Cyclist Injuries near Queensborough bridge 2012-2018
About a year ago I stopped taking the subway for work. I did this because I couldn't get motivated to get to the gym, and it was the "Summer of Hell" of 2017. I was totally disgusted with the 30 to 90 minute transit delays turning my crowded-yet-mercifully-short subway commute into the world's worst carnival ride.
When I turned in my Metrocard for a Citi-bike membership, I was stunned by the cost savings. Bike share membership was much cheaper, more efficient, and often more pleasant than taking the subway. Even after I enrolled in a gym near work just to use the shower (getting over the bridge can be strenuous), my monthly costs were less than $50.
Biking, jogging and walking was often faster, healthier, and more pleasant way to get to work. However, my nerve was shaken when I saw a bad bicycle accident. I had seen a bikers take spills before. Once I was riding to Queens from Manhattan after work on a summer day, and one biker coming towards me wiped out and landed face first on the blacktop. He wasn't wearing a helmet. It was a bad fall but he got back up and started riding.
The worst accident I saw was when one cyclist was overtaking another on the down slope from Queens into Manhattan. He changed lanes too early and clipped the other cyclist's front wheel. I'm not exactly sure how, but the two got entangled and he either broke/sprained/or dislocated his knee. It looked really bad. I called EMS and me, one cyclist and the two EMT had to lift him over the guardrail separating the bike lane from the road. After that I didn't cycle to work anymore. It seemed dangerous and made me nervous.
It's been a while since and the memory is less vivid than it was. So after some nightmarish subway rides to work, I'm considering a return to a subway-free commute. So to assuage my fears, I gathered some data to assess just how dangerous it is to bike across the bridge.
My first thought was to use EMS data of the area from NYC open data website. However, there are several problems with the dataset. One problem is the data set is too broad and doesn't have any geocodes (e.g. latitude and longitude) to pinpoint where an incident occurs. The closest proximity I could get is zipcode. Also, the dataset is massive. For the 5 zip codes in which the Queensborough bridge spans, there were over 100,000 incidents.
Instead, I found a data set of NYC motor vehicle collisions which includes number of cyclists injured or killed in a given incident.
The data set I decided to use is limited to motor vehicle collisions, but it's easier to narrow down to the specific geographical area in which I'm interested. The data may not include cyclist on cyclist accidents (I didn't see any). However, those types of injuries are relatively minor compared to a car on cyclist collision. Motor vehicle accidents involving cyclists are more likely to cause major injuries.
That said, here's an analysis of cyclist injuries from motor vehicle collisions.
The first page has two charts. The first shows collisions by day of week and time of day. I broke the day into 5 parts, each around four or five hours long because this seemed easier to read than bars for each hour. The second chart shows collisions by month and year. Each small bar represents a month and the months are clustered by year.
The second page has several pie charts showing some detail of the cause of collision and the vehicle type involved in the collision. Bikes are excluded from the list of vehicles. There is also a heat-map showing number of incidents by borough at different times of day. I had initially expected Rush hours to have more incidents since I assumed most bike riders were commuters. Since most collisions occur at midday, it made me think that a good portion of the cyclists may be people who use their bike for work, e.g. deliveries.
The third page contains an interactive google map with layers of collision data with Geographic coordinates (latitude, longitude). This was the reason I chose this data set, so I could plot the coordinates on a map and see where the hot-spots were. Each color on the map indicates how frequently a collision occurred there. Clicking the top left icon will expand the legend panel which includes the meaning of each icon color. Hovering your mouse over a given icon in the legend will highlight them in the map.
After doing this analysis, I was surprised in a few ways. I had initially expected more injuries from cars, considering my experience with how crazy rush hour commuting is on a bicycle. It turns out there are around 100 per year, which is about one every three days. Considering my anecdotal data about how many cyclists I see commuting every morning, this seems pretty low.
Another surprise was the number of collisions that happen during at midday. I would think that gridlocked car traffic would reduce the risk to cyclists since car speeds would be lower. The numbers indicate that the number of collisions is still relatively high but this may be due to increased bicycle traffic during those times, e.g. lunch deliveries.
