This is probably the best explanation of this phenomenon that I've seen. I'm going to refer to this in a future video for sure. One of the most frustrating "objections" when it comes to alternatives to driving is that many people need to drive, and (today at least) there may be few people who can take advantage of these alternatives in a car-dependent city. It's annoying that this "objection" ignores future travel demand, but it's even more frustrating because you actually only need to take a small number of cars off of the road to make a hugely positive difference to congestion. You didn't quite say it in this language in this video, but you hinted at it: the difference between free-flowing traffic and stop-and-go traffic can literally be just a few cars. Which means that even if, say, a dozen rush hour drivers cycle instead, it can have a massive contribution to better traffic flow. I recently made a video about how driving is awesome in the Netherlands, but I have a feeling that most suburbanites are never going to understand this. Edit: by the way I searched for lazer dix on my channel and he's a car enthusiast who was triggered by my noise video. Some of these people do not appreciate being reminded of what the damage their "hobby" causes to society, and when somebody does, they try to paint them as irrational car-haters. There will always be selfish people who want to stand in the way of progress, but that's why we need to keep fighting for better cities.

"You are the congestion." That might be the single biggest eye-opener here.

This is your best one yet, by far. I'm sure these kinds of videos take significantly more effort than the top 10 lists, but this is the one that made me subscribe. Keep it up!

That first possible solution is more or less the one everyone think will work: "1 more lane will do the trick!" It just makes the cities even worse for pedestrians with massive stroads and these intersections so large that even with pedestrian traffic on crossing it is basically suicide.

When I was a kid in the 90s my central Mexico hometown started building more road infrastructure like overpasses and interchanges, when the city was about 500K inhabitants. My relatives joked that all those new bridges were excessive for our "small" city. Now, 25 years later, the city is over 1M inhabitants, lots and lots of viaducts and underpasses have been built, most of them on the access roads to the city and the ring roads, however when a new underpass is under construction it always gets messy. Public transport was and still is lackluster, so even working class people bought cars to move around. It has just gotten worse. I recall my childhood years of empty roads, they're gone. I don't know if the civil engineers are in bed with the local government or the fact we have the largest Nissan plant on Latin America has fed the car culture.

This is a great explanation, but as a math/cs professor, I have a quibble: this shouldn't be described as exponential growth. The effect is clearly super-linear, but would be better described as quadratic (x^2) than exponential (2^x). The effect here is quadratic because each additional car ADDS time that grows with the number of cars; exponential growth would mean that each additional car MULTIPLIES the time by a constant factor, which would have a much, much bigger cumulative effect. In this context, either quadratic or exponential would be bad, but in other contexts (like disease spread) there's an enormous difference between these growth rates!

Would love to see a video detailing “smart” traffic lights in the US. The one for my street seems incredibly dumb, like it is only programmed to ever save time on the busier cross street rather than consider the actual traffic conditions at the time.

This is also kind of why airport security lines (for example) appear either empty or backed up significantly. When the system is at or just below capacity, all is empty. When it's just 20-30% over capacity, it's 100% lined up.

It has been my experience that cities which have successfully eliminated or seriously reduced traffic congestion have used none of the solutions mentioned in the video. Instead they have implemented measures that make walking, cycling or public transit the fastest, easiest and most convenient way of safely going from A to B. I will give three examples. The first is Groningen in The Netherlands. This uses a neighbourhood zone strategy. This means that people walking, cycling and using public transit can go straight from A to B. But concrete and steel barriers prevent private automobile drivers from going directly from one zone to an adjacent zone. For private automobile drivers to go from one zone to the next, they must drive out of the zone to the surrounding ring road. Then drive around the ring road to the outer edge of the zone in which their destination lies. Then drive in through the zone to their destination. This has the effect of making walking, cycling or public transit the fastest, easiest and most convenient way of going safely from A to B for a very large percentage of trips in Groningen. The second example is Ghent, Belgium. The introduction of its Traffic Circulation Plan put affected streets into three categories. Category 1 Car-free. Category 2: Motor vehicles are restricted to local residents, local merchants and their suppliers, public transit vehicles, taxis and emergency vehicles. Category 3: A neighbourhood zone strategy, similar to the above one for Groningen. Ghent's Traffic Circulation Plan resulted in rmaking walking, cycling or public transit the fastest, easiest and most convenient way of going safely from A to B for a very large percentage of trips in Ghent. The third example is the city of Utrecht in The Netherlands. It has an extensive car-free downtown. The motorway that was built upon the city moat was removed and the moat restored. And on other streets, there has been a steady shift in use that took away space that was previously used by motor vehicles in order to create fantastic cycling infrastructure. Once again, the effect has been to make walking, cycling or public transit the fastest, easiest and most convenient way of going safely from A to B for a very large percentage of trips in Groningen. I strongly disagree with the statement: "Every city has problems with growing traffic congestion." These three cities do not, and neither do many other cities around the world. The one thing that they tend to have in common are measures that make walking, cycling or public transit the fastest, easiest and most convenient way of safely going from A to B. I highly recommend Mark Wagonbuur's blog, "Dutch Cycling." It provides an excellent description of Utrecht and many other places in The Netherlands.

