Hacker News Reader

As a casual cyclist commuter, I can coast probably 100 meters after stopping pedaling. It's not the slowing that is hard though, it's the anxiety of needing to exert a lot of effort to get back up to speed. From that angle, I'd posit that cyclists may be doing this out of laziness, while scooter riders don't have that same mental/physical cost to fear when they stop.

That said, I don't run lights, and am just as baffled how some cyclists will barrel through a red light at intersections without even slowing down...

I don't understand that, as a cyclist I find it much more relaxing to just follow the rules and stop at a red light, if there is a car already waiting I wait behind it as he got there first. I would be much more anxious running a red light and doing a track stand trying to see if the intersection is free of other vehicles as I see many fools doing.

Exactly this.

I'm cycling for a bit of outside time and exercise in my day. If I have to stop at a light, then fine, it's a bit of extra exercise and outside time.

I'm not there for the excitement of some kind of death race with stressed out drivers.

Could it be your physical fitness? When I was physically unfit I always dreaded the work of building up speed on the bike.

Shift into an easier gear as you stop. Makes it significantly easier to accelerate. Stopping at lights is no longer "anxiety-inducing".

Bonus points, use a bike with internal geared hub and shift down after you stop.

Getting up to speed from a stop feels super slow and takes a bunch of energy. Combine that with cycling being a game of energy conservation, and the result is that slowing down to a stop feels like an utter defeat in a video game.

With that mentality, one can see why so many cyclists disregard safety around intersections and stop signs - mentally, the feeling of doing an unsafe thing doesn't suck as much as the feeling of "losing" (even if it's just an emotion not grounded in any logic or reason). Regardless of whether it's logical or not, there's also this: https://en.wikipedia.org/wiki/Idaho_stop

Idaho stop: What a great sounding idea. Would like to see some data showing there is no negative accident/health effect to have some solid arguments.

There’s been some pack of the napkin research on this, and the conclusion is that a cyclist stopping at every intersection and then speeding up to 20mph vs a cyclist that doesn’t stop and just cruises at 20mph burn 500W vs 100W. Unfortunately I can’t find a link right now but based on my own experience I’d say that is spot on. Stopping and resuming at every stop sign or stop light is far more taxing, so much so that many people wouldn’t even be able to do it and maintain a better than walking reasonable pace through a city.

I think the reasonable alternative then is to walk, not to dangerously run through intersections when it's not your turn.

Edit: I will admit that I rarely come to a complete stop at a stop sign. It definitely is a considerable inconvenience to get back up to speed, but a lot of safety precautions are inconvenient.

The problem with walking is the same issue. Unless the lights are timed for a walking pace you will almost always have to wait 3-8 minutes at a light for your turn. Walking in a place with a lot of crosswalks is excruciatingly slow. Especially since a lot of crosswalk signals will only add you in on the next cycle of green rather than let you cross on a green right away like a car or a bicycle.

I made an estimate and while it was higher than I expected, the increase in power is not anywhere near what you mentioned.

I'm using the value of Cd * A from here (0.6 m^3): https://www.sheldonbrown.com/rinard/aero/formulas.html

That page also indicates that the vast majority of the energy expenditures on a bike are due to drag.

The fraction of energy needed to get up to speed divided by the energy expended due to drag works out to: (mbike + mhuman) / (Lblock * rho * Cd * A) where mbike is the mass of the bike, mhuman is the mass of the human, Lblock is the length between stops, rho is the mass density of the air, Cd is the drag coefficient, and A is the projected area. I can post the derivation if anyone is interested.

Assuming an 8 kg bike, 90 kg human, 200 m between stops, and an air density of 1.2 kg/m^3, stopping adds 68% more energy.

This is dependent on the weight. If I crunch the numbers for my own weight, stopping adds about 56% more energy.

So, this is higher than I expected, but I don't think it's so much higher that it justifies running red lights.

