Reported: 2019 Year In Review
I wanted to take a moment and reflect on the past year for Reported. This post has several sections:
- An assessment of 2019 Goals
- Highlights and lowlights
- A review of the Reported Data in 2019
- Data export
2019 Goals
At the beginning of 2019, I had set an ambitious goal of getting 5x as many reports as 2018 — we were able to do 4x. To do this, I set several key results for the year. Here’s what we achieved.
Get 10 articles (or press) about Reported
- Ben Verde wrote for Streetsblog “The Reported App is Reducing Repeatedly Reckless Taxi Driving”
- Lisette Voytko wrote in Gotham Gazette “How One New Yorker’s App Pushed 311 to Modernize Traffic Violation Reporting (to an Extent)”
- Reported was specifically mentioned in a Toronto City Council Hearing (!)
- Daniel Villaveces wrote Dude, where’s my car? An analysis of bike lane tickets in NYC
- Chris Polansky recorded a podcast for CUNY’s Journalism School which “explores how cyclists use technology to stay safe from reckless drivers”
- There was this solid Reddit post
- Reported was involved in the Beta NYC Mobility for All Abilities Hackathon on National Day of Civic Hacking
- Liam Quigley wrote in AMNY “NYC cyclists call out blocked bike lanes after fatal midtown crash”
Rebuild the app
Ryan took on the task of rebuilding the entire app in React Native. This was a major undertaking and he was able to release a new version of the Android app.
Fully automate the entire Reported process
Behind the scenes of Reported is a complicated patchwork of scripts that submit complaints directly to the 311 website. This year I set out to make these scripts run in a fully automated way so my presence (for example, answering a RECAPTCHA question) is not necessary. I was able to do this and have found more clever ways of continuing to run Reported seamlessly without having to be involved. The KPI (key performance indicator) for this is what I call “Time to submission” or TTS — this is the amount of time that passes from when you click “submit” in the app and when that submission is relayed and re-submitted into the 311 website. The goal is to keep this time down to under 15 minutes and ultimately make it real time.
Give 10 talks about Reported
- Jackson Queens TA meeting
- Presented at the NY School Of Data in a session called “Reported: making it easier to report drivers blocking bike lanes”
- Joseph and I presented at CHEKPEDS’ monthly meeting
- Joseph presented at the BetaNYC “Cyclists with cameras” Meetup (which was an amazing line up of inspiring people working on safe streets). Watch the full video here.
Increase Slack group to 100+
Reported has an invite-only Slack group. This year it exceeded 100 members! If you’re interested in joining for discussion of Reported, TLC hearings, bikes, activism and more, email me at reportedapp@gmail.com and ask for an invite.
Get on The War on Cars Podcast
I didn’t achieve this, but I also didn’t even email them to ask so shame on me.
Create a daily tear off calendar of cars in bike lanes
Yes! I created a daily calendar in which every day has a photo of a car blocking the bike lane on that day. Full description in this post.
Highlights and Lowlights
We are optimistic that congestion pricing passed, a meaningful plan for Complete Streets passed, a range of bike infrastructure has been built, and more. However all of this progress is tainted by the fact that 2019 was an awful, deadly year for cyclists — 29 cyclists were killed — and pedestrians. I urge you to read Streetsblog’s 2019 In Memoriam, which painfully recounts all of these stories.
And so here we are, doing our small part in this fight. Reported is making it easier to hold reckless drivers accountable by submitting your reports to 311 and the TLC.
Highlights
- Record breaking numbers of users on Reported
- 17,700 submissions from 919 users
- Rebuilt the Android app v3.0 and released
- New system where all users now have their own 311 accounts
- Reported submissions account for a third of all complaints the TLC receives and (we estimate) over half of all summons issued by TLC
Lowlights
- A major 311 upgrade over the summer (on a Friday night!) caused our system that submits to 311 to fail for over a week while I updated our code to reflect the changed web interfaces.
