“Give Me Your Hunches”

Posted on 24 March 2011 by


Last week, when Eric Olson joined Dave and me on the Police Led Intelligence Podcast, we were discussing ways in which agencies could better prepare patrol officers to gather and report on the rich intelligence they naturally come by on a day-to-day basis.

Eric was talking about how he thinks about ways to automate thought processes which, at first glance, are highly mysterious – such as the way some cops can stand on a highway overpass and select the smuggler car out of hundreds that are driving by.

That guy probably couldn’t tell you how he knows. But he knows, and this kind of information, which is often so difficult to articulate, is exactly the kind of thing we hope to help analysts articulate.

Dave mentioned that this process to which Eric referred was merely what cops called a “hunch.” This is, of course, the most storied power that cops have, but it’s a double-edged sword.

In the movies, the hunch is that which guides the grizzled-yet-amiable, maverick-yet-ultimately-law-abiding-or-at-least-justice-seeking, divorced-yet-loving-and-misunderstood street cop to the ultimate solution to the baffling crime pattern. And in the movies, the captain who prototypically castigates the cop for listening to the hunch is saying something like,

Listen, Frisco. I don’t wanna know what you think, I wanna know what you can prove. Don’t tell me about shivers down your spine, or hairs standing on end, or coply intuition, or tea-leaves…get me proof! You got 24 hours!

Give Me Your Hunches
We’re here to suggest that, in fact, the properly-articulated hunch may be among the most valuable pieces of intelligence an officer can provide on a shift. Anyone who has gone on a ride-along can tell you that it’s astounding what patrol officers can see – that guy, running behind him? The inspection sticker out-of-date by one month on the car passing the opposite direction, two lanes over? That guy entering the building, who looks as if he doesn’t live in that building?

All those kinds of things are examples of something that patrol officers see because they have worked, since the first day they hit the streets under a Field Training Officer, to become finely-tuned anomaly detection devices.

If it’s weird, the officer will likely notice it.

Similarly, analysts need information about what is weird. It’s that old game of “One of these things is not like the other.”

But as I said in the podcast, formalizing this is problematic, because the human intelligence assets are the range of people cops come in contact with every day, and the officers don’t have a mission more specific than “protect and serve:” which is pretty darn broad. They don’t know what can be important, but they know that the cost of wrong is very high. So they must constantly be processing the data they receive, which is bandwidth intensive. They’re casting a wide net. And at the end of this chain sits the analyst seeking rich nuggets of data pertaining to suspicious activities, drug and gang violence, terrorism … all priorities of each department.

The metric looks like this:

data==(officers) x (contacts) x (minutes per contact)

To make matters worse, each contact provides information of unknown and unknowable value prior to harvest.

So here’s something actionable, cheap, low-effort and potentially high-reward: buy a case of 3×5″ index cards and give out five cards to each patrol officer.

Let’s ask each patrol officer to take a minute when he notices something that he thinks is important on the next shift.

And let’s see what we get back. And then go from there.

Any takers?