Crash prone 'modern roundabouts'

[Tarkus]

Having 50% less fatalities in Carmel isn't that impressive when the typical Carmel roundabout has 50% less traffic though.

Exactly.  It's about the per MEV rate.

I'm still looking around to see if the crash data is easily accessible for Carmel.  As much as I love to bag on ODOT, at least they have a publicly-searchable crash statistics database.

[tradephoric]

Have you compared Carmel's crash stats to other cities of the same population?

Dearborn, Michigan has a population of 98,153 compared to Carmel's population of 85,927.  Over the past 3 years there have been 3 fatal crashes at Dearborn's top 100 most crash prone intersections.  There have also been 3 fatal crashes at Carmel's roughly 100 roundabouts during the same time period.  Maybe Carmel has been on the wrong side of the bell curve lately, but for whatever reason there has been a string of roundabout fatalities in the city of Carmel.

[jakeroot]

Have you compared Carmel's crash stats to other cities of the same population?

Dearborn, Michigan has a population of 98,153 compared to Carmel's population of 85,927.  Over the past 3 years there have been 3 fatal crashes at Dearborn's top 100 most crash prone intersections.  There have also been 3 fatal crashes at Carmel's roughly 100 roundabouts during the same time period.  Maybe Carmel has been on the wrong side of the bell curve lately, but for whatever reason there has been a string of roundabout fatalities in the city of Carmel.

Why not expand the year measure to 10 years? 20 years? That's solid data. You seem hard-set on the last three years, which are clearly Carmel's worst.

[tradephoric]

Why not expand the year measure to 10 years? 20 years? That's solid data. You seem hard-set on the last three years, which are clearly Carmel's worst.

Roundabouts didn't exist in Carmel 20 years ago.  Even 10 years ago there were maybe only a half-dozen to a dozen roundabouts in the city.  It's hard to analyze data that doesn't exist.

[jeffandnicole]

He's well known to cherry pick data. And because an accident occured near a roundabout and had nothing to do with the functionality of a roundabout (such as a drunk driving straight into the center) doesn't change his opinion that the roundabout itself is defective.

[jakeroot]

Why not expand the year measure to 10 years? 20 years? That's solid data. You seem hard-set on the last three years, which are clearly Carmel's worst.

Roundabouts didn't exist in Carmel 20 years ago.  Even 10 years ago there were maybe only a half-dozen to a dozen roundabouts in the city.  It's hard to analyze data that doesn't exist.

You'd be analyzing before/after data.

[tradephoric]

Carmel, the self-proclaimed roundabout capital of America, had the same number of fatal crashes as Dearborn, Michigan (a city with ZERO roundabouts).  That's a fact everybody can excuse away all they want.  Have fun with that. 

[lordsutch]

Carmel, the self-proclaimed roundabout capital of America, had the same number of fatal crashes as Dearborn, Michigan (a city with ZERO roundabouts).  That's a fact everybody can excuse away all they want.  Have fun with that. 

What you call "excus[ing] away," statisticians call "controlling for other variables." But sure, we'll play your game. Compare the number of fatal accidents in Dearborn to fatal accidents in Carmel, not over your cherry-picked last 27 months, but the last 10 years - a nice, even, round number, nothing up my sleeve.

Until you do at least that, as far as I'm concerned you're not engaging in honest research, you're just engaging in a vendetta against a type of intersection you seem to personally and irrationally dislike.

[tradephoric]

^Carmel currently has 95 roundabouts.  I'm comparing the fatal crashes of 100 Dearborn intersections to the fatal crashes of 95 Carmel roundabouts.  Each year you go back there are fewer and fewer Carmel roundabouts to analyze.  It wouldn't be fair to compare the number of fatal crashes at Dearborn's 100 most dangerous intersections over the past decade to the number of fatal crashes at Carmel's roundabouts (since 10 years ago, there were only about a dozen roundabouts in Carmel).  That's apples to bananas.  Admittedly this is a back of the napkin analysis, but your suggestion totally throws off the numbers.  While we are at it, you might as well compare the fatal crashes in Chicago to Billings, Montana. 