I don't think any amount of effort justifies running red lights.

I think this calculation is basically right, but might be a bit worse for the average commuting cyclist. Many commuting bikes are heavy and badly maintained, so rolling resistance is likely to be a larger factor. Also at the low (max ~25km/h) speed of many commuting cyclist rolling resistance is almost as significant as air resistance. [1] indicates that up to around 17km/h it is as big a factor as drag, but that is on a well maintained bike.

[1] https://www.gribble.org/cycling/power_v_speed.html

20 mph sounds way over the top for average city cycling. Anecdotally most people running stoplights seem to not exceed 12 mph, as I frequently overtake them shortly after getting my green light.

I was thinking that too, but it turns out that the percentage increase in energy is independent of the speed in a fairly simple model: https://news.ycombinator.com/item?id=22257938

(20 mph is really fast for a bike, eh?)

Anyway, 400W to avoid getting creamed by a car seems cheap to me.

Eat a Clif bar; stop at red lights.

Goddamnit.

The worry is that you won't be able to output 400W. My approach is to reduce speed earlier and hope the green comes before I'm forced to stop.

That's a good idea. What I want is some gadget that caches my momentum (in a little compressed air cylinder or something) for the stop.

"Caching the momentum" is essentially what "regenerative breaking" does.

An engineering student designed something like that, using a flywheel: https://www.treehugger.com/bikes/engineering-student-builds-...

But if you want something commercially available, you have to get an e-bike with that feature, and those models seem quite pricey.

I have an ebike w/ regenerative breaking (a Bionix conversion kit, $1500 w/ labor) but I can't say it seems to do much. But then again, with no other non-regen bike to compare to I can't be sure. It doesn't recover anything like the charge it spends going up a hill (with assisted pedaling) when you're coming back down, but that's to be expected, eh? Thermodynamics and whatnot.

The flywheel thing is exactly it, except I think it would be safer and lighter to use compressed air, also stores longer than a flywheel.

With something like that you'd eliminate most of the work penalty for brief stops.

Compressing air isn't a very efficient operation, pumps usually have leak, spring compression, friction, and heat losses.

> pack of the napkin

That's back of the napkin. Bone apple tea!

As a cyclist, there is a much bigger energy expense in starting and stopping than on an e-scooter. With the bike you have to dismount (or track stand) when you stop then get back on the bike and get your momentum going. With the scooter you just stop, you are already standing up. Similarly getting going again on the e-scooter you just push off and you are moving.

I don't see a problem with bikes running red lights. It's not that much different from crossing any other intersection where cross traffic isn't stopping. Generally I avoid traffic signals on my commute, but there is one place on my commute where I almost always run the signal. In this particular intersection, it's less dangerous crossing by myself against the signal than crossing with the signal next to a bunch of cars trying to pass me.

The fundamental problem is in most places traffic controls are designed around the needs (and safety) of people in cars. Bikes and pedestrians are usually an afterthought.

Many veteran cyclists I know treat red lights as stop signs. It can be safer and better for the flow of traffic to get a head start on the cars stopped at the red light rather than jockeying with them for road space out of the gate when the light turns green.

Treating red lights as stop signs is sensible and safe - I'm even fine with not stopping if you have clear line of sight at all angles. One thing car owners like to forget is that cars make you mostly blind to your surroundings. While it feels incredibly unsafe to blow a red in a car, it's not the same on a bike- you can see so much more.

Cyclists who speed towards or through an intersection when they can't see what might come into conflict with their path are as stupid as drivers who run reds.

I've been cycling for transportation for over 10 years. I don't know anyone who advocates for running reds who has rode a similar amount of time or longer. I have heard that argument before, but I think it underestimates the damage to the reputation of cyclists involved. Also, more experienced cyclists tend to take the lane at intersections, which would be a better way to avoid jockeying for space.