Now I'm curious how a similar analysis for a roundabout intersection plays out.

I used to take an exit in Mountain View (101 N to Shoreline N) that would back up at the light from 8 am to 11 am, and let through ~30 cars a cycle. I once (while siting in the traffic) figured out that every car that decided to use the exit caused a car after them to miss a light at EVERY traffic cycle, which was about 2 minutes long. If a car showed up at 8:02, the next 90 cycles would each have one car shifted to the next cycle, and therefore be delayed by 2 minutes. That works out to 180 minutes of delay PER CAR!!! Congestion is non-intuitive... I usually showed up around 10:30, so I was only delaying a few cycles after me, but this realization strongly encouraged me to take my bike.

I really appreciated the graphic representation, helps explain something that would otherwise be quite abstract

I would love to hear more about congestion pricing! I live in San Diego, and our grand transportation plan just lost its mileage tax provision because the city council cut it due to public opinion/messaging. Now the city's big goals on many things--including multi-modal transportation upgrades--are possibly not paid for in the long run... developing story!

Traffic congestion, like other issues related to city/community life, requires a unified, system based approach. Number of car trips is related to number of dwellings and to work/school/shopping locations and is closely linked with income. (not everyone can afford to own a car) Part of our challenge is that all these factors are viewed/managed by separate agencies/interest groups. Lack of unity is an overall limiting factor as regards mitigation of problems or improvement of the quality of life.

Could you do a video talking about how citizens can help influence public policy in their own cities around issues such as this? Congestion pricing sounds like a great idea, but given how the recent attempt to implement a parking management systems (paid parking) in my own city was received by the population, it sounds like it would be a really hard sell. Not saying specifically to do a video talking about how to get congestion pricing implemented in a city, but how to advocate for positive change in regard to urban transportation. Maybe a top 10 examples of times urbanist advocacy groups made stuff happen in cities in North America and how they did it?

Since I only derive utility from sitting in traffic, this is the best news I have ever gotten. I will watch the rest of your videos while driving around my F150. *I* am the congestion.

Thank you for normalizing that it's OK to be an advocate for active transportation and transit while also thinking that congestion is bad.

i like how tight you edit your episodes. not a word wasted in 12 minutes. Impeccable!

This is one of the factors, but a really significant one. Other factors contributing to exponential growth of traffic also include these: - Land use: as cities sprawl more and more, the average trip length grows. Each trip then contributes to congestion along more portions of the road network. Suppose you have an arterial with signalized intersections every kilometer. Someone living in the inner suburbs 5km away would likely affect up to 5 intersections; but someone living in outer suburbs 10km away would affect up to 10 intersections, if they're heading to the same city center. All of the congestion-causing phenomena effects mentioned in this video manifest at more places, generating gridlock. - Induced traffic demand: believe it or not, people have a surprisingly high tolerance and flexibility for congestion. If a major route has its capacity increased through geometrics (ex. more through lanes, more turning lanes at intersections, grade separations) this reduces travel time in the short-term. The effect of this is people who were using strategies to avoid congestion (alternate routes, alternate trip times, carpooling, alternate destinations, trip avoidance transit as a back-up option when they'd rather drive) will pour into the main route, and them vacating the alternate route attracts more trips elsewhere. In the long-term, the expanded capacity makes long-distance trips more attractive, encouraging sprawl. - Traffic incidents: situations which obstruct traffic can have an outsize effect on congestion, such as car crashes, disabled vehicles on the roadside (or sometimes in the lane if there's not enough shoulder/berm space available), inclement weather and so on, will create localized disruptions to the traffic network. Add up enough of these, or do just one of them on a busy route, and this will propagate congestion big time. There's also a risk of secondary traffic incidents caused by the initial incidents, which may mean the first incident's effects alone can create a prolonged disruption.

