The UK has "average-speed cameras", which use number plate recognition to see when you enter/exit a long stretch of road, and from that measure your average speed over a significant period. You can't just slow down for the camera. They are extremely effective - in practice traffic does slow down to the posted speed limit. Unpopular as they are, I wonder if they should be deployed widely.
I agree. I can’t say that I love speed cameras, but these do genuinely have a big impact. The other thing they tend to do is even out the traffic - everyone is travelling at a speed close to the vehicles around them. (It’s rare for vehicles to drive a lot slower than the camera-enforced speed - whereas there is a lot of speed variation between vehicles on motorway stretches without cameras.)
So in addition to the reduction in speed, you’ve got smaller speed differences between vehicles and also a much more even and steady traffic flow which can help prevent that traffic jam inducing stop/start behaviour.
I reluctantly think they should be used more widely. (And I say that as a U.K. driver who is not unfamiliar with receiving the odd speeding ticket in the past).
This is such an obvious extension of the concept, but I've never figured out why they aren't used in the jurisdictions I'm familiar with. Privacy related to license plates? Numerical calculations not being valid grounds for incrimination?
In some places (like California), it's illegal to use a calculated average speed to issue tickets, regardless of whether it's done by a person or with cameras.
I live in SF. I feel like speeding is not really a problem in the city proper as the roads are narrow and etc. The "natural" speed limit is the same or lower than the legal speed limit. On the highways I go 80-something mph, which is legally speeding but is the typical "flow of traffic" in the left lane.
That's good because I'm sure they could catch tons of people speeding in the fast lane from their fastrak transponders showing up at consecutive readers too fast.
I sense constituents would be livid and would vote accordingly. I suspect the same backlash around red light cameras and mobile speeding vans would arise and the program would be cancelled after about a decade or so.
It really depends on the municipality: I would vote for this in a heartbeat, and so would a lot of other people in my area. There would also probably be some opposition, but I don't think it would be particularly lopsided against these kinds of cameras.
I think it is one of those things people might broadly support, until they themselves get a speeding ticket and start throwing a fuss. Traffic laws, in particular, are things people want strongly enforced on everyone else.
This feels like doubling down on something that's not super effective.
The way to make traffic slower, is to make road lanes narrower. You can also add curves/trees or overhangs to add shadows to the road, which breaks up the roadway and makes it harder to reason about from a distance.
I don’t understand why adding things that usually make people crash is more beneficial than letting traffic normalize at a speed that may be above the legal limit.
Around me people speed, but you know what’s more dangerous? People who go 55 on the left lane. And trucks that don’t move to the right lane. They are not breaking the law in regards of speed, but everyone that’s going at the normalized pace has to avoid an obstacle.
I say leave the speed limit at 75, put everyone through a more difficult test, enforce lane hierarchy and enforce tickets above 90-95. This equals on a traffic speed of 80-85 that’s safer than adding trees and making narrow lanes
Edit: after posting this comment I come to realize what living in the US has done to me. I immediately assumed that a road with a speed trap is a highway. Which is all I see every time I get on the car anyway. The American dream.
Highways probably shouldn't have speed limits. Honestly, close to nothing should have a speed limit. How fast you can go on a highway should be entirely decided by how good your visibility is. So adding things on the sides of the road could help lower visibility by decreasing the amount of sunlight on roads.
Most people drive at a speed that they deem to be safe. Conditions such as road shape and size, and traffic, are the biggest factors that influence this. If you want people to drive more slowly, you should first figure out why you want that. If you're still determined, then you should figure out why drivers believe those speeds to be safe on that road. Changing the road conditions will change the speed at which people drive.
The fact that perceived safe speed and real safe speed may differ is not relevant. If we want people to drive slower, what's important is the perceived safe speed.
What is relevant is that people have a perceived safe speed, and that this perceived safe speed is able to be influenced. Change the environment and you will change the driver.
Some people are totally oblivious to whether they are safe or not and also reckless with the safety or other road users. This isn't some binary option there are far more factors at play than perceived safety. Even speed limiting cars on dangerous corners could have unforeseen safety implications in some situations.
