February 17, 2014
Using Open Data to Predict When You Might Get Your Next Parking Ticket

If you own a car in the city for a long period of time, then you certainly have gotten a ticket or two.  As a data nerd, I’ve always wondered if the officers giving tickets have any sort of pattern to the way they travel around the neighborhood.   After all, it is extremely difficult for a human to be truly random.  

So I was very excited to learn that NYC started sharing Parking Ticket Data recently, and I was able to get my hands on the data set between August 2013 and February 2014.   

My mission was to learn a few things about how ticketing works in my neighborhood, Carroll Gardens.   We’ve got a “No Parking, Street Cleaning, 11:30AM-1:00PM” sign on our block, so I decided to learn everything I could about the way tickets were given during that time period in our neighborhood.

First, I plotted out all the locations of tickets given (assuming an address was given with the ticket).

image

From the look of it, the officers have very good coverage, hitting up all of the streets in our neighborhood that have street cleaning on Thursday at 11:30AM.    So there seems to be no where to hide from tickets.

But what about timing?  I assume the officers have some sort of routine. First, we can see how long after 11:30AM most tickets are issued:

image

What we see here is that officers are honoring the 5 minute grace period put in place by the City Council last 7 months.  We also see that there have been no tickets given in the last 10 minutes of Street Cleaning in the neighborhood in the lat 6 months.  If you are one of those people who sit in their car every Thursday, and you are willing to bet that this pattern will continue, you just saved about 10% of your wait time, (assuming of course that the Street Sweeper has already passed.) 

The problem with the above figure is that it does not tell us anything about where in the neighborhood tickets are given.  I figured the best way to look at this was to color code blocks with the earliest time that a ticket has been given in the last 7 months. 

image

So if you are parked in the Eastern part of the neighborhood, the officers are ticketing immediately after 11:35AM.  But if you are parked in the Western part of the neighborhood, you seem to have about a 20 minute grace period before the tickets come knocking.   (The map shows me that no one has ever gotten a Street Cleaning ticket on our street before Noon, despite the 11:30AM no parking time.  Good to know)

We can also do the opposite, and study when the last ticket is given on each street:

image

The light green streets have never gotten a ticket after 12:02PM, which is only 32 minutes into the 90 minute restriction.  The pattern is a bit harder to see here, but I’d say that the center of the neighborhood seems the most subject to later tickets (in red).

So, after years of suspicion, I’ve finally confirmed that officers are not random.  They ticket in certain patterns.  

Keep in mind, ticketing is important, and serves a public good. But if you have already seen the street sweeper clean your block, and you are wasting gas sitting in your car, this type of analysis might help you decide your next move.  A staggering amount of pollution is added to NYC air each year by people idling in their cars after the Street Cleaner has passed.  

I used this ticket data, this street data, and this police precinct data, along with QGIS,  the MMQGIS plugin (for “Hub Distance”) and Excel Pivot Tables (eek) for this analysis.

     photo tiny_little_mail_icon_zps8e7ba5c7.gif Mailing List
2:39am
  
Filed under: nyc opendata QGIS parking dot maps NYPD 
  1. iquantny posted this