- A subsequent change to 311 (that put a “RECAPTCHA” at the end of every submission) meant we had to rebuild the way these submission were handled and then re-submit 1600+ reports so TLC could properly process them. All of this made our backend unstable so it was hard to make any meaningful progress on Reported since we were just trying to keep the ship afloat.
- We still have a major bug in the iOS app where the map doesn’t allow you to update the location. This is due to Google deprecating their Places API. We have not yet had a chance to make the update to the app store.
A review of the Reported Data in 2019
In 2019, 17,700 submissions were made on Reported. (That’s nearly 4x the submissions in 2018!). Here’s the breakdown of how many were submitted to 311 (note: a single complaint could be submitted to both TLC and NYPD as an illegal parking/blocked bike lane complaint):
- 7,729 TLC complaints (2.4x from 2018)
- 11,681 Bike lane blocked complaints submitted to 311
- 3,254 illegal parking complaints submitted to 311
We crunched the data and are excited to share the Reported 2019 Year In Review.
Reported is an app that lets New Yorkers hold dangerous drivers accountable for their behavior behind the wheel, for blocking bike lanes & crosswalks, failing to yield to you when you’re crossing the street, running a red light or stop sign, or parking on a sidewalk.
TLC-licensed vehicles
First, let’s focus on TLC-licensed vehicles (yellow & green taxis and “for hire vehicles” like @Uber & @Lyft). TLC vehicles represented 43% of the reports, down this year because of much more reporting of private vehicles. Reported then submits these to 311/TLC on the user’s behalf. The TLC often issues a summons to the driver.
All For-hire vehicles (Uber, Lyft, Juno, Carmel, Via, Elite, Dial 7, etc) have a base that the car’s license plate is associated with. This is a legacy aspect of the TLC. Everyone knows there is no exclusiveness here — Uber drivers also drive for Lyft, Via, and other services. So while we cannot determine exactly which rideshare service the vehicle was driving for at the time of the observed report, we can provide data on the primary base for the vehicle. It follows that the primary base should be most responsible for the safety of the drivers whose vehicles are associated with that base. Ok, with that out of the way, onto the reveal…
Of the TLC vehicles that were reported, 76% were associated with — you guessed it — @uber. 2.7% are @ridewithvia, 2.5% are @lyft & the remaining ones are older NYC players like Dial7, Elite, Carmel, and Skyline.
76% of all TLC vehicles that were reported have Uber as their primary base. Uber has by far the most opportunity to educate their drivers to never block bike lanes or crosswalks.
Vehicle types:
- 38% FHVs (For Hire Vehicles are Ubers, Lyft, Juno, Carmel, etc, all the rideshare vehicles)
- 6% Yellow taxis
- 47.5% private drivers (any license not considered a taxi, FHV or a NY commercial vehicle)
- 7% NY Commercial vehicles (where the license matches a 00000MX where 0 is any digit and X is any letter like 65454MD)
- 0.2% green taxis
- 114 reports of NYPD vehicles blocking bike lanes (an all too common sight). If we want NYPD to enforce #VisionZero policy, they can start by not blocking bike lanes.
The TLC (@nyctaxi) will issue drivers a summons (like a parking ticket). According to our data, about 60% of drivers will pay this summons. Fines for blocking the bike lane & blocking the crosswalk (the most common complaints) are $100.
If a driver does not respond within 6 weeks, the TLC schedules a hearing at OATH where the driver can plead their case in front of a judge. If the driver doesn’t show up to a scheduled hearing, that generally counts as a guilty plea and they have to pay the fine (a default judgement). Fines are doubled if they are found guilty in a hearing. We estimate (based on internal data) about 30% of drivers heard by a judge are found guilty. (Down from 50% in 2018)
For every 100 drivers issued a summons by the TLC, 60 will plead guilty & pay a ~$100 fine. The remaining 40 will eventually go on to an OATH hearing where 12 (~30%) will be found guilty and pay ~2x the fine ($200).
Each Reported submission to the TLC ultimately yields $84 for the city. In 2019, Reported filed 8,528 TLC complaints on behalf of users. That translates to ~$716,000 in revenue for the city!
