Crash prone 'modern roundabouts'

[tradephoric]

Crashes are up at a new 2x2 roundabout at Highway 22 & Madison Ave. in Mankato, Minnesota.  One thing to note is this roundabout has standard pavement markings (as opposed to the fishhook pavement markings).  This is likely a direct result of the study done on the Richfield, Minnesota roundabout that studied the effects of standard vs. fishhook pavement markings. 

The lessons learned from the Richfield, Minnesota roundabout were applied to the Mankato roundabout yet crashes are still up.  What's concerning is officials seem resigned to the fact that accidents will go up at these high volume roundabouts.  Of course, there is something else in common between the Richfield & Mankato roundabouts.... small central island diameters.  Perhaps officials should focus on the size of the roundabouts before resigning themselves to the fact that accidents will increase at these roundabouts.  Don't accept mediocre performance.

The tradeoffs were discussed in the planning stages of the roundabouts. In exchange for more accidents, the roundabout would make said crashes much safer. And improve traffic flow.

http://www.keyc.com/story/28328544/crashes-at-mn-hwy-22-more-frequent-but-much-safer

Sure, it's great that roundabouts reduce injury crashes, but which would you prefer?

A.)  A roundabout with a central island diameter of 150 feet that has 20 crashes and 1 injury crash.
B.)  A roundabout with a central island diameter of 100 feet that has 60 crashes and 1 injury crash. 

I'll pick Option A.

[tradephoric]

The lowest crash rate roundabouts on this thread have had central island diameters between 130 and 200 feet.  This equates to Inscribed Circle Diameters of roughly 200 to 300 feet (60 to 90 meters).   Take a look at the graph below that looks at the crash rate of Czech Republic roundabouts in relation to their inscribed diameter.  The multi-lane roundabouts with ICD's between 60 to 90 meters perform well.  Smaller than 60 meters and the crash rate goes up; larger than 90 meters and the crash rate goes up. 


http://www.degruyter.com/view/j/tvsb.2010.x.issue-1/v10160-010-0009-0/v10160-010-0009-0.xml

The "˜goldilocks' inscribed circle diameter appears to be between 200 and 300 feet (60 to 90 meters) for multi-lane roundabouts. 

[Sykotyk]

Rotaries in the northeast work very well. I like them, such as this one I drove on US44 just east of I-495. I drove it east and west, and though traffic seemed to be endless from it, it cleared out well because of the spacing between entry points (though I really think that Friendly's entrance should be closed and only access it from the side road).

https://www.google.com/maps/place/Plympton,+MA/@41.9027404,-70.9590075,19z/data=!3m1!1e3!4m2!3m1!1s0x89e497d5f44178cb:0x3d50f1892f6c5149

The smaller modern roundabouts aren't safe because drivers treat a yield sign as nothing. Just as an on-ramp has a yield sign, the mindset is if you can get in front of that car, you can speed up and get in front of them. This is tolerated on freeway on-ramps, but dangerous in roundabouts.

The idea you must 'yield' to traffic in the circle only makes it worse. The goal then, is to get in that circle at all costs, so everyone else has to yield to you. Small traffic circles, though with a yield sign, should be treated just the same as a four-way stop. Each entry point taking turns letting one car into the circle (regardless which exit point the cars take). You can have more in the circle than one car, but traffic to your right gets to go first (i.e., you yield to the next entry point).

But, our signage laws AND the way our drivers use those signs is not possible. Yield doesn't mean yield to most drivers. I've seen roundabouts where a line of 5-10-15-20 cars all breeze into a roundabout from one entrypoint, choking the entire thing because nobody else can go until every last car gets through from that one point. Meanwhile, traffic backed up the other entry points, and as soon as they got their shot, they floored it. Only for two entry points to clog the center, not allowing any new traffic in as that initial burst had to disperse.

A four-way stop with each entrypoint alternating N/S to E/W with left turn traffic yielding to through and right turn traffic first would been the most efficient means to clear our a huddled intersection.

