3 am with flash bang grenades and SWAT teams actually works. Whether it’s the right thing to do is still open to debate, but to have honest debate you have to have good information.

See the real bias in analytics isn’t torturing the data. That’s actually fairly easy to spot. The real bias is asking the wrong question to begin with, particularly when it fulfills a predetermined biased hypothesis. 

So this I pledge to you, my reader (hopefully readers). Whether you are democrat or republican, I will do my level best to provide you with objective information to inform your opinions. Not based on proving one side or the other right, even when I’ve made up my mind. I will present the evidence and let you decide. 

Now, back to Ferguson. Why is it I can find a list and accounting of police killed in action in a single click, but an hour of searching for victims shot by police turns up nothing more than articles about how journalists are stonewalled if they try to investigate this? *Sigh* I guess governmental accountability will have to come next.

You can bet I’ll have a damned good viewpoint on it, because I’m always right.

2 thoughts on “The dangers of being right

  1. Well the data does correlate to your assertion. For the sake of speed, from the Justice Department, here are robberies, which are in my mind more indicative of overall crime than other crimes.
    1998 447,186
    1999 409,371
    2000 408,016
    2001 423,557
    2002 420,806
    2003 414,235
    2004 401,470
    2005 417,438
    2006 449,246
    2007 447,324
    2008 443,563
    2009 408,742
    2010 369,089
    2011 354,772
    2012 354,522

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s