I'm not suggesting to make poor quality roads. I am suggesting to not put up a 25mph sign on a road wide and straight enough to feel safe going 50mph. A main road near me is such a 2 lane road, 4 lanes widths wide, and is completely straight, but marked 25mph. It's no wonder most people go at least 10-15mph over. I do so unconsciously if I don't pay attention to the speedometer.
The overestimation comes partially from the design of the road - for example wide open roads with many lanes feel safe to speed on. You can correct this perception by tightening lanes or adding trees etc.
You're trying to place blame on drivers of certain types of vehicles, and make a distinction between the behavior of those drivers and the "rest of us", putting them in an "other" category. This is misguided and ineffective.
Even people who drive small cars do so in a way that they believe to be safe. Even people who drove before "modern SUVs" existed did so in a way they believed to be safe. Drivers of SUVs are not the ones risking the lives of others, while the "rest of us" are being safe and responsible.
When car accidents are as common and tragic as they are, the fault isn't with individual drivers. Drivers try to drive safely, and generally do their best under the circumstances. They're just regular people, like all of us. The fault is with the system: road design, vehicle design, incentives (city design, etc), and education.
> You're trying to place blame on drivers of certain types of vehicles, and make a distinction between the behavior of those drivers and the "rest of us", putting them in an "other" category.
The behaviour of SUV drivers and non-SUV drivers can be exactly the same, but because of the sizes and masses involved with SUVs, the same event can cause much more damage when an SUV is involved.
The word "accident" has the connotation that the event(s) couldn't be helped but ((almost) had to) happen. There is growing evidence that this is false:
I'm not a big fan of speed limits and speed traps, but some people just have an mis-calibrated sense of what's safe. Heck, even if you're well intentioned, calibrating to "safe" yourself is very hard because hitting and killing a pedestrian even once in a lifetime is way too much.
The parent comment is acknowledging this. They're also admitting this is true for everyone... not just the people you think are bad drivers (everyone is a bad driver).
Speed traps don't actually enforce/fix this. They just make people unhappy.
If you want a street to feel "safe" for pedestrians, you have to make it feel "unsafe" for drivers, and make them think consciously about how they're driving.
In a local municipality where I used to live, they have a number of double-wide (two lanes in each direction with a median) roads in residential areas, with houses on each side. People often sped on these roads, going 45-55 miles an hour in a 35 zone, causing accidents and hurting pedestrians and cyclists.
So they changed the speed limit from 35 to 25mph.
Now, people routinely speed on these roads, going... 45-55 miles an hour in a 25 zone.
Why don't we just make the speed limit zero? That oughta fix it. What's the alternative?
I guess we could develop roads that are actually sized for the speed you want people to travel at, and thereby not make residential areas 100% road-centric, but I'm told by every civil engineer in America that that's crazy talk for some reason.
What I've seen as part of a solution to that particular problem is to shape/decorate the roads according to the speed you want people travelling at. That's more of the case in Europe than the US. For lower speeds, plant more trees along the road, add more swerve-y features like alternating street parking on the left and right side of the road (as seen in Switzerland); avoid large, long visibility areas, asking people to pay attention and slow down.
Making a 1mi visibility 2-lane each direction road with a 25mph is stupid, as it just won't be followed or if it is enforced - lead to endless frustration. Just use natural features to cause people to behave the way you want.
It's a bit funny (sad) how it works. Like studies in my city show how removing a lane of parked cars and instead using that as a bicycle lane, increases speed of traffic. The added visibility, "feeling" of a wider road, and less chance of a scratch in your side overrules the chance of maiming the cyclists next to you.
Stop building double-wide roads in areas where speed control is desirable.
Start prioritizing humans over cars.
Locally, they've put several such roads on "diets" with good success (usually cutting lane count in half, using the space for bike lanes). Accident rates are down, speeding is down, and throughput only marginally impacted.
I guess we could develop roads that are actually sized for the speed you want people to travel... I'm told by every civil engineer in America that that's crazy talk for some reason.