Most of the cyclists I’m referring to have been cycling since childhood, so 15-20 years of experience, a lot of it in cities. Taking up the lane at the intersection is unrealistic in a lot of city scenarios, nobody I know advocates for getting in front of cars in the crosswalk or waiting in the middle of a long line of stopped cars at a red light.

> Taking up the lane at the intersection is unrealistic in a lot of city scenarios, nobody I know advocates for getting in front of cars in the crosswalk or waiting in the middle of a long line of stopped cars at a red light.

My understanding is that taking the lane at an intersection is often recommended, and some cities create special "bike box" infrastructure to encourage cyclists to do that, though few do.

I wouldn't recommend blocking the crosswalk, but my experience getting in a long line of cars is good.

I'm definitely not a veteran cyclist and almost always end up on the other side of the intersection faster than the first car. Cars have a lot of inertia. I don't believe going on red is necessary to be in front.

From what I’ve heard (UK study reported about a decade ago in local news), drivers and cyclists jump red lights at about the same rate. However, because cyclists cycle at cycling speed and drivers drive at driving speed, any given cyclists sees fewer cyclists and more drivers than a driver, and vice versa, so both groups see evidence of more red light jumping by the other group than by their own group.

Anecdotal observation is that drivers might run the red light for the first few seconds. But cyclists will run the red the moment they arrive at he light, independent of when it turned red. These are two very distinct types of running a red light.

I’ve seen both behaviours in both groups. But the point is worth caring about in any quantitative study.

I wouldn't buy that.

I sat in a corner desk right on top of a major central business district intersection. Cars never ran the light, it's too busy and complex. Cyclists did reckless, stupid shit all of the time, running lights, cutting across the diagonal on the wrong side of the road, etc.

From my POV, drivers are more predictably dumb, and dangerous when they perceive low risk, for example, they run red lights in a low traffic neighborhood and are awful about a number of things. Cyclists are less consistent and perceive risk differently. Younger people who perceive themselves as invincible in particular.

End of the day, we should have better dedicated infrastructure for bicycles and other things.

Having cycled in both the US and the UK, I think you both are right.

I have never seen a cyclist in the UK run a red light at a complex intersection (often only running the 'quiet' side of a T-junction, or an empty junction with a 20MPH speed limit and no cars in sight). Also worth noting is that the roads in UK cities are much narrower, I would guess 40-60, and often have less lanes and often have a bike lane of some sort.

In the US (California) I have seen all the crazy stuff. People running red through traffic, people doing a U-Turn, people going the wrong way, people crossing to go the wrong way, people blowing through a red light whilst on their phone.

However for cars, the UK has more stop lights that are for cars giving way to pedestrians, i.e. there is no junction. I have sometimes seen cars run a red on these when they don't notice/care. However in the US the red lights are at junctions, and are therefore more dangerous for cars (getting T-boned) I have however seen more people run these still.

I'm both a driver and a cyclist and I've seen orders of magnitude more cyclists running red lights than cars.

In Illinois, bicyclists are allowed to proceed through a red light if the the fails to turn green after 120 seconds and after they've yielded to traffic [1].

I can understand the spirit of the law but in practice cyclists interpret this as "it's ok to red lights if I don't see any oncoming traffic". This turns out to be extremely dangerous because a situation can change in seconds especially when high velocity metal objects are concerned. Blindspots are always a problem.

This issue has salience with me because a cyclist who ran a red light at the intersection near my house got hit by a car that was following the rules of the road. There was a dramatic video on Youtube of the cyclist getting hit and flying off his bike. I don't know if he survived.

As both a cyclist and a driver, I can't emphasize enough: cyclists need to follow the rules of the road. It is not ok to run red lights.

[1] https://www.cyberdriveillinois.com/publications/pdf_publicat...

That guy is clearly in the wrong and I'm surprised how intense your reaction was. There's got to be something else in that.