Interested to see if this compounding effect is apparent for freeway lane closures, and the effect that more lanes of traffic get better or worse with the same number of lanes closed

Really enjoyed the video, getting into the math is really interesting and I think you could even cover more advanced or complicated math. The graphics were really helpful, how do you make them? I would love to see a video exploring congestion pricing along with induced demand, and the effect (or lack thereof) of transportation alternatives on congestion!

I work as Vertical transportation consultant (read elevator guy) and I can relate to this 100%

It would be amazing to see you tackle all this issues with Cities Skylines. It would be so interesting to watch you build your own city, with public transport, traffic and motorways. You should try it!

I think that's the thing that really gets me about autocentric transportation planning. The only thing that makes it viable is a roughly static number of drivers driving a roughly static distance, and socioeconomically speaking, the justification for pouring all the resources into that type of transportation development is essentially that the economy will always be growing and more people will always be driving more.

So glad that I've found your channel; you have lots of interesting videos related to urban planning, especially on transportation! Can't wait to watch your past videos!

Wow! What a great video. Keep it up man!

I'd love to see a video on how to make congestion pricing work on surface streets.

I love hearing this kind of analysis. Thank you for this peak into your chosen field.

Fantastic video! Great insight! Love the technical outlook on the problem!

As someone who lives in Fort Myers and has taken the intersection shown in the video hundreds of times, the solution shown does not work even though it looks nice. That intersection does not get congested often, but the left turn wait times are heinous for all turn lanes. Usually 3 minutes or more between cycles and so when it does get congested during our tourist season people are left waiting 7 or 10 minutes to turn left onto 41. And therefore it’s my most despised intersection in the city. Funny to see it on here, your best video yet!

I was randomly thinking about that yesterday, and here is the answer. Thank you!

Great, I've been waiting for this one since you teased it a while back!

Great video as usual! I love it when videos reinforce what I'm learning in my classes, and this works through a lot of what we covered in doing a corridor study (things like modeling queues and the impact of implementing tolls with a transit cross-subsidy).

One other thing that can compound the aforementioned inefficiencies is bad weather/road conditions (especially when you're like me and driving in Canadian winters) Since cars take longer to accelerate and have to be slower in general on ice, road capacity can be temporarily reduced and cause the same exponential traffic jams. It can pretty easily turn an otherwise normal intersection into waiting 20+ minutes for a left turn...

I just recently saw your videos. I love it, it’s way in my interests and I’ve subscribed and talk you up regularly. For me, it’s really eye opening that a very few people making different choices can massively improve things. I’ve taken to doing a fair bit of my work transportation on a commuter train with a scooter or EUC to and from the station, and lots of my local errand trips on an ebike. People really don’t recognize what removing traffic lights from your trip accomplishes. Many of my trips are faster than driving. This is in the DC suburbs. Great work!

Wonderful content. I really appreciate the effort in these and look forward to a congestion pricing video

Honestly one of your best videos. Needs more views.

This is great! I wish I could share this on my local Facebook community group and that people would actually get it... instead of constantly demanding local reps try and make their drive smoother, instead of considering alternatives.

Awesome video! I’ve always had trouble explaining these types of congestion scenarios to friends/colleagues but this is the clearest I’ve ever seen it explained. Will definitely be sharing this video to others instead of trying to explain this myself!

Great video and I loved all of your visuals!! Super helpful explanation

Fantastic video. I'll certainly share with folks anytime I have an argument about solving traffic congestion!

The visuals are awesome, thank you! Makes it much more understandable. :)

This was an incredible clear and well explained video!

Great video, very informative. I'd love to see your take on how best to implement congestion pricing and where/when it would be necessary.

Wow, this was an awesome video! I really can tell you spent a lot of time on this, especially the graphics. What did you use to make those? I'd love to see more of these style. Exciting to see this channel growing!

This was awesome! Thank you!

Really enjoyed this video and felt like I learned something!

One of the best explanations of this topic!

High quality video, congrats!

Very clear, important info. Thank you

This is probably one of my favourite of your videos, the mathematical explanation of traffic phenomenon is great

Love the more theoretical videos! You bring awesome information to the table! Keep it up

Good stuff, explaining in intuitive terms how congestion works. Thanks!

Great work!

Good video...but I actually think a better way to illustrate exponential congestion is with on/off ramps as opposed to traffic lights. The problem with a freeway is there is a finite space-time window to enter/exit a freeway via the far right or far left ramp. So a 10 lane freeway doesn't have 5x the capacity of a 2 lane freeway, because the cars can't use the lanes equally. Each lane change creates slowdowns for traffic behind you, and many vehicles have to stay close to the exit lane to get off...which in turn bottlenecks the cars trying to get on. On top of this roads take up a lot of space which results in lower density development and forces longer commutes. A final thought experiment to prove my point...would a 100 lane freeway still suffer traffic jams? I would argue yes, because the ramps and not the lanes are the bottlenecks most often. I actually think a 100 lane freeway would be a good video topic to prove many concepts.