[realjd]

Rotaries in the northeast work very well. I like them, such as this one I drove on US44 just east of I-495. I drove it east and west, and though traffic seemed to be endless from it, it cleared out well because of the spacing between entry points (though I really think that Friendly's entrance should be closed and only access it from the side road).

https://www.google.com/maps/place/Plympton,+MA/@41.9027404,-70.9590075,19z/data=!3m1!1e3!4m2!3m1!1s0x89e497d5f44178cb:0x3d50f1892f6c5149

The smaller modern roundabouts aren't safe because drivers treat a yield sign as nothing. Just as an on-ramp has a yield sign, the mindset is if you can get in front of that car, you can speed up and get in front of them. This is tolerated on freeway on-ramps, but dangerous in roundabouts.

The idea you must 'yield' to traffic in the circle only makes it worse. The goal then, is to get in that circle at all costs, so everyone else has to yield to you. Small traffic circles, though with a yield sign, should be treated just the same as a four-way stop. Each entry point taking turns letting one car into the circle (regardless which exit point the cars take). You can have more in the circle than one car, but traffic to your right gets to go first (i.e., you yield to the next entry point).

But, our signage laws AND the way our drivers use those signs is not possible. Yield doesn't mean yield to most drivers. I've seen roundabouts where a line of 5-10-15-20 cars all breeze into a roundabout from one entrypoint, choking the entire thing because nobody else can go until every last car gets through from that one point. Meanwhile, traffic backed up the other entry points, and as soon as they got their shot, they floored it. Only for two entry points to clog the center, not allowing any new traffic in as that initial burst had to disperse.

A four-way stop with each entrypoint alternating N/S to E/W with left turn traffic yielding to through and right turn traffic first would been the most efficient means to clear our a huddled intersection.

On small roundabouts, you don't treat them as a 4 way stop. You treat them as a roundabout. If two cars arrive at exactly the same time, it's still treated as a roundabout, i.e. yield to the car on the left.

Where do they put yield signs at the top of on ramps? That's dumb. It's a merge, not a yield.

[tradephoric]

Rotaries in the northeast work very well. I like them, such as this one I drove on US44 just east of I-495. I drove it east and west, and though traffic seemed to be endless from it, it cleared out well because of the spacing between entry points (though I really think that Friendly's entrance should be closed and only access it from the side road).

https://www.google.com/maps/place/Plympton,+MA/@41.9027404,-70.9590075,19z/data=!3m1!1e3!4m2!3m1!1s0x89e497d5f44178cb:0x3d50f1892f6c5149

The US44 rotary has an abysmal crash record though.  From 2009 to 2011, the rotary averaged 123 crashes a year.  The rotary has a inscribed circle diameter of 400 feet which leads to high circulating speeds. 

According to the Department of Transportation, approximately 40,000 cars use the rotary each day and there were 58 accidents there a year on average between 2007 and 2009 before the numbers jumped to about 123 a year between 2009 and 2011. Built around 1932, the rotary encircles a green lawn used as an American Legion memorial where flags are periodically flown as a way to honor deceased servicemen and women.

https://www.bostonglobe.com/metro/regionals/south/2014/03/12/after-decades-gridlock-revamped-rotary-middleborough-will-open-region-some-say/6CoApiCZds5HWcPyTyhl2I/story.html

[johndoe]

Here's an interesting article: http://journalstar.com/news/local/govt-and-politics/mayor-picks-elevated-roundabout-for-warlick-intersection/article_064e9de0-f6b5-5bd0-9c6d-8eba655c3c4f.html

It alludes to previous changes at another local roundabout: http://journalstar.com/news/local/fixes-aimed-at-simplifying-superior-street-roundabout/article_088eec03-c54d-5f34-9808-36ffa4ead11c.html

[Sykotyk]

The US44 rotary has an abysmal crash record though.  From 2009 to 2011, the rotary averaged 123 crashes a year.  The rotary has a inscribed circle diameter of 400 feet which leads to high circulating speeds. 

123 crashes may sound impressive... but what's that to traffic volume compared to other rotaries with similar volume?

[tradephoric]

123 crashes may sound impressive... but what's that to traffic volume compared to other rotaries with similar volume?

Assuming the 40,000 volume cited in the article, that equates to 8.4 accidents / million vehicles.  As a general rule, anything over 2 accidents / million vehicles at an intersection is considered high.

