49

We have had a number of roundabouts installed near us in South Western PA in the past decade, and a quick internet search shows other states doing so as well. This seems to be an almost universally frustrating trend with those I've spoken to.

A quick online search leads to a number of pages such as this one from PennDoT: Road Design > Roundabouts. These pages extoll the safety of roundabouts and quote statistics about how much safer they are than intersections. This sounds great and, if true, I'm on board.

However, my personal experience with roundabouts, even ten years after installation of the first one in our area, involve regular instances of:

  • drivers switch lanes inside the roundabout at the last minute without looking because they are unclear which lane to be in, or how to properly get to the inner lane
  • drivers fail to yield when entering the roundabout since they're used to most cars flowing straight
  • drivers slow nearly to a stop in confusion while they try to understand how to use the roundabout (hopefully this will disappear eventually)

I experience at least one of the above almost every visit to a roundabout. There have also been at least 2 accidents that I've heard of in our local roundabout, despite it being a medium-low traffic road. This makes me question whether the statistics being presented are cherry picked or presented out of context to appear to support an untrue premise.

Are roundabouts actually safer than traditional intersections, and I'm just seeing issues caused by the adoption of the new and unfamiliar? Or are these statistics false or misleading?

EDIT

Thank you, all, for the wonderful answers.

Since the question has been asked a number of times, I wanted to post a link to the particular roundabout that prompted my question: https://www.google.com/maps/@40.2038168,-80.1263559,151m/data=!3m1!1e3

Ironically, the day after accepting the answer about them being safer, I had to slam my brakes to avoid being hit by an entering car that ignored the yield and just continued through at 30+ mph. The universe is messing with me again...

13
  • 2
    Reminder: Comments are for clarifying and improving the question, not discussing your opinions on traffic around the world.
    – Oddthinking
    Sep 9 at 5:25
  • 35
    Can you define "safer"? I will gladly accept 100 more crashes with property damage to save 1 life so te me a huge increase in crashes with a reduction in deadly crashes falls under "safer". If you just count any kind of collision or accident, no matter their consequences the result will be completely different. Moreover as @Oddthinking is saying there is no single type of "intersectin" and also there is no single type of roundabout, so it's likely that some kind of roundabouts are worse and others are extremely better (at least coming from the EU where they are way more common...)
    – GACy20
    Sep 9 at 7:38
  • 6
    Welcome Hot Network Question Readers: Here at Skeptics.SE, we actively DO NOT WANT your opinions or your anecdotes in the comments. Comments are for clarifying the question. Answer boxes are for fully-referenced answers.
    – Oddthinking
    Sep 10 at 6:31
  • 1
    'Safer per person-mile' is not the same as 'safer per person-hour' is not the same as 'safer per person-trip' is not the same as 'safer per intersection-hour'. (For an illustration of this: changing a road to have an enforced speed limit of 1 furlong/fortnight would likely make it safer per person-hour, but not per person-mile, and would likely cause many people to avoid said road, which would make it safer per intersection-hour but which could have mixed effects on safety per person-trip depending on the alternatives people end up using.) Could you please clarify which of these you mean?
    – TLW
    Sep 10 at 21:52
  • 2
    When I see a roundabout I don't like, I get in it and go round and round and round like Chevy Chase in one of his Vacation movies.... As far as I can tell, it's perfectly legal to perform a denial of service attack in a roundabout.
    – Michael
    Sep 11 at 19:17

5 Answers 5

94

Here is a study of traffic safety at roundabouts in Minnesota. In this study they looked at all roundabouts in Minnesota, and look at the recorded crash data at these sites, comparing the data before the construction of the roundabouts to that after the construction.

The findings of this technical report both confirm the increased safety of roundabouts and your anecdotal report of increased incidents at round abouts. In particular they found

  • a significant decrease in the number of serious accidents that were fatal (-86%) or lead to serious injury (-83%)
  • an increase in the number of "property damage only" crashes, which were up 50%.
  • a slight increase (+15%) in the overall number of crashes.