Exceptional video! One thing to also keep in mind for perspective on the inefficiency of cars is that each cycle in this example there are ~7 people moving through the intersection.

Loved the video. I personally have to say I do enjoy these kind of videos more than the top 10s (which are still very interesting) and I really would be delighted to see more theoretical videos in the future. One completely unrelated thing, I was wondering if you could maybe do a video on the public transport system in Karlsruhe (Germany). I find that an very interesting one because 1. The city isnt really big, and it's always interesting to see smaller citys develop public transport systems. 2. It is an early adopter in multy system trains with trains running on the normal tram Tracks and the Deutsche Bahn Tracks even running on the same tracks as the ICE (high speed rail) 3. They just finished a huge infrastructure projects with an downtown city Tunnel which replaces trains running through the pedestrian zone. 4. The network is huge for a city that size with trains running from the city centre to quit remote places in the mountains (Freudenstadt is 80km away) 5. The system grew extremely nicely by using existing tracks which where underused and is completely electrified 6. It kind of shows that quite good service can be given to very remote region's I don't know if you would be interested in this system, but I always found it very interesting especially as an model for city's which may be interested in improving there systems cheaply.

Please keep making these videos this is more addicting then car-centric development in North America.

So glad you made this video! I enjoy all your videos but I would like to see more theoretical videos like this

Great video. Also, I want to add to the voices encouraging you to do more along these lines, in terms of video structure. The top 10 lists are entertaining (and possibly useful for somebody considering a move), but stuff like this is way more interesting and rewarding.

Well expressed the facts behind. Nice Video. I liked. Subscribed to your channel.

Thanks a ton! I also would love to see studies about how likely people are to be on their phones not paying attention and the like at various levels of cogestion - I feel like observing other drivers anecdotally (and myself) suggests that not only is it more likely for there to be a person doing that, but everyone prone to doing that is more likely to do so the more congested (and thus disengaged/frustrated) drivers are

This is the best and most important video on the internet.

What an excellent teaching video. Kudos. 👏🏽

I remember like 2 months ago you were celebrating 1K subscribers, now pushing 15K. Great work!

Commenting for the algorithms! Great content.

Great video.

Excellent video, I write you from one of the most congested cities in the world, Bogota Colombia. I love your math and technical take on this issue, that is the non trivial explanation we all need for the exponential nature of traffic jams. In what concerns solutions, the main ones are to pay for congestion (i.e. to make one-driver cars a very expensive privilege) and to have massive, humane, accesible transit in particular, metro lines. Greetings from Colombia and excellent content!

Good explanation!

Not so sure about congestion pricing, but parking pricing - definetely yes. The ideal situation is where you have a car at home (if you want to own one), but the parking at your destination is expensive enough so that you don't actually use it. Now you might ask why own a car if you wont use it. My experience tells me that public transport is almost impossible in rural areas, forests, minor intercity routes, etc. And I do road trips a lot. At least one kind of lifestyle requres a car even if not using it in the city for everyday transport.

Brilliant presentation. Quantitative analysis always convinces me more than hand-waving qualitative arguments.

I live in London (UK) and the congestion charge definitely helped matters but I don't think it's gone far enough. Congestion is still terrible, despite having a great commuter rail system and the Tube, I still see a lot of single occupancy vehicles driving into the city every day. I think the zone needs to be enlarged and charges increased further. I say this as a driver. I drive into the city (occasionally) for work because I have to (I carry a lot of gear in my car, which I couldn't take on the train or bike), but I see many people who really don't look like they need to be in a car. If the charge was higher, a lot of those would go and things would get better for everyone else who have to use the roads.

Awesome video! Really good way of putting it! 😊

Couple ideas for future videos: 1) Top 10 (or bottom 10) cities in terms of the condition of infrastructure. As a San Francisco resident, I'm consistently shocked by the condition of the roads here and how the city (and state for that matter) completely shut down any time it rains. I feel like other cities handle much worse weather far better. 2) Does it take longer than it used to build big projects in the US, and if so, why? Again, as a CA resident, it seems like even simple projects, like adding a BRT Lane to Van Ness, take an incredible amount of time to complete.

Hey, you made it already! Nice!

