Computer Predicts Executions With Uncanny Accuracy

Found this at http://www.acsblog.org/cat-criminal-law.html



Original source: Using software to model death row outcomes

:sarcasm:

WTF is this all about? I read it and am still not sure.

And using only those factors, along with type of crime, the computer can predict who will get death 90% of the time.

it's not talking about predicting who gets sentenced to death.

It's about who, out of those ALREADY sentenced to death, actually get executed.

also, it's doesn't look like the result is really based on social status.
It's hard to tell without more information but it looks like location was used as a factor.

==============
The clues are already there, Harper says. For instance, the state where someone is convicted is key. California has the largest number of inmates on death row and has executed just 12 people in the last 38 years, whereas Texas has been "an execution machine," he says.
.......
"If you can die in Texas three times faster than you can die in Louisiana, or you happen to have committed a murder in one of the 38 states where the death penalty occurs, then that in itself is an arbitrary thing," Harper says. "You can be one foot over the line in the next state and not be subject to this kind of penalty."

I presume your question "WTF is this all about?" was rhetorical. If death penalty outcomes in capital cases are so closely correlated with social, economic, and racial issues, then it either means that those are arbitrary determinants in the sentencing phase, or that they are highly correlated with the nature of the crimes themselves. That is possible, I suppose, although my understanding is that the incidence and nature of violent crime is pretty evenly distributed across the various demographies in America.

It suggests that courts are very likely to hand out stiffer sentences-- death penalties in this case-- to certain socio-economic groups, and less likely to kill members of other groups, based more on group membership than upon the nature of the crime. This has long been maintained by death row advocates.

Even the most ardent death penalty supporter will admit that there is a Certain Kind of Person who gets sentenced to death. Look at socioeconomic factors and you will see: Poor people get the needle. Rich people get life in prison. Men get the needle. Women get life. Single men get the needle; married men get life. Attractive people don't get executed; ugly people do.

At this point we've got to stop and point out that Karla Faye Tucker breaks all the expected stereotypes--white female, pretty, enough money. I think it was the pickaxe thing.

What the authors of this study have proven is that the Certain Kind of Person thing is so institutionalized that you can basically pick off who these people are without even knowing the details of what they did.

of a civilian in iraq today, and an american soldier.

My computer predicts people will die in every nation today.

My computer predicts more people will be born than die.

My computer predicts that the US prison population will
grow towards 2% of the population as they increase the
petty criminalization to further the new slavery.

My computer predicts that most of today's murdering will
be done with weapons made by americans.

And all that, without even turning it on.

I love that word. Whatever people use it to refer to is usually only "uncanny" to them. :eyes:

it's too predictable. That's the problem.

That factor is often HUGE.

Did you see that study just done that found a defendent does about equally well representing him- herself? That says a lot. If you haven't seen it, I'll try to dig it up, or it might be posted at talkleft.com.

I wonder if it considers location too - that seems to be the real determining factor sometimes.