The explanation for this comes from the fact that the nature of crashes at roundabouts is very different from that at traditional crossings. Roundabouts experience a decrease (-35%) in the number of right angle collisions (typically the most dangerous crashes), while experiencing in increase of the number of sideswipe (+681%), and "ran-off road" incidents (+272% or +373% depending on the side of the road).

EDIT: In the comments @Oddthinking makes the fair point that this data mainly shows that placing a roundabout can make a dangerous intersection safer, but it does not necessarily tell us whether a roundabout is a safer option than other improvements one could make to an intersection. The report does actually contain data comparing different intersection types (this time using data from the 2015 Minnesota Intersection Toolkit, based on 5 years of crash data from 2011-2015.

Traffic Control Device Crash Rate Fatal and Serious Injury Crash Rate
Urban Thru-Stop 0.18 0.33
Rural Thru-Stop 0.25 1.05
Signal - Low Volume/Low Speed 0.52 0.42
All-Way Stop 0.35 0.57
Single Lane Roundabout 0.32 0.31
Signal - High Volume/Low Speed 0.70 0.76
Signal - High Volume/High Speed 0.45 0.48
Unbalanced Roundabout 0.76 0.15
Dual Lane Roundabout 2.18 0.00
All Roundabouts 0.51 0.24

And here's a color-coded version:

The same table as above, but with color-coding to show the outliers

While roundabouts (particularly dual lane roundabouts) show higher overall crash rates than other intersection types, they show lower rates of fatal and serious injury crashes. This is consistent with the other findings. Of course, there could still be some sort of selection bias due to the fact that certain solutions are built in certain situations based on where they would best fit.


Source: Leuer, Derek, and Safety and Technology Minnesota. Dept. of Transportation. Office of Traffic. “A Study of the Traffic Safety at Roundabouts in Minnesota,” October 30, 2017.

36
  • 21
    A source of bias in this approach is that the intersections were not randomly selected, but very likely selected for upgrade because of a history of accidents or congestion. If roundabouts hadn't been an option, they may have had lights or other safety measures installed. The comparison should be roundabouts versus similar-priced alternatives (e.g. lights), not roundabouts versus unimproved dangerous intersections.
    – Oddthinking
    Sep 9 at 5:21
  • 5
    @barbecue That's the problem. They say nothing about the "control group", making the report unconvincing. At the end they quote the crash rates for alternatives (lights) from "the 2015 Minnesota Intersection Toolkit", but - again - any decision to upgrade an intersection is predictably very far from random.
    – kubanczyk
    Sep 9 at 7:38
  • 10
    "While roundabouts (particularly dual lane roundabouts) show higher overall crash rates than other intersection types" - the table shows single-lane roundabouts having lower overall crash rates than every intersection type except thru-stops, which doesn't quite match what you wrote here.
    – NotThatGuy
    Sep 9 at 12:05
  • 8
    Maybe this is a cultural difference in terminology, but having not been able to find an explanation online - what is a "Thru-Stop"?
    – Zibbobz
    Sep 9 at 12:32
  • 7
    Also the "Dual Lane Roundabout" stats look like they're derived from much smaller numbers of journeys than everything else: the incident rate is abnormally high yet barely affects the low "all roundabouts" average (and it's striking there hasn't been a single fatal accident yet). Sounds to me like there's only one or two "Dual Lane Roundabout"s in the area, recently built, and local drivers are still getting used to it. If true, their stats are almost certainly not statistically significant or stable, and I'd advise not drawing conclusions from those until there's more data. Sep 9 at 13:18
46

Germany started to reintroduce roundabouts in 2000, so quite a lot of research has been done, and there is plenty of experience.