Fortunately all of my countrymen understand math, so they are all in with this. The hands down discussion of congestion is a brilliant piece of work. Best math explanation imaginable, concrete and accessible. CGPGrey like brilliance.

Very interesting. It really goes to show how important public transport and bike infrastructure are, when you can just get say 20% of the cars off the road and reduce congestion by half.

What a wonderful explainer and great mathematical analysis - a good example of this is where dolley madison meets lewinsville in tyson’s corner

wow I just learned so much. teach us more plz!!

This is fricking fantastic. Go CityNerd go!

Congestion pricing is super interesting, but it becomes a real problem real quickly if you don't solve the transit problem first.

great video. yes please to the congestion pricing video!

That was mind-blowing! And with such a simple and graphic explanation! The only thing that was missing IMO, was showing us the car pile-up with >160 cars in the bar diagram, that would have been fun!

I love how he's totally bored by his own specialization and as this "you know... whatever" kind of tone. That's authenticity. 👍

Been driving for 40+ years, professionally for 13+ of those years, and I would argue that 60%-80% of traffic problems (particularly on freeways) are due to a small percentage of 'oddballs' - people who don't (usually won't) conform to surrounding traffic. While they're halfway tolerable when congestion is minimal, a single oddball can potentially shut down an entire freeway if traffic is at least half of the design capacity. The first mantra of professional drivers is "go with the flow". It's almost hilarious when some looney flies past me going ±15 MPH faster than the rest of traffic, weaving between lanes (usually without a single turn signal), only for me to pull up alongside them at the next traffic light. The same happens in the opposite direction when someone insists on driving slower than the rest of traffic, particularly when they do so anywhere other than the rightmost lane. Even then, it doesn't take long before other drivers become more prone to take undue risks in order to switch lanes and get around the oddball. When traffic is light, the impact of either extreme is negligible - there are plenty of opportunities to change lanes at minimal risk. But as soon as congestion begins to build, changing lanes (especially if going a different speed than prevailing traffic) becomes exponentially more dangerous. Furthermore, drivers who are overtaken at speed have a tendency to panic and reflexively hit their brakes; seeing brake lights ahead, the next driver hits their brakes (likely harder than the person ahead of them had), etc. You very quickly have people slamming to a complete stop in the middle of the freeway for no reason whatsoever. Which means that someone is likely to switch lanes suddenly, bringing adjacent lanes to a screeching halt as well, expanding to include the entire roadway - sometime even ensnaring oncoming traffic in the process. And all because ONE driver insisted on going 10 MPH slower than everyone else on the freeway.

Great vid

As usual, a great video. If only we had more planners and engineers in the system that grasp how transport works like you do

Your top 10 videos are good... But I really like these explainer videos even more. I especially like connecting the engineering issues with the policy issues.

More of this kind of content please. Also, if you could discuss bus bunching in the same manner, that would be great!

Don’t mind me just paying respect to the algorithm

Thanks!

The idea of congestion pricing sounds very interesting. Is there a country that has a similar policy already in place so we could analyze the effects that it has on transit? Also, I might have a video topic suggestion. I noticed you mentioned Montreal in a few videos, like in the city pairs for high speed rails in North America and high speed rail vs air vs car, Canada edition. I was wondering if you would be interested in doing a video on Montreal's city planning/transit, and what are some of the ways it could be made better. I have a hard time finding articles on that topic. Lastly, I want to thank you for creating this channel, I have been enjoying the content a lot and I'm looking forward to your future videos!

5:00 This was a great visualization.

This is excellent stuff, as a current resident of one and former resident of four other top 11 metro areas in the US, the key to all of my public transit willingness is how big the difference in time is between driving and other modes of transportation. Of course there are days I drive instead of bus/train if I had other errands that day, but traffic congestion factors like this one are a huge influencer in commuter decision making. This leads right into your content on bus only lanes and why metros or commuter trains that aren’t at the mercy of freight are critical. The real question is how much time is saved or at least roughly equal to take public transit (or carpooling) versus driving and being stuck in intersections like this one.

Really shows how car pooling and taking transit makes a difference. Great video. I don't get why all these city planners suggest fees though for driving. That's not really fair as it punishes the poor and those who don't live close to where they need to be.

I love these policy videos! I'll be back for congestion pricing!

That's why you should plan your route to minimize time consuming left turns during rush hour. Not to mention that making a left turn puts you at higher risk of getting involved in an accident. There's a reason why UPS instructs drivers to avoid left turns like the plague.

This is quickly becoming my favorite polo.

Ooh congestion pricing, when you do a video on that i cant wait to watch it!!

more of this please