Reminder: Reported is a passion project. We do not get paid by users or the city for this project.
According to the 2019 Mayor’s Managerial Report (MMR), the TLC received 26.5k complaints, which led to ~14k cases they prosecuted (eg a summons was issued).
This is an excerpt from that report (emphasis mine):
…Overall, the number of complaints received from the public increased by eight percent in Fiscal 2019 compared to the previous year. Of the 26,532 complaints received, TLC had sufficient evidence to prosecute 13,865 cases, an increase of 12 percent. Due to the high volume of cases that proceed with a TLC prosecution, the average number of days to close a case steadily increased to 95 days in Fiscal 2019. As it became clear that processing times were exceeding the target level, the agency took steps to reduce the processing time for consumer complaints. These efforts were largely a success. By June 2019, the number of days to close a case decreased to 56.3 days, and it is expected to fall within target in early Fiscal 2020.
Reported submission to the TLC account for a third of all TLC complaints.
For context, there are more than 125,000 TLC-licensed vehicles and 185,000 licensed drivers. Reported submissions in 2019 were made for 6,069 unique TLC-licensed vehicles, which seems like a lot but is less than 5% of all vehicles on the road.
So these fines may seem high or unfair, but out of those 6,000 unique TLC licenses reported in 2019, only 7% were reported more than once. We think low recidivism shows that fines are effective and drivers change their behavior.
For other reading on TLC licensed vehicles and drivers, there is a June 2019 Congestion Report that is worth a read and this amazing set of charts that visualize vehicle, driver and trip data from TLC.
Blocking bike lanes
These summons generally document dangerous behaviors where TLC drivers put lives at risk. Many are bike lane blockages and stopping in the bike lane. Stopping in the bike lane “for just a sec”, known as the #deblasiostop, is illegal because it forces cyclists into traffic. This isn’t a minor inconvenience. It’s deadly.
A modest fine can be a memorable way for professional drivers to obey rules that protect the most vulnerable road users. That’s what Reported is all about.
We just want the bike lanes, crosswalks and sidewalks to be clear of cars and for drivers to not threaten our lives.
65% of reports were for blocking a bike lane (an epidemic for painted bike lanes, aka #murderstrips, that render them useless quite often); 18% illegal parking (eg @placardabuse); 10% blocking a crosswalk; the remaining 7% were for driving recklessly, running red lights, failure to yield.
Reported automatically submits your “bike lane blocked” complaints to @nyc311 as “illegal parking” as well as to the TLC. So you may get 2 emails from 311 from 1 submission. We want to make sure bike lane blocking registers in 311 reports that city officials review.
People often ask if they can file complaints for non-FHVs and the answer is yes! In fact, 10,000 reports were for non-TLC vehicles (including private drivers, USPS, UPS, Fedex, NYPD, etc). That’s over 10x what was submitted in 2018. Most of these reports were filed to 311 as illegal parking and bike lane blocking complaints.
Reported makes it super easy for cyclists and pedestrians (in particular) to snap a photo and submit later. 86% of reports had at least 1 photo or video.
We love twitter particularly because it’s a great way to hold drivers accountable in a more public way. Virtually all reports were tweeted and include the hashtag of the police precinct, community board and any speeding or red light tickets the driver has.
Vehicle Violations
Reported automatically looks up the vehicle’s prior violations (speeding and red light tickets, illegal parking, etc) and includes a vehicle “report card” in the tweet. (Similar to @howsmydriving, a fantastic tool.) Here’s an example:
Not so surprisingly, 83% of the vehicles submitted to Reported had at least 1 violation. Most of these drivers have been caught speeding in school zones and blowing through red lights.
Virtually all NYPD Precincts get reports
When we tweet a report, we include the NYPD precinct the incident took place. The hope, of course, is that the police will notice that there is a real problem in their precinct and enforce the laws that protect cyclists and pedestrians, the most vulnerable road users.
This year, we saw 75 NYPD Precincts seeing reports (the NYPD has 77 precincts, so all but 2 were represented). Precincts 13 (Gramercy/Flatiron) and 17 (Midtown East) had 15% of complaints between them.