Some deficiencies in the analysis are:
When I turned in my Metrocard for a Citi-bike membership, I was stunned by the cost savings. Bike share membership was much cheaper, more efficient, and often more pleasant than taking the subway. Even after I enrolled in a gym near work just to use the shower (getting over the bridge can be strenuous), my monthly costs were less than $50.
Biking, jogging and walking was often faster, healthier, and more pleasant way to get to work. However, my nerve was shaken when I saw a bad bicycle accident. I had seen a bikers take spills before. Once I was riding to Queens from Manhattan after work on a summer day, and one biker coming towards me wiped out and landed face first on the blacktop. He wasn't wearing a helmet. It was a bad fall but he got back up and started riding.
The worst accident I saw was when one cyclist was overtaking another on the down slope from Queens into Manhattan. He changed lanes too early and clipped the other cyclist's front wheel. I'm not exactly sure how, but the two got entangled and he either broke/sprained/or dislocated his knee. It looked really bad. I called EMS and me, one cyclist and the two EMT had to lift him over the guardrail separating the bike lane from the road. After that I didn't cycle to work anymore. It seemed dangerous and made me nervous.
It's been a while since and the memory is less vivid than it was. So after some nightmarish subway rides to work, I'm considering a return to a subway-free commute. So to assuage my fears, I gathered some data to assess just how dangerous it is to bike across the bridge.
My first thought was to use EMS data of the area from NYC open data website. However, there are several problems with the dataset. One problem is the data set is too broad and doesn't have any geocodes (e.g. latitude and longitude) to pinpoint where an incident occurs. The closest proximity I could get is zipcode. Also, the dataset is massive. For the 5 zip codes in which the Queensborough bridge spans, there were over 100,000 incidents.
Instead, I found a data set of NYC motor vehicle collisions which includes number of cyclists injured or killed in a given incident.
The data set I decided to use is limited to motor vehicle collisions, but it's easier to narrow down to the specific geographical area in which I'm interested. The data may not include cyclist on cyclist accidents (I didn't see any). However, those types of injuries are relatively minor compared to a car on cyclist collision. Motor vehicle accidents involving cyclists are more likely to cause major injuries.
That said, here's an analysis of cyclist injuries from motor vehicle collisions.
The first page has two charts. The first shows collisions by day of week and time of day. I broke the day into 5 parts, each around four or five hours long because this seemed easier to read than bars for each hour. The second chart shows collisions by month and year. Each small bar represents a month and the months are clustered by year.
The second page has several pie charts showing some detail of the cause of collision and the vehicle type involved in the collision. Bikes are excluded from the list of vehicles. There is also a heat-map showing number of incidents by borough at different times of day. I had initially expected Rush hours to have more incidents since I assumed most bike riders were commuters. Since most collisions occur at midday, it made me think that a good portion of the cyclists may be people who use their bike for work, e.g. deliveries.
After doing this analysis, I was surprised in a few ways. I had initially expected more injuries from cars, considering my experience with how crazy rush hour commuting is on a bicycle. It turns out there are around 100 per year, which is about one every three days. Considering my anecdotal data about how many cyclists I see commuting every morning, this seems pretty low.
Another surprise was the number of collisions that happen during at midday. I would think that gridlocked car traffic would reduce the risk to cyclists since car speeds would be lower. The numbers indicate that the number of collisions is still relatively high but this may be due to increased bicycle traffic during those times, e.g. lunch deliveries.
Some deficiencies in the analysis are:
- The data only includes collisions with motor vehicles. To my knowledge, it does not include bicycles colliding with other bicycles.
- The data only includes incidents that are reported to the NYPD. This may be lower than the true number if some people choose not to report it. For example, a delivery person who gets hit by a car and chooses to continue working rather than wait for police.
Despite these shortcomings, I think it's a pretty good picture of the risks involved with bicycle riding near the Queens-borough bridge. I feel much better about walking and riding over the bridge after seeing the data.
Great job for publishing such a nice article. Your article isn’t only useful but it is additionally really informative. Thank you because you have been willing to share information with us. Gym Near Kelambakkam
ReplyDeleteGreat blog ! I am impressed with suggestions of author.
ReplyDeleteDefect Inspection Gold Coast