[tradephoric]

There have been 13 crashes in 1 week at a new multi-lane roundabout in Lakeville Minnesota.   It's on track to have 676 crashes the first year of operations!  That won't happen, but it's still bad PR when you have 13 crashes in a week.

http://kstp.com/article/stories/s3898071.shtml

https://www.youtube.com/watch?v=PQGTfBAAWac

Many of the design recommendations that are found in the Richfield, MN study (which studied a crash prone 2x2 roundabout) have been implemented in the new roundabout in Lakeville (standard pavement markings are being used as opposed to fishhook markings).  It will be interesting to see if the lessons learned at the 2x2 roundabout in Richfield will be able to reduce the crash rate at the 2x2 roundabout in Lakeville (based on the first week of crash data though, it's not looking good).

[jakeroot]

(based on the first week of crash data though, it's not looking good).

I think it's far too early to be predicting long-term crash data. Hell if I know the stats, but I'm willing to bet that most roundabouts, when they first open, have a spat of collisions as people figure things out. Give them a chance.

[tradephoric]

(based on the first week of crash data though, it's not looking good).

I think it's far too early to be predicting long-term crash data. Hell if I know the stats, but I'm willing to bet that most roundabouts, when they first open, have a spat of collisions as people figure things out. Give them a chance.

Every multi-lane (2x2) roundabout cited on this thread where crash data is available has seen an increase in total crashes.  Can you give me a compelling reason why the multi-lane roundabout in Lakeville will be any different?  It's not too early to predict a trend. 

Driving through a multi-lane roundabout shouldn't be a foreign experience for most Minneapolis drivers.  There is a multi-lane roundabout 2 miles down the road from the new roundabout in Lakeville.  In addition, there are roughly 30 multi-lane roundabouts in the Minneapolis region. 

[jeffandnicole]

Every multi-lane (2x2) roundabout cited on this thread where crash data is available has seen an increase in total crashes.  Can you give me a compelling reason why the multi-lane roundabout in Lakeville will be any different?  It's not too early to predict a trend. 

LOL 

Why would someone post about non-accident prone roundabouts on a thread titled "Crash Prone Modern Roundabouts"?  Of course most roundabouts cited on this thread have seen an increase in total crashes.  That's the whole fricken point of this thread.

Should we start up a thread "Non-Accident prone roundabouts"?  Can't wait for the exciting dialogue on that thread.

[tradephoric]

Previously in this thread, a comprehensive list of 2x2 roundabouts in America was queried out from a database of over 5,100 modern roundabouts.  This was in direct response to a request you had made jeffandNicole.  I attempted to find before/after crash data for each one.  I could only track down crash data for 14 out of the 40, but of those 14, all had an increase in total crashes.

It would probably be fair to see a longer list, which will show both 100' diameter roundabouts that don't have high crash frequencies, and roundabouts with larger diameters and how they rank. 

Here is a list of 40 multi-lane roundabouts that I think could be meaningful to this conversation.  This was queried from a database of over 5,100 modern roundabouts.  This was the criteria used to query out the list:

-roundabouts constructed within the last 10 years
-all approaches have 2 entry lanes
-roundabouts have 4-legs
-main & side streets are major state or county routes (ie. likely high AADT roundabouts)
-interchange roundabouts (and frontage road roundabouts) not included



Now the hard part is finding accurate before/after crash data to make any type of analysis.  Here is a google KMZ file that includes the 40 roundabouts in the chart above:

http://www.mediafire.com/download/cd570rarros2c3g/Multi-Lane+Roundabouts+%28with+high+AADT%29.kmz

[tradephoric]

Here's the 14 roundabouts i eventually found crash data for.  Increases in total crashes ranged from 36% to 1400%. 

[johndoe]

Interesting data, some of those are shocking!  171 crashes in one year? ! That's hard to believe! (Not doubting you)   Can I ask why you exempted interchanges ?

[Rothman]

Yeah, I questioned the data because I travel through roundabouts on that list frequently (more than once weekly).  If the data was true, I'd come across more accidents than I do, even from just an anecdotal sense.

[NJRoadfan]

After braving this roundabout, I'd have to think there are plenty of accidents: https://goo.gl/maps/7sKaT

Compared to others in the area, it has a small diameter island. Its pretty difficult to "jump in" during rush hour.