Yes, I was angry, and I'm sorry about that. My reaction was emotional. But I think it's perfectly reasonable to be angry when someone almost causes you serious harm. I think how unseriously the scooter rider took the incident is much worse.

If you don't think it was that close, consider that the wide angle lens on my helmet camera makes the incident appear farther away than it actually was. Out of all the incidents I've had in the past year, this was one of the worst in terms of potential for serious injury. If I was a foot or two closer, there would have been a collision.

The main thing I learned from this incident was to not focus too much on people turning from my right. That's where my attention was approaching the intersection. I need to also consider the possibility that someone will run the light.

You did good in everything else, you honked and the guy snapped out of his stupidity and you gave him a good blasting. The guy was brain dead to not even look, had headphones on whilst riding (???) and ran a red light. Moron.

I don't know how bad it would have been if he hit you. A broken bone and a lot of scraped off skin. For some people that's a risk worth taking. Does that mean falling behind in work, insane medical bills and stress on the family? I don't know if cycling is worth it for everybody. It's not my place to say any of those guesses about what it means to you in real life. I'm curious what the stressors are for cyclists. Being in a car crash at least gives you the peace of mind that in a reasonable accident the car will take all the damage and you can walk away and get a rental.

I really like personal vehicles and I think we need to move in that direction, pollution aside (incl break pad dust etc.) cars (or worse, the trucks that are so popular with Americans) don’t scale. It’s really not possible to move accross DC in a car in a reasonable amount of time (even when you’re moving between suburbs and not just DC proper.) Hopefully all that is enough to convince you how important I think personal vehicles (including bicycles) are.

All that being said: DO YOU GUYS ACTUALLY WANT TO DIE? I’ve never seen any group of people ignore traffic laws and common sense more than cyclists. They consistently ignore signals and four way stops and ride on two lane highways with high speed limits, no shoulders, and blind curves.

Yes the roads need to be shared (and personally I think cars should be banned before things like bicycles and scooters) but dang people need to behave.

Meh, likely you're just used to the ways that cars fail to obey the law. Every day on my driving or biking commute I see cars failing to signal, rolling through stop signs "because you can't see from the actual stop sign", speeding, honking because they're mad at someone, directing traffic, illegally turning, failing to yield right of way, etc.

Straight up running red lights or stop signs does seem like it is overrepresented by cyclists, but I really don't they they're overrepresented for generic 'breaking the law'.

You seem to be conflating law with rules of the road. The rules of the road are just how (mostly) everyone (mostly) agrees to behave (most of the time). Basically it's social consensus. The end result is similar but not the same thing as the law (which makes sense since laws reflect norms). Cars mostly follow the socially agreed upon rules of the road regardless of the law.

Cyclists and to a lesser extent scooters throw a wrench into things because we as a society haven't yet had time to reach a consensus on what the norms for these classes of traffic are and how they should behave so people fall back to the law.

This is why people get all "but the law" when cyclists run reds but don't care when cars roll stops or speed (under appropriate circumstances). We've established that the latter is ok but haven't yet reached a consensus on if the former is to be expected as part of normal traffic flow.

As a pedestrian in San Francisco I have much worse experiences with cyclists than I do with cars, despite cars outnumbering cyclists, mostly because of cyclists running reds and taking shortcuts by getting on crosswalks and sidewalks.

Unfortunately, the 25 pedestrians killed by cars every year in San Francisco are not here to share their anecdata.

Sure, a collision with a car is more likely to kill me than a collision with a bike, but I do think I am significantly more likely to be in a collision with a bicyclist than a car. Cars do not often drive up on the sidewalk or drive 4 inches away from you at 20mph on a crosswalk

Bicyclists also don’t routinely drive into buildings and destroy them. And this is all while working with a tenth of the public space we give cars!

Same can be said in Chicago.

I find many (other) cyclists' safety arguments don't place enough emphasis on the danger of annoying someone in a car. They can hurt you intentionally and they can get away with it.