It's only crazy if rule #1 is maximizing auto throughput, which has historically been true in the US. And it's hard to do retroactively, because other transit options usually aren't sufficient (and cost a fortune to retrofit). But, designing such places from scratch shouldn't be too difficult.
Windy roads. Double wide gives you optionality here: convert them into single lane each way and wind through the double wide, turning the remainder into parks or returning it to the adjacent plot.
Alpine towns in Europe have long ago solved this problem. Those roads have a 90kph speed limit that abruptly changes to 50kph or less for a few hundred meters through a town, then goes back to 90kph.
As you can imagine, people speed. A lot.
So towns started adding a gratuitous sharp turn at the entrance into town. You fly in doing 90kph, slow down for the turn, and are greeted by a 50kph speed limit. You won't speed back up because why would you.
Portuguese rural towns, I've seen, have a similarly brilliant solution – a stoplight outside town. If you approach at speeds above the speed limit, the red light comes on. Forcing you to keep stopping outside very tiny little town. You quickly learn it's faster to just go the speed limit.
Another good solution I've seen from little alleys in San Francisco. The speed limit is whatever, but you really shouldn't drive more than 10mph there. Such nice straight road though ... so they stopped being straight. The 1 lane of driving and 1 lane of parking switch sides every quarter block. Forcing you to slow down.
This is fundamentally exactly what it comes down to. For driving to be safer for people it must become less convenient, more stressful and frustrating for drivers. It seems like most americans think of themselves as drivers first, and reject that.
Some places are doing what they're calling 'road diets' where they narrow the lanes, add a bike lane if it doesn't already exist, or an additional buffer between the car lanes and the bike lanes. Near me, in a few places they also changed some 3-lane roads to 2 lanes.
Is we wanted to remove the revenue aspect, we could eliminate fonts and court costs for traffic violations and just rely on license demerit points. If one accumulates enough demerit points within a given time frame, then their license is suspended. If they're caught driving on a suspended license, then they're arrested and remain in jail until trial.
That's true, but they still assess fines and court costs. A lot of people will actually opt to pay an increased fine to avoid having demerit points on their license. This is because they know that the demerit points will affect their insurance rates and make it more likly their license could be suspended if they commit another offense.
If you remove the option to opt for a fine instead of demerit points, then people won't be able to get out of the consequences of getting demerit points and there would be no income based discrimination. For example, someone who makes a decent living would have no problem paying an increased fine compared to someone who is living paycheck to paycheck.
In my experience, speed traps aren't catching dangerous people, just people that are unaware of the existence of the speed trap.
For example, where I am, there's a hill with a road that winds around it with a 50 MPH speed limit, except right before the bottom and around the turn, suddenly it's 35 MPH before becoming 45 MPH highway. A cop hides around the turn and pulls over people who don't know that their 50 MPH road will suddenly become 35 MPH before they've cleared the hill. Locals know to go slow enough down the hill, as in less than 50 MPH, in order to not get pulled over.
Some municipalities love to pull the trick with a yellow road geometry sign with a yellow limit for the road feature. It’s a speed limit, but not on the traditional white sign with “Speed Limit”.
This entire article is predicated on a local college student who measured speeds after the speed trap happened. That student then concluded that the speed trap was ineffective. Shouldn't he have also measured speeds prior to the speed trap?
Perhaps, if speeding in this case was measured on a continuum. If you look at it as a binary choice, either someone is speeding or not, you can conclude that people going over the speed limit are indeed still speeding.
Agreed. But I would think that a speed trap would slow some people down, not all people. Lots of reasons for that, such as people unaware there was a speed trap there previously, people forgetting, etc.
I know that areas in my town that are well known for speed traps definitely have a lower occurrence of people speeding. On a neighborhood road near my house we had a really bad problem with speeders. Over the course of several months, there were numerous speed traps. Today, there are less speeders than before.
There are small towns on small highways spotted all around me - and some are well known in the area to be "speed trappy" and others are known to be "Wild West" - and the speed limit through all are the same 35 mph or whatever, but only one group really has noticeable slowdowns.