[Mrt90]

This is a two lane roundabout in Pleasant Prairie, Wisconsin.  The problem that I have with this roundabout is that there is an inconsistency regarding which lane(s) can exit the roundabout.  If you exit onto 39th Avenue going north or south or if you exit onto Hwy165 going west you can be in either lane to exit. However, if you are exiting on Hwy165 going east then you must be in the outside lane in order to exit and the inside lane is only to continue in the roundabout.  Why is that one exit different that the other 3?  What usually happens to me is that I'm trying to continue east on Hwy165 so I know that I have to be in the right lane the way the lanes are drawn, but someone will pass me in the inside lane and continue east on Hwy165, cutting me off. If there is another car on the inside I'm basically forced to stop because everyone thinks they can exit from the inside lane, and because of the other roads in the area almost everyone that enters the roundabout from the west is exiting south or east, not north. I'm not sure why the inside lane isn't always a "continue in the roundabout" lane and the outside lane is the only exit lane, but only 1 of the 4 exits in this roundabout is drawn this way.  Is this typical?


https://www.google.com/maps/@42.5225151,-87.8537268,138m/data=!3m1!1e3?hl=en

[johndoe]

Mrt90, it's because of the dual EBL.  (You could argue dual lefts probably aren't necessary because look what happens to those two lanes a whole 50' north of the intersection...)  To me it looks like there are lots of signs and markings; it's unwise for drivers to assume a 2 lane approach always means two through lanes.  If people would drive roundabouts like signals many problems would be minimized.

[Mrt90]

johndoe I see what you mean, but it probably would have been better if the road going east had two lanes merge into one after the roundabout like the other roads do, so that the exiting lanes were consistent.  The other problem is that if you look at the map the I linked and go a little bit west, there is another roundabout (the link below show both).  So someone going east on Hwy 165 through both roundabouts can use either lane to continue east through the first roundabout, but then is supposed to use only the outside lane to continue east on the 2nd roundabout.  And my experience is that almost everyone gets that 2nd roundabout east exit wrong.  It's probably safer for me to do it incorrectly just like everyone else.

https://www.google.com/maps/@42.5256339,-87.8568222,1106m/data=!3m1!1e3?hl=en

[UCFKnights]

That is among the best signed roundabout I have seen. Hard to believe people are getting that wrong... crazy

[tradephoric]

Found crash data for the multi-lane roundabout at Venice Ave & Jacaranda Blvd.  It has a crash rate of 4.16 (the highest crash rate intersection in Sarasota County).  The county considers any intersection with a crash rate of over 2 as "critical".   

http://www.heraldtribune.com/article/20150815/ARTICLE/150819785

[tradephoric]

Are there any multi-lane (2x2) roundabout in America where there was a reduction in total crashes?  I'm starting to wonder.

[jakeroot]

Also in the article;

Oddly, as Bennett noted, a by-product of the roundabout is more lives have been saved even though crash rates skyrocketed. People drive slower in roundabouts — the posted speed limit is 15 mph in Venice – lessening the chance of serious injuries, but the lack of knowledge of how to maneuver them causes plenty of sideswipes and insurance claims in Venice.

Yes, there does seem to be an increase in collisions at 2x2 roundabouts. Fine. But if the overall number of critical-injury collisions is falling, that's good, yes?

Do you have some data for 2x2 roundabouts over time? I seem to remember some previous data that you provided showing overall collisions falling over time. I know this roundabout had 52, 57, and 50 over the last three years (overall # of collisions). So this most recent year was its best ever (despite the previous year being the worst ever). I think we need to give this intersection a few more...uhh, decades, before we declare it a loss.

[tradephoric]

Do you have some data for 2x2 roundabouts over time? I seem to remember some previous data that you provided showing overall collisions falling over time. I know this roundabout had 52, 57, and 50 over the last three years (overall # of collisions). So this most recent year was its best ever (despite the previous year being the worst ever). I think we need to give this intersection a few more...uhh, decades, before we declare it a loss.

If an agency is aware that the roundabout has a crash rate of 4.16 (over 2x what they deem "critical") and does nothing to address the problem, they could potentially be found negligent.  "The roundabout will work well in 20 years"  is not a strong defense.  The roundabouts best performing year had total crashes 350% higher than when the intersection was signalized.  That's a bad result.  When total crashes increase by 350%, there is no guarantee that injury crashes will drop (several roundabouts have been cited in this thread where injury crashes have gone up).