The German "Bundesanstalt für Straßenwesen", federal road research institute, conducted a study about roundabout safety in 2021. The results are published in https://bast.opus.hbz-nrw.de/opus45-bast/frontdoor/deliver/index/docId/2512/file/V343_BF_Gesamt.pdf. The document is in German, but has an English abstract and summary on pages 3-9.

They state "Literature consistently finds a higher level of safety at roundabouts compared to other types of intersections on rural roads." and "severity of accidents is lower at rural roundabouts then [sic] at other types or rural intersections", then continue with evaluating the main dangers and constructive measures to avoid them.

The UDV (Unfallforschung der Versicherer), an accident research organization funded by Insurers, came to a similar conclusion, but stated that there are some problems with cyclists, in a report compiled in 2012 (https://www.udv.de/udv/themen/sicherheit-inneroertlicher-kreisverkehre-75470, in German). The research was specifically done to evaluate roundabouts in towns, as they said

Kleine Kreisverkehrsplätze gelten außerorts zu Recht sowohl in Deutschland als auch im Ausland als ausgesprochen sichere Verkehrsanlagen. Unterschiedliche Untersuchungen ... kommen jedoch zu dem Ergebnis, dass innerorts Sicherheitsprobleme mit dem Radverkehr auftreten können.

"Small roundabouts outside of towns are considered, rightfully, in Germany and abroad, as very safe traffic installations. However, various examinations conclude that in towns there may be safety problems with cyclist traffic" (my translation).

They analyzed all accident counts in roundabouts in the state of Nordrhein-Westfalen and each individual accident at 100 selected roundabouts in all of Germany, and they used video to analyze driver behavior at 10 roundabouts. (As they are run by insurers, they have access to detailed accident reports whenever insurers are involved.) As a result, they came to the conclusion that

Es stellte sich heraus, dass Kreisverkehre auch innerorts ein höheres Sicherheitsniveau erreichen als signalisierte oder unsignalisierte Knotenpunkte. Während Kraftfahrzeuge und Fußgänger vom Sicherheitsgewinn deutlich profitieren, kann allerdings die Sicherheit für den Radverkehr nicht immer verbessert werden.

"The conclusion was that, also inside towns, roundabouts reach a higher level of safety compared to crossings (with or without traffic lights). While motorized vehicles and pedestrians profit clearly from the safety gains, the safety for cyclists cannot always be increased".

So, both studies, conducted by organizations whose main function is road safety, come to the conclusion that

  • Literature commonly says roundabouts are safer than other types of crossings
  • their own research confirms this
  • there are rules and best practices that have to be followed to eliminate common causes of accidents; if those are omitted, safety may be reduced.
27
  • 17
    The problem with using data from Germany is that in general the level of driving proficiency and discipline in Germany is much higher than in the US. It could be that roundabouts are safer when used by competent road users, but unsafer when the road users do not know what to do at a roundabout.
    – TimRias
    Sep 9 at 7:48
  • 22
    @TimRias It's also possible that the level of driving proficiency in Germany, which you claim to be higher (source?), is due to road users having used roundabouts for twenty years. In places where roundabouts were introduced more recently, users may still be in the learning curve. In particular, newer generations of road users may have an advantage over people used to "just driving through", for example. The US may reach similar results in a decade.
    – Simone
    Sep 9 at 7:57
  • 18
    @Simone Every driver in Germany has at least 12 * 45 minutes of driving instruction by a professional driving instructor (this is the legal minimum, the typical time is probably double that) and has to pass a practical driving test (which is not entirely trivial to pass). So the minimum level of driving lessons is much higher in Germany than in the US as far as I understand the US system.
    – Mad Scientist
    Sep 9 at 8:22
  • 21
    I think comparing driver "quality" is a red herring when looking at roundabouts. European drivers encounter them all the time wherever they go, so they're routine and everyone knows what to do. American drivers are no less capable of learning them, but even if they are experienced on roundabouts, they still have to deal with other drivers that are not because roundabouts are rare. That makes for more unpredictable situations and thus crashes.
    – Cyrus
    Sep 9 at 10:01
  • 17
    @fredsbend I don't think you're "supposed" to cut people off in a traffic circle/roundabout. My understanding is that you should wait for a safe gap before entering the circle.
    – Onyz
    Sep 9 at 14:14
5