Oh you have soooo much space to add some cycle lanes. Here in the UK we are fighting for shared space. There in the US you _have_ the space. Holy shit make some cycle lanes. US is seriously carbrain damaged.
>If we want drivers to slow down, we don’t ignore human behavior and the 85th percentile speed,” said Strong Towns president Chuck Marohn. “We change the street design to make it feel less safe for drivers. In other words: we change the street design to reflect reality."
I don't get this. Isn't this doing the opposite of reflecting reality and ignoring the 85th percentile speed by modifying the street design to add danger?
>The 85th percentile speed is the speed at or below which 85 percent of the drivers will operate with open roads and favorable conditions. The assumption underlying the 85th percentile speed is that most drivers will operate their vehicle at speeds they perceive to be safe.
Should be: "An individual speed trap has no long-term effect on speeding: anecdote".
However, enforcement as a whole certainly has an effect on the behavior or drivers.
Aint gonna work. Design affects behavior. If the road encourages speeding but the sign says "hey go slower", drivers will end up speeding. This is known for decades. Design roads for slower speed: complexity e.g. chicanes, trees whatnot
Average Speed traps work great. I used to hate those in the UK. With all the toll roads in the USA I'm surprised it hasn't been done already. Eg on the interstate for an hour or two, sensors can tell when you entered and exited & average speed.
According to these studies, section speed cameras can reduce the number of accidents by 30% on average, compared to 20% for conventional speed cameras. Even better: section speed cameras reduce the number of serious accidents by 56%!
Better, the study highlighted a "contagion" effect: the average speed also dropped in the areas located before and after the controlled section.
Finally, the speed seems to have become more homogeneous, and the band changes less frequent in the studied area.
Basically all of my accidents were caused by boredom and losing attention, not by speeding, losing control in corners, etc. It's a complicated topic with a lot of variables, but those are my 2c.
The few times I've driven in italy...
Average speed cams from toll to toll booth would be only half the fines collected for tailgating....at high speed.
Many drivers would defeat the average by blasting along at their regular 180kmh, stopping for a thimble-full of coffee that somehow takes an hour to prepare and two hours to stir three grains of sugar into and then off again to the next toll booth flashing lights at anyone in their way doing less than 200kmh - to make up for lost time.
Come on Meloni - legalise cannabis and average speed cams...all bills paid and no tax hikes.
Most roads in the US are not toll roads. Where there are toll roads there are generally places to exit without counting as leaving - McDonald's often has a drive thru right on the toll road so you can stop for coffee and get right back on without dealing with the tolls. I've also seen gas stations on a toll road. These things are enough that for a significant number of people their average speed is legal despite spending most of their time on the road speeding.
I think the UK also has non-toll exits (I think every country with a developed highway network does?).
The obvious solution is just to put the cameras on the non-toll exits as well. They tend to be spaced generously on US interstates, so it probably wouldn't be a significant expense relative to the overall cost.
Engineering solutions are best. A nice safe design where cars are away from people. Where turns are limited. Where pedestrians only cross at designated, high visibility areas, ideally timed to gaps in intended traffic flow patterns.
Arterial routes are necessary; designate them, build them accordingly. Add a small buffer between them and residential areas.
Time the dang lights on an arterial to encourage the most common flow to drive at the posted speed limits to continue hitting green lights.
Avoid red lights to 'calm' traffic patterns, that just encourages bypasses through whatever alternatives exist, which will be neighborhood streets.
That still sounds like a world designed primarily for cars. It's hard for me to picture it honestly but it doesn't sound pleasant to move through as a person.
A world designed for persons needs a __massive__ overhaul compared to what exists literally everywhere I've seen. Even the nearby major cities are still begrudgingly designed for cars. They lack external interfaces and vehicle storage warehouses sufficient to break from that paradigm.
A city designed for humans would need to be far closer to Azimov's Caves of Steel world, where everything is closer together, beltways and moving sidewalks replace in city bus transit. You'd still need the vehicle layer for emergency services, construction, and bulk product delivery; those vehicles will probably be electric though, or possibly transported by in city dedicated vehicles.