I used to get that mad. haha. I'm so jaded at this point that I just swerve and go on my way. It's hard on your psyche to fly into a rage a couple times every week.

I'd say one main reason why cyclists want to avoid stopping is that with a well-adjusted bike seat you should only barely be able to reach the ground with your foot when stopping making the whole process quite awkward.

I sincerely doubt that's the main or even minor reason. With a well adjusted bike seat you likely would not be able to reach the ground at all. You want your leg to be at near full extension when you are at the bottom of your pedal stroke.

To solve this problem, just get off the seat when you stop.

i'm surprised by how many people on this forum have not stopped indefinitely at a red light on a bike. most lights i've come across are not calibrated for bikes so this isn't a rare occurrence.

many states have a "dead red"[1] law for when lights are not operational. which permits the cyclist to proceed with caution after stopping.

[1] https://en.wikipedia.org/wiki/Traffic_light#%22Dead_Red%22_l...

Yes, cyclists should not run red lights. But we need to improve cycling infrastructure a lot. Right now there is a lot of incentive to break the rules in small ways here and there, because the situation is pretty bad.

There are quite a few places in my city, where I have no idea what the legal way to get from one side of a large road to the other. Sure you can always act like either a car or a pedestrian, but that is either dangerous or slow. For example, I find it quite unfair, that many left turns require me to either ride between cars or stop twice.

Some bike paths just look like the planners reserved space at the roadsides and simply skipped coming up with a solution at the intersection. And that is frankly disappointing.

  cyclists are more likely to run red lights than scooter riders

E-bikes are lumped in with bicycles, so those stats became meaningless to me at that point.

On flat terrain a bicycle can coast a very long way without peddling. If your bike comes to a stop quickly then you have something out of alignment or loose.

Or a cheap bike

Okay, I think I was unclear. I was thinking 100 meters or longer. I made a slight edit to my comment. My point was that for starting and stopping to be a major contributor, getting up to speed had to be comparable in energy to drag. Turns out that getting up to speed took more energy than I expected when I did the math in a later comment: https://news.ycombinator.com/item?id=22257938

Personally I like coming to a complete stop because it gives me the opportunity to practice accelerating from a stand still.

your reaction to the guy crossing the street is hilarious.

Speeding is dangerous but running a red light is straight up reckless.

I think drivers speeding is considerably more dangerous than cyclists running red lights.

In my experience, the vast majority of the time cyclists run red lights, they check both directions before going and there are not conflicts. The examples shown in the article are fairly rare.

Plus, the danger from a cyclist running a red light is mostly to the cyclist, while the danger to a driver speeding is to everyone nearby on the road.

Add on top of that the fact that there are many more drivers on the road than cyclists. The net effect is that speeding is a much bigger safety problem in total than cyclists running red lights.

(Note that I'm not endorsing running red lights.)

the danger from a cyclist running a red light is mostly to the cyclist

It's also dangerous to pedestrians. I see cyclists run red lights with pedestrians in the crosswalk almost every day, and many of them seem to assume that people will stop walking and yield to them. If somebody doesn't notice the cyclist then this can easily lead to a collision.

Fair point. That hasn't been my experience in pedestrian-heavy areas, though I figure this varies from location to location.

Without wanting to endorse it: I live in a Southeast Asian country where people run red lights all the time. As in, I see it happen probably a half dozen times every time I drive.

I've never seen an accident arise from running a red light.

Which isn't to say that I think running a red light is "safe" but I think the context in which one runs a red light is probably a bigger factor than blanket denunciations imply.

In America, running a red light usually means something like "while driving a 4,000 pound car at 40 miles per hour [6.4 million mv^2], the driver blows through the red light due to complete unawareness of the signal."

Whereas where I live it is more like "a guy on a scooter (360 pounds net), slows down (somewhat), sees a gap in the traffic and zips through at 15-18mph." [116,000 mv^2; the car is 55x worse]