One simple idea is to get more @NYPDnews officers out of their cars and on bikes so they can experience the kind of road hostility #bikeNYC knows all too well.
@NYPDnews might benefit by getting the majority of their officers out from cars and into their communities on foot and by bike. If they themselves experienced life as vulnerable road users, like most NYCers do, they might understand driver hostility and dangers better.
Community Boards should treat this as community feedback
Community Boards play an important role in street safety as they can work with DOT and support (or block) much needed projects like protected bike lanes or safer intersections. When we tweet a report, we tag the Community Board in which the incident took place.
CBs can take these reports as community feedback, by the most vulnerable road users, that they should be pushing more aggressively for bike infrastructure, traffic calming measures and other improvements, like this project led by CHEKPEDS:
Tied for 10% of all complaints each, @cbsix (East side from 14th st to 59th st) and @CB3Man (Lower East Side) and @manhattancb5 (Central business district, from 14th st to 59th st) should be — and in several cases, are! — pushing for more car-free areas, protected bike lanes and slow zones. Brooklyn CB 2 and @ManhattanBoard4 are the next most common areas for Reported complaints. Again, unlike 2018, there are no clear “winners” in 2019. All community boards are represented here and should see this data as a clear reminder that the current infrastructure is not good enough.
App Platform
Reported is available on iOS, Android and on the web. About 75% of users submit on the iOS app, 20% on the Android app and 5% on the web app.
More eyes on the street
We ❤️ our reporters. More than 900 people saw something dangerous on their walk or bike ride, paused and took the time to submit reports in 2019. That’s nearly 4x the users from 2018!
Reported users may be a small group but we are powerful and dedicated; we’re committed to holding dangerous drivers accountable and shifting driver behavior.
Massive growth, spurred by tragedy
It’s remarkable how much the app has grown this year. We believe this was in direct response to the particularly deadly year in which drivers killed 29 cyclists. In the wake of each of these tragedies, people turned that pain into purpose. Reported provides at least a small way to take action.
- On July 16, we saw the most reports in a single day came, with 168 reports from 49 people.
- On September 27, we saw the most users file a report in a single day, with 54 people making 138 reports.
TLC submissions are decreasing as a percent
The Reported app started off as just a Taxi complaint app which naturally upset taxi drivers who thought the app and its reporters were unfairly only targeting taxis while non-taxis were also blocking bike lanes and should be reported. Reported allows non-taxis to be reported and we have made a few UI changes to make it very clear that we accept all vehicle types (private, taxis, commercial, USPS, NYPD etc).
Over time, reporters have learned this and the trends are clear.
By the summer, TLC vehicles represented less than 50% of all reports.
The chart below shows the percent of reports that were (1) submitted to 311 as a TLC complaint because it was a taxi or FHV, (2) submitted to 311 as an “bike lane blocked” complaint (which can apply to any vehicle type) and (3) submitted to 311 as an “illegal parking” complaint (which can apply to any vehicle type).
Keep reporting
If you ride a bike or walk in NYC and are tired of being bullied by reckless, impatient, selfish drivers, snap a photo and report them. Let’s continue fighting for safer streets together in 2020!
2018 Year in Review
For additional reading, check out the 2018 Reported Year in review part 1 and part 2. I created 2018’s YIR as a Tweet-storm, thinking that was a more digestible way to share the data, but have decided that Twitter is not the best form of distribution of this information.
Full export of data
Finally, if you would like to do your own analysis or just take a peek at the data treasure trove Reported users have generated, I’ve made that possible here. This is a full export of all Reported data, in a public Google sheet. There are 3 tabs: (1) Reports, which is every unique report, one per line, (2) Daily Aggregate, which sums up the counts per day of things like reports and users submitting and (3) Changelog which provides the date of the export and some notes.
This is a work in progress. I’d like to export the data several times per year to make it more timely. If you have specific requests, email reportedapp@gmail.com.
Onward, to a safer and more bike friendly 2020! 🚀