Walt Disney World comes to mind as a larger area intended to have several hubs of primarily human scale transport. EPCOT has the well earned acronym 'Every Person Comes Out Tired' for good reasons; it's too far of a dang walk and the humidity around that lake is murder on a hot day. The inner facade stages are roughly at the sensible limits of human scale, but they're connected either by utility corridors or paths or direct abutment to roadways behind the parks. Even in these _designed_ areas, economics dictates both vehicle and human existence mostly at the same or similar levels and very few over or under passes; where there are it's often exactly two effective layers. Some might have multiple modes of transit. E.G. between EPCOT and Hollywood there's a boat and pedestrian system, which has bridges for the pedestrians, and is crossed in one location by a major road bridge.
There is an "urban redesign" going on - it has started in places like Barcelona (or maybe Amsterdam) and is (slowly) spreading out.
It has three or so basic precepts
- car centric as opposed to people centric is not building economic or pleasent places to be
- denser living is more efficient - even if it's hard to rub shoulders with, you
know, other smelly people. You need more apartments / condos and less gardens.
- if you live dense enough to have viable public transport, then you live dense enough to share other public amenities like libraries, schools, hospitals, streets, toilets cafes. getting along with people you don't like is a social necessity.
It turns out not many people like these implications.
The turning down the dial on car-first urban design has lots of implications - some of it looks like enforcing speed limits (yes this whole article is about enforcing - speed cameras can happily work along vast swathes of modern roads.
Imagine speed cameras on every motorway or highway that automatically pumped out fines as you went over the limits. yes annoying and expensive but also suddenly that 45 minutes commute takes 1.5 hours and the train looks a good option. As does the coach.
Just enforcing the current rules will probably have 50% of the effect desired.
Canberra Australia recently dropped the cbd speed from 50kph to 40kph and setup speed traps the same week. Apparently they made $2mil in revenue.
I think that speeding is bad, but cheap shots by law enforcement drive a deep us vs them mentality that makes people resent road law.
Surely ‘little Timmy died because someone was speeding, don’t kill little Timmy’ appeals to peoples emotions would work better, if law enforcement actually wanted to make a difference..
Well said. I would add they do the same once they jump out of the car, especially when placing speed traps that cause artificial traffic interruptions and possibly accidents. And when they unnecessarily pull over and harass drivers on the highway for driving normal speeds, posing great danger to others.
There will always be people like that, I suspect there would be more impact in a compassion driven campaign over blatant revenue raising bass punishments
History shows you to be wrong. When you put people inside a steel cage where they are protected and cut off from everyone else, something happens with their psychology.
A strong conclusion from a weak study. I do think the way the police intervened probably wasn’t effective- a one shot intense enforcement effort.
A counter example is my kids school. Their was a real problem with people speeding through the school zone. School district police started showing up at least one a week for about 6 months and occasionally after that.
I don’t have quantitative results, but qualitatively speeds have remained down for years afterwards, especially because they still park a police car there every month or two.
I’m not saying speed enforcement is the right solution generally, but done correctly in the right circumstances it does seem like it can lead to persistent change.
90km/h? I've come to the same conclusion. It felt too slow at first but I got used to it. It's safer, more economical, speed cameras aren't a problem and traffic pretty much dissappears. I don't think I can ever go back to driving faster, there are too many benefits.
> This is what’s known as “design speed,” and it may be substantially higher than the legal speed limit. Engineers often use the 85th percentile speed (the speed that 85% of drivers are going at or under) as an indication of design speed. According to Staggs’s findings, on Delta Drive, the 85th percentile speed falls at 45 mph, a full ten miles per hour over the posted speed limit.
They came so close to the solution without saying it. The point of the 85% rule is that's what the speed limit is supposed to be. The reason so many people are speeding is that the speed limit is too low, and the right answer is to raise it.
While I don't disagree with the general notion that ideally the "natural" speed of a road should match the intended target speed, especially in the case of existing infrastructure changing the road design isn't always possible easily and/or quickly, and yet there might still be good reasons for lowering the speed on a particular stretch of road. So I still disagree with a blanket "well, then never post a speed limit that's lower than what people would drive at naturally".