When talking about "safety", you could mean one of many metrics. Some common ones include:

  1. Deaths per day per intersection.
  2. Accidents per day per intersection.
  3. Deaths per car entering the intersection.
  4. Accidents per car entering the intersection.
  5. Deaths per person-mile traveled.
  6. Accidents per person-mile traveled.
  7. Deaths per person-hour traveling.
  8. Accidents per person-hour traveling.
  9. Deaths per person-trip.
  10. Accidents per person-trip.
  11. Person-lifetimes per person-mile traveled.
  12. Person-lifetimes per person-trip.

Now consider three alternatives:

Intersection A lets through 2 vehicles per unit of time, on average. Each vehicle takes ten units of time and distance for its trip. One in a million vehicles crash; 1 in ten vehicle crashes result in a fatality.

Intersection B lets through 1 vehicle per unit of time on average. Each vehicle takes a hundred units of time and four units of distance for its trip. Two in a million vehicles crash; one in forty vehicle crashes result in a fatality.

Intersection C... no-one uses intersection C, because a vehicle would require a literal lifetime to go through it, and so people have changed their routes, instead taking a thousand-unit of time and distance to avoid it.

Now. Which of these is safest?

Intersection C wins for deaths and accidents per day per intersection, fairly trivially (no-one ever uses it). Intersection C loses for person-lifetimes per person-trip or person-mile, again fairly trivially.

Intersection A wins on accidents per car entering the intersection. Intersection B wins on deaths per car entering the intersection.

Intersection A wins on deaths per person-mile traveled. Intersection B wins on deaths per person-hours traveled.

Given all of the above, it would be misleading to say that intersection B was safer than intersection A, even if you provided (correct) backing statistics showing that intersection B wins on deaths per car entering the intersection. It would also be misleading to say that intersection B was less safe than intersection A, even if you provided (correct) backing statistics showing that intersection A wins on deaths per person-mile traveled.

So. Coming back to your question:

Are roundabouts actually safer than traditional intersections, and I'm just seeing issues caused by the adoption of the new and unfamiliar? Or are these statistics false or misleading?

Roundabouts are safer by some metrics than traditional intersections; these statistics are arguably misleading for the claim that roundabouts are safer in general.

Roundabouts tend to be superior w.r.t. deaths per day and deaths per vehicle, and you'll see many statistics that show this (e.g. your link, or https://rosap.ntl.bts.gov/view/dot/35084 , etc.)

Table one of https://dr.lib.iastate.edu/server/api/core/bitstreams/6d1e271e-1a94-4b98-afbb-95f62e805832/content shows 5.6 fatal accidents per billion vehicles entering the intersection for high-speed signals, and 1.9 fatal accidents per billion vehicles entering the intersection for high-speed roundabouts. This is a difference of 3.7 fatal accidents per billion vehicles in favor of roundabouts.

Roundabout accident rates are often higher per vehicle, especially in unbalanced and multilane roundabouts. (e.g. the Minnesota paper with "Many of the [unbalanced roundabout] sites have seen an increase in the frequency of crashes, and the overall total crash rates. However, unbalanced roundabouts are achieving a noticeable reduction in fatal, serious injury, and other injury crashes.").

It is difficult to directly find comparative statistics on the other metrics for roundabouts versus signaled intersections. You generally have to combine multiple statistics and hope you haven't introduced bias in the process.