I just believe the safety benefits of roundabouts are being over estimated.  Keep in mind, the national safety numbers so often cited in literature is based on a study that is over 15 years old.  Roundabouts have evolved at lot since 2000 when the Insurance Institute for Highway Safety released their report.  The more recent safety studies on roundabouts have found that in many cases total crashes are going up (ie. Wisconsin study found a 12% increase in total crashes, not a 39% decrease as cited in the 2000 IIHS study).  You mention "honest research", yet there are so many DOTs that cite the 2000 IIHS study like it is gospel to push for a roundabout.  That to me isn't honest either.

[jakeroot]

^Carmel currently has 95 roundabouts.  I'm comparing the fatal crashes of 100 Dearborn intersections to the fatal crashes of 95 Carmel roundabouts.  Each year you go back there are fewer and fewer Carmel roundabouts to analyze.  It wouldn't be fair to compare the number of fatal crashes at Dearborn's 100 most dangerous intersections over the past decade to the number of fatal crashes at Carmel's roundabouts (since 10 years ago, there were only about a dozen roundabouts in Carmel).  That's apples to bananas.  Admittedly this is a back of the napkin analysis, but your suggestion totally throws off the numbers.  While we are at it, you might as well compare the fatal crashes in Chicago to Billings, Montana.

Deerborn has 100 conventional signals. 10 years ago, Carmel had mostly signals, so 10 years ago, the cities had much more in common. Now that Carmel has more roundabouts, the comparison would be whether or not fatal collisions in Carmel have dropped over time with the new roundabouts, and whether or not the not number of fatal collisions in Deerborn has dropped or stayed constant since they didn't change anything.

The comparison is Carmel, who changed over to roundabouts, to Deerborn, who changed nothing.

[Landshark]

Carmel, the self-proclaimed roundabout capital of America, had the same number of fatal crashes as Dearborn, Michigan (a city with ZERO roundabouts).  That's a fact everybody can excuse away all they want.  Have fun with that.

Carmel & Dearborn are apples to oranges.   One is older dense innercity, the other sprawly modern suburbia.   Not only do the physical designs greatly differ, but so do the demographics.  I am sure Carmel has lower % of transit ridership, longer average auto trips, etc.   I also wouldn't be surprised if the average speed limit was also higher in Carmel vs. Dearborn. 

[tradephoric]

Carmel & Dearborn are apples to oranges.   One is older dense innercity, the other sprawly modern suburbia.   Not only do the physical designs greatly differ, but so do the demographics.  I am sure Carmel has lower % of transit ridership, longer average auto trips, etc.   I also wouldn't be surprised if the average speed limit was also higher in Carmel vs. Dearborn. 

Take Troy, Michigan as the comparison.  Troy has roughly the same population as Carmel and is the poster child for "suburban sprawl" .  Over the past 4 years the 100 most crash prone intersections in Troy has been the site of 3 fatal crashes.  This is the same number of fatal crashes as Carmel's roundabouts over the same time period.  All things being equal, Carmel should have had fewer fatal crashes:

#1.  This analysis is comparing the fatal crashes of 100 Troy intersections to only 95 Carmel roundabouts.  And back in 2013, there was only about 60-70 roundabouts in Carmel.  Even though fewer intersections were analyzed, Carmel had the same number of fatal crashes as Troy. 

#2.  Average traffic volumes at Troy intersections are likely much higher than the Carmel roundabouts.  Some of the intersections in Troy experience 70,000 vehicles per day.  I don't know of a single roundabout in the entire country with traffic counts approaching that.  Even though the Carmel roundabouts have lower traffic volumes, they had the same number of fatal crashes.   

The fact is over the past 4 years Carmel roundabouts have been the site of 3 fatal crashes.  Other similarly sized communities in the United States have had the same number of fatalities as Carmel has had.  Where is the 90% reduction in fatal crashes?

[roadfro]

We really ought to consider Pelican phasing, like in the UK. They use the standard R/Y/G signal setup, with solid red during the WALK phase, and flashing amber during the DON'T WALK phase. If the iffy compliance with HAWK signals has anything to do with failure to understand the signals themselves, perhaps compliance would go up when people meet a standard traffic light.

https://www.youtube.com/watch?v=aI3FiOf4rJI

Just catching up on this thread after several weeks. I know it's a bit off topic, but I had to chime in to agree with this statement.

I have not been a fan of the HAWK signal style since I saw it proposed for the 2009 MUTCD. The Pelican crossing signal demonstrated here makes much more sense intuitively, in the context of existing signal indications, and addresses two things I don't like with a HAWK (signal is dark until ped activation, and the wig-wag red). The only downside is that you couldn't use the Pelican signal at the entrance to a roundabout, as was mentioned earlier in the thread–although really, a signal at the entrance to a roundabout is counter-intuitive to the roundabout controls anyway...

[renegade]

I think I'll leave this here ...

http://www.mlive.com/news/ann-arbor/index.ssf/2016/04/road_commission_improves_state.html#incart_river_home

Number of crashes at State and Ellsworth roundabout drops following initial spike

It's getting better.  The idiots are beginning to learn how to navigate it.  And it's not nearly as bad as someone who does not live here makes it out to be.  I have to deal with it every day, and have yet to see an accident.  I don't think anyone has died there, either.

[tradephoric]

I think I'll leave this here ...

http://www.mlive.com/news/ann-arbor/index.ssf/2016/04/road_commission_improves_state.html#incart_river_home

Number of crashes at State and Ellsworth roundabout drops following initial spike

It's getting better.  The idiots are beginning to learn how to navigate it.  And it's not nearly as bad as someone who does not live here makes it out to be.  I have to deal with it every day, and have yet to see an accident.  I don't think anyone has died there, either.

Here is ten years of crash data for State and Ellsworth (green = pre-roundabout; yellow=roundabout construction year; red = post-roundabout).  You can spin it however you like, but the fact is there has been a spike in total crashes without a reduction in injury crashes.  I didn't see 2015 injury accidents listed in the article renegade cited so that was not included in the chart.


http://semcog.org/Data-and-Maps/High-Frequency-Crash-Locations/Point_Id/81016689/view/RoadIntersectionCrashDetail

[kphoger]

#2.  Average traffic volumes at Troy intersections are likely much higher than the Carmel roundabouts.  Some of the intersections in Troy experience 70,000 vehicles per day.  I don't know of a single roundabout in the entire country with traffic counts approaching that.  Even though the Carmel roundabouts have lower traffic volumes, they had the same number of fatal crashes. 

Then what we need is a comparison between signalized intersections and roundabouts where traffic volumes are comparable as well as city sizes.  This would yield apples-to-apples data to review:  similar city sizes make for a meaningful comparison of total crashes and fatalities, while similar traffic volumes made for a meaningful comparison of intersection designs.

[tradephoric]

I just find it ironic that Carmel Indiana, the roundabout capital of America, has had the same number of fatal crashes as similarly sized communities with zero roundabouts.  Articles will keep regurgitating the IIHS safety stat that says modern roundabouts reduce fatal crashes by 90%.  If you say it enough times it must be true.  A lot of people on this thread believe it.

[paulthemapguy]

I can't see roundabouts reducing the number of fatal crashes if they occur in low-speed areas...But then again you shouldn't be putting roundabouts on highways with high speeds in the first place!  If someone proposed a roundabout on a high speed roadway, I'd say, "you're gonna put a big circle in the middle of a road where the speed is 45-55mph?  Are you serious?"  Doesn't stop some agencies though.

[jeffandnicole]

I can't see roundabouts reducing the number of fatal crashes if they occur in low-speed areas...But then again you shouldn't be putting roundabouts on highways with high speeds in the first place!  If someone proposed a roundabout on a high speed roadway, I'd say, "you're gonna put a big circle in the middle of a road where the speed is 45-55mph?  Are you serious?"  Doesn't stop some agencies though.

See NJ 70/72. https://goo.gl/maps/c6XhdHMXcVn  Although it's a Jersey traffic circle, it does have roundabout features (everyone yields prior to entering).  Speed limits on 3 of the spokes are 55 mph; on the other 2 spokes, limits are 50 mph.

No issues here.  You wouldn't think it by looking at the aerial photos, but it can jam up during rush hours!

[JMAN_WiS&S]

The multi lane roundabout at US 53 and Town Hall Rd south of the US 53/I94 interchange in Eau Claire Wisconsin has been subject to quite a few crashes recently and now the DOT is proposing changes such as building the mound up in the center, and replacing the current advanced warning signs with fluorescent ones, even though the ones on US 53 have flashers on them. The major issue is the speed limit right in the area is 55 mph and the roundabout limit is 15.

[kalvado]

I can't see roundabouts reducing the number of fatal crashes if they occur in low-speed areas...But then again you shouldn't be putting roundabouts on highways with high speeds in the first place!  If someone proposed a roundabout on a high speed roadway, I'd say, "you're gonna put a big circle in the middle of a road where the speed is 45-55mph?  Are you serious?"  Doesn't stop some agencies though.

See NJ 70/72. https://goo.gl/maps/c6XhdHMXcVn  Although it's a Jersey traffic circle, it does have roundabout features (everyone yields prior to entering).  Speed limits on 3 of the spokes are 55 mph; on the other 2 spokes, limits are 50 mph.

No issues here.  You wouldn't think it by looking at the aerial photos, but it can jam up during rush hours!

Just a single one? Not a big deal!  How about a real chain?
https://www.google.com/maps/place/Albany,+NY/@42.6428116,-73.8497982,16z/data=!3m1!1e3!4m2!3m1!1s0x89de0a34cc4ffb4b:0xe1a16312a0e728c4
55 MPH posted throughout.

I often compare roundabout construction with prescribing insulin: in the right situation is a life saver, but insulin doesn't help broken bones or pneumonia...

[kphoger]

Just a single one? Not a big deal!  How about a real chain?
https://www.google.com/maps/place/Albany,+NY/@42.6428116,-73.8497982,16z/data=!3m1!1e3!4m2!3m1!1s0x89de0a34cc4ffb4b:0xe1a16312a0e728c4
55 MPH posted throughout.

That doesn't even make sense to be signed at 55 mph.  Looks like 45 might be more appropriate.

[tradephoric]

^On streetview it appears Slingerlands Bypass is posted at 45 mph through the roundabouts.  It only increases to 55 mph once you get past the string of roundabouts.   Kalvado, is it possible the speed limit was just recently reduced?  The streetview images are from September 2015. 

[kalvado]

^On streetview it appears Slingerlands Bypass is posted at 45 mph through the roundabouts.  It only increases to 55 mph once you get past the string of roundabouts.   Kalvado, is it possible the speed limit was just recently reduced?  The streetview images are from September 2015.

I wish I remember. I drive there once every few months. I am usually on a slower side of things going 60. But I never been there during rush hour.
Before roundabouts that used to be posted 55.

[Sykotyk]

Just one clarification for those discussing the Carmel/SE Michigan statistics.... In Carmel you're counting ONLY the 100 roundabouts, while in SE Michigan you're counting the top 100 most crash prone intersections, rather than ALL intersections.

Let's say there's 10,000 intersections covered under the data used for SE Michigan.

Now the data is 4/100 to 29/10,000.

These are supposed to be safer. In a lot of instances, they are not. And the main culprit, to me, is that many people do not know what the YIELD sign truly means. They're used to STOP, and in the absence of STOP their thought is to go. We see it on on-ramps all the time where the YIELD sign on the ramp is never heeded. They weave into traffic, cut off cars in the right lane, flip you off for not letting them over, etc. They don't consider it an actionable sign.

This has been ingrained for decades for some of these people and now suddenly throwing a 'new' type of interchange at them will not get them to stop their bad habits. It just amplifies them.