You know what seems to really work to slow people down?
Speed-triggered red lights. Sure, people are more than willing to drive a speed they consider "safe" even if it's above the limit. But most aren't willing to run a red light.
So in many parts of Portugal, that's how they slow you down. If you go above the limit, it triggers a red light. Then you can sit at the red light as other motorists (and even nearby pedestrians) give you a death stare. Thanks for slowing everybody down, dumbass!
Never underestimate the power of an old Portuguese woman (usually dressed in black) staring at you in disgust!
This is the common situation with human brains. We rationalize something because of perceived risk not calculated risk. The article shows people were speeding because of perceived risk but doesn't explain the calculated risk. The undeniable fact is, if car makers, auto safety ministers and everyone else with something in the matter cared about speeding your car wouldn't be able to go over the speed limit set. But, alas, money is at stake.
> The undeniable fact is, if car makers, auto safety ministers and everyone else with something in the matter cared about speeding your car wouldn't be able to go over the speed limit set.
Until the moment when you need to speed away from the danger or you need the extra power to escape the skid. But yeah, let’s just put enforced speed limiters on cars. What could go wrong.
>The undeniable fact is, if car makers, auto safety ministers and everyone else with something in the matter cared about speeding your car wouldn't be able to go over the speed limit set.
How exactly do you do this?
If your answer is GPS, this means that highways will either become very dangerous, or will need to reduce speeds to ~15 mph in many places, because there's tiny access roads parallel to them with low speed limits. GPS is so horribly inaccurate that Google Maps frequently thinks you're driving on a parallel road, so to avoid some people's cars suddenly slamming the brakes on (because their GPS thinks they're on the access road), you'll need to just limit the total speed limit to the lowest speed limit of all roads in the vicinity.
In other words, your suggestion is technically infeasible.
Wow, surprise surprise, a speed trap that is no longer there is no longer working. I bet it would be a different story for a fixed camera speed trap that is well signposted.
I've found those just cause people to brake hard before them and speed up after. If they become well known, they make people speed even more before/after, as it's not expected to have another speed control measure so close to a trap.
In the UK fixed speed cameras are actually painted bright yellow, for the sole purpose of making them more visible to drivers. The reason being is they are positioned in places where you want to force the traffic to slow down for specific safety reasons (merging of two lanes, exits etc). They are not meant to change the behaviour of drivers on the road as a whole.
I don't know about the US, but I think we'll have on every major road in 10 years here in Australia.
They also do this thing where there are two sets of cameras, and if your average speed is above the limit between two points your in trouble. At the moment they target heavy vehicles, but my guess its only a matter of time before they target all vehicles.
Virginia? If not, we do too. I think it's 80mph OR posted limit+20. Either way is an automatic reckless driving ticket (which can be criminal, with arrest and impounded car).
Odds are still in your favor to speed. What was the traffic volume over the course of that week to calculate the odds?
I see more and more in residential areas at 25mph where they install digital signs indicating current speed through highly traveled streets to hopefully slow down drivers but idea if that theory works. I'm assuming it doesn't hurt given how drivers are easily distracted but I'm curious if these sign log any data.
Our (Northern Canadian) municipality stopped doing speed traps near school zones. Instead, they now have an automated radar that just tells you what your speed is.
It seems to have solved the problem. Most people will slow down.
They still do photo radar speed traps on the highway, but it's not really a deterrent or safety thing. More like an automated toll booth for people who insist on speeding.
There's also not anywhere near enough speed limit signs on that road. I'm looking at the photo and I can see any.
Add a whole bunch more of them and it would help. They need one by every single intersection - otherwise people entering the road could travel for quite a distance before realizing there's a low limit on that road.
>Deciding to conduct a speed study, Staggs placed himself in an inconspicuous spot away from school zones or intersections that may have influenced the results. He spent two hours tracking oncoming traffic speed from both directions to collect the study data.
Make it so that at the third ticket in a year, your car is seized and sold off and your car debt becomes non-dischargable with a credit card interest rate.