There are situations where roundabouts are slower than alternatives. (There are also situations where a roundabout is faster than alternatives.) (To disambiguate: 'slower' in the sense of 'taking longer for a specific vehicle to get through the intersection'.) Consider the intersection of a high-traffic high-speed road and a low-traffic road, for instance. A controlled intersection can be set up with traffic sensors, where the light normally stays green for the high-traffic road, only changing when a car actually shows up on the low-traffic road. This results in most cars not having to slow down, and very little average delay. Meanwhile, a roundabout will always require cars to slow down and turn, regardless of how much traffic is on the sidestreet. This shows up as increased average intersection control delays.

Compare table 3 of https://onlinepubs.trb.org/Onlinepubs/circulars/ec083/42_Kaubpaper.pdf (the first paper I could find that included estimated average delay for both a signal and corresponding roundabout, if you're wondering). In this case, moving from a signal to a roundabout increases estimated average delay from 11.9/15.0 seconds to 20.6/33.4 seconds, or an increase of 8.7/18.4 seconds in AM/PM, respectively.

Assume as a ballpark guess that average vehicle occupancy is roughly the same as average fatalities in a fatal crash, and so 3.7 fatal accidents/billion translates to ~3.7 fatalities per billion vehicles entering the intersection. (I'd love to hear about better numbers here.) Meanwhile, US life expectancy in 2019 was 78.8 years, or ~700 thousand hours. 3.7 person-lifetimes per billion vehicles entering the intersection translates to about 9.2 seconds.

You'll note that 8.7 < 9.2, but 18.4 > 9.2. So in this case, the roundabout won in person-lifetimes per transit in the morning, but lost in the afternoon.

(There are several issues with this analysis. The average age of people driving is greater than zero. On the other hand, time spent in transit is arguably not completely wasted. On the other other hand, average vehicle occupancy is more than one. On the other other other hand, average fatalities per fatal crash is also more than one.)

4

This could depend on your definition of 'safer' - fewer accidents, or fewer injuries/deaths.

If you watch YooToob's 'dash cam' compilations, you quickly spot that Americans can't use traffic lights & Brits can't use roundabouts. What you will also see is the enormous difference in how serious the results are.

By their very nature, many of these videos use the most spectacular footage they can find, which makes them a biased view; but of the ones who concentrate on US footage, the sheer number & carnage resulting from 40mph+ T-bones & 'turning left against a straight-through light' incidents is remarkable. [No-one asked for my opinion, but I actually think this is a result of having the traffic lights so high up, well out of the natural sightline.]

Comparatively, the British roundabout incidents are often someone changing lanes foolishly, or entering when unsafe. These usually result in either some heavy braking & horn blowing, or some dented metalwork. They don't end in lamp posts & telegraph poles down, cars spun right off the road in tangles of twisted metal.

By my definition, that already makes roundabouts 'safer'. There are fewer deaths and serious injuries.

One thing we have in the UK is a lane system, where if you start in the correct lane, you will naturally be fed out into the correct exit. This can make a huge difference in big roundabouts, where you can't even see your exit from the entrance. This one is small enough that you can see how the lanes work from a drone photo. This only qualifies as a medium-large roundabout. We have them a whole lot bigger.*

If you start in the right-hand lane [up at the top of the pic] to turn right, then without crossing any lanes except in the initial entry phase [which is also helped by shorter white markings], then you are fed out to the correct exit.

enter image description here

Not all large roundabouts have this system. The ones that don't are far more hazardous to negotiate, even for one practised at the British structure.

From the OP's PennDoT link - this is how not to lane a roundabout.

enter image description here

There are too many areas of 'guesswork' if both lanes are used simultaneously. With this method you only need one idiot in the outside lane trying to turn left for it to quickly become chaos. [This is reminiscent of what I've seen in Spain, where half the drivers stay in the outside lane, no matter where their exit might be. You need eyes in the back of your head to negotiate one of those. It is one of those times where the 'bad' behaviour actually becomes safer, and so often you'll see everybody going round the outside, leaving the inside empty & effectively treating it as though it had no lanes & was just a single-track system.