A few locals learn where they are, slam on the brakes just before them, then accelerate to normal speed afterwards. Thus making the total more dangerous.
If we jump to table #2 we can learn about driver-related causes for accidents:
Recognition Error 41%
inattention
internal and external distractions
inadequate surveillance
Decision Error 33%
Driving too fast for conditions
Too fast for the curve
False assumption of others’ actions
Illegal maneuver
Misjudgment of gap or others’ speed
Performance Error 11%
Overcompensation, poor directional control, etc.
Non-Performance Error 7%
Sleep was the most critical reason
Other 8%
Note that just over 2/3 the critical driver-related reasons had nothing whatsoever to do with speed. If we just add-up the categories that do not include speeding we get 41 + 11 + 7 + 8 = 67%.
Even worse, the 33% group in "Decision Error" does not include speed as an absolute isolated critical reason. Speed always requires context. In this case, the definition only lists road conditions, turns or the proverbial freeway stops and a car plows into the car in front of them.
BTW, not sure where blood alcohol level might be on this table. My guess is that it is a factor in all categories. Again, absolute speed, in isolation, isn't a root cause.
People have been killed by someone driving the posted speed limit while intoxicated.
The point isn't to dismiss speed as a contributor but rather to make an attempt to use facts rather than incorrectly focusing on speed as a the most important contributor or a root cause.
Speed is, by far, not the most serious issue on the road. This is empirically evident to almost anyone with reasonable driving experience and the ability to think critically. If that fails, watch a few of these videos:
From my perspective, driver training and horribly bad/lowest-common-denominator licensing requirements are a huge part of the root cause.
For those outside the US who might not be familiar with this, here's the basics:
In California a 16 year old kid can drive after completing six hours of training with a licensed instructor plus six months and at least fifty hours of practice with any licensed adult (older brother, grandpa, parent, the neighbor, anyone). After this they can take a very simple driving test and obtain a license.
What do I mean by "very simple"?
It does not include any accident avoidance or car control tests at all. Something as simple as driving towards a cone and quickly changing lanes when commanded tells you a lot about the degree of control someone might have. Hard braking and hard acceleration. Driving figure-8's around a set of cones. Driving figure-8's in reverse. Braking and lane changing on wet pavement. Etc.
Not one of these skill tests is included at all. And yet, some of these are absolutely critical for accident avoidance. My family and I almost had a horrific accident on the highway about a year ago. A pickup truck engaged in highway maintenance suddenly drove into the fast lane --where I was driving-- from the construction zone to the left (between the north and south-bound lanes). Where it not for my performance driving training and vehicle control skills this would have been a horrific t-bone accident with our SUV plowing into a pickup truck entering the highway at almost a 90 degree angle and zero speed. An SUV hitting a wall at full speed. Thanks to may awareness, constant strategic positioning in traffic and ability to execute a rapid lane change at speed while maintaining control this became a "holy shit! what was that idiot thinking!" event rather than a trip to the hospital.
As a side note, that's when my wife made the most interesting comment: Now I understand why you insist on taking the kids to race driving school at the track.
This license, after 50 house of substandard training, allows a 16 year old to get on the road, by themselves, driving anything from a 76 HP Mitsubishi Mirage to a 500 HP 200 MPH Corvette and even a 1000+ HP monster if mommy and daddy can afford it.
What could go wrong? Useless training and a 500 HP monster? Heck, even a 150 HP car is dangerous given that starting point. Regulating speed isn't going to do a damn thing to mitigate what could (and usually does) happen with such inadequate training and licensing thresholds. Seriously, why do we allow any driver to get into a 500 HP monster without requiring serious training? The cars are not cheap, which means that requiring they spend $3000 in training and certification should not be an issue.
In terms of skills, licensed instructors are pretty bad. They are the result of the same substandard training. What they teach does not help make better drivers. It's just enough to pass the test, and nothing more.
I can get into how I teach kids to drive during those 50 hours after the certified instructor. It's a bit long, so I'll leave it for a follow-up comment if anyone is interested.
(These are usually deployed near roadworks)