In the UK a modern roundabout this small wouldn't be laned at all which makes everyone treat it as a single lane & follow rather than try to drive side-by side… mimicking the Spanish system, but officially. If it was laned, as older roundabouts tended to be, then it's a single dotted line right the way round, but with no other messages. That's why they're doing away with lanes on the smaller ones, it removes a lot of dangerous guesswork.

*Bigger roundabouts - I've never measured it but this one might be a mile round, under & over two motorways. We don't really do clover-leaf junctions so much in the UK. This one has great laning right round, except for one exit which makes it a nightmare. [On the right in this pic, but the pic is of an earlier laning structure, since replaced & easier to see] btw, they must have taken this photo at 8am on a Sunday, it's never this quiet ;)

enter image description here

3
  • 1
    The safety advantage of a roundabout is that it almost completely eliminates the possibility of a head-on collision. Almost.
    – Criggie
    Sep 11 at 19:45
  • 3
    If you are going to show pictures of mad roundabouts, how can you leave out swindon?
    – User65535
    Sep 12 at 11:54
  • Aside from the fact I was in danger already of being too anecdotal, I had thought to include Hemel [which is the only 6-er, the others all seem to just have 5 'satellites'], but decided against it. I went round the one at Denham only this morning, going to & from Pinewood ;)
    – Tetsujin
    Sep 12 at 12:04
4

First, make sure you're talking about the same form of intersection. The safest form is called a Modern Roundabout, which was developed in the UK in the 60's. This form was first constructed in the US in 1990.

Large diameter traffic circles and rotaries are not built anymore because the large diameters enable speeds that are unsafe and cause congestion.

Many people confuse other and older styles of circular intersections with modern roundabouts. High speed, east coast rotaries, large multi-lane traffic circles (Arc D’Triomphe, Dupont Circle), and small neighborhood traffic circles are not modern roundabouts and UK 'roundabouts' are not the same as North American 'modern roundabouts'. The Brits even call a merry-go-round a kid’s roundabout.

What is, and is not, a modern roundabout: FHWA: https://safety.fhwa.dot.gov/intersection/roundabouts/ UMass video: https://www.umass.edu/newsoffice/article/new-umass-transportation-center-video WA DOT: http://www.youtube.com/watch?v=vsCoI7lERGE

The magic roundabout in Swindon, UK, is a form of Ring Junction.

Modern roundabouts are the safest form of intersection in the world - the intersection type with the lowest risk of fatal or serious injury crashes - (much more so than comparable signals). Modern roundabouts require a change in speed and alter the geometry of one of the most dangerous parts of the system - intersections. The reduction in speed to about 20 mph and sideswipe geometry mean that, when a crash does happen at a modern roundabout, you might need a tow truck, but rarely an ambulance. Visit the Insurance Institute for Highway Safety or FHWA for modern roundabout FAQs and safety facts.

Modern Roundabouts are proven safer than signals (FHWA): https://safety.fhwa.dot.gov/provencountermeasures/roundabouts/ https://safety.fhwa.dot.gov/intersection/innovative/roundabouts/

A Review of Evidence-Based Traffic Engineering Measures Designed to Reduce Pedestrian–Motor Vehicle Crashes: https://ajph.aphapublications.org/doi/full/10.2105/AJPH.93.9.1456

Freakonomics story: https://freakonomics.com/podcast/roundabouts/

Since their introduction to the US in 1990, with over 8,000 modern roundabouts in place, there have only been about 190 total reported fatal crashes at modern roundabouts. Compare that to over 30,000 annual roadway deaths in the US and about 8,000 fatal crashes per year at intersections. https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/810682

Regarding 'safety' I would advocate for only considering crashes that cause injury, and exclude property damage only crashes, however many statistics used for comparisons count all reported crashes, so you need to assume unreported crashes at different forms of intersection are comparable. In the US, what is required to be reported varies from state to state, only state-based comparisons may get closest to an accurate answer. Injury and fatal crashes are more likely to be reported, and required to be reported.

New contributor
Scott B is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .