The Paradox of Road Choice

Two physicists and a computer scientist used Google maps to study traffic in Boston, London, and New York, and found that when people use real-time driving maps to try to pick the fastest routes, traffic slows down. Their solution: close a few roads so drivers have fewer options. (HT: Kottke)[%comments]

Leave A Comment

Comments are moderated and generally will be posted if they are on-topic and not abusive.



View All Comments »
  1. Mike B says:

    Hasn’t Braess’s Paradox been covered here before?

    Thumb up 0 Thumb down 0
  2. Cheryl says:

    Did they discover this before or after NYC started closing roads to vehicular traffic?

    Thumb up 0 Thumb down 0
  3. Rob says:

    The paper only makes the point that if *all* motorists use real-time maps to optimize their routes then the total throughput to decrease. This is unlikely to ever occur in practice and is easily prevented (limit accessibility of real-time data to some fraction of all drivers).

    Just as with the study of law-breaking drivers summarized at , there is likely an optimal percentage of real-time-optimizing drivers that is greater than zero and less than one hundred. My OTA estimate is that the number who do use real-time routing is much much less than the optimal number; I thus find the mischaracterization of this study (“using real-time routing hurts traffic flow”) counterproductive.

    Thumb up 0 Thumb down 0
  4. Joe says:

    what happens if you take this to an extreme and close all the roads except one?

    Thumb up 0 Thumb down 0
  5. Robot Mistake says:

    I thought that the problem in these studies is a poor definition of real-time systems.

    The system changes faster then we can recieve updates when drivers are allowed to use the information to make decisions.

    Oh. And yes. I have always believed that traffic would move faster if no one was allowed to changelanes (TM).

    Thumb up 0 Thumb down 0
  6. YX says:

    Interesting read, though after driving in Boston a couple of years, I’m convinced there is nothing short of shooting every driver on sight can improve driving conditions here.

    Thumb up 0 Thumb down 0
  7. Doc says:

    This study is a good example of the risks of empirical work. Choosing routes between Harvard Square and Boston Common to study driver behavior is akin to choosing an Inuit village to stidy children’s winter play patterns. They have WAY more winter than the rest of the world and they come from a different culture than most of us. Same is true for Boston/Camridge travelers. Thay have WAY more congestion and arcane routes than most of the rest of us and they practice a very different set of driving behaviors. Check out a copy of How to Be a Boston Driver for details.

    I’d like to see this replicated with a true destination and a tightly defined origin, say Dodger Stadium and season ticket holders who live between Manchester and Artesia Blvd. in the South Bay. On game day, they could represent the marginal entrants casuing congestion on alternate routes.

    Thumb up 0 Thumb down 0
  8. Eric M. Jones says:

    @7 Doc is absolutely correct.

    My first experience of so-called “driving” in Boston evoked rage and outrage regarding the behaviour of other so-called “drivers”. I was amazed; coming from LA where drivers are extremely polite and occasionally armed. The movie “Death Race 2000″ comes to mind.

    Knowledgeable locals avoid the Boston area at all costs. The comic-tragic Big Dig (Now 22 BILLION dollars and still not finished) has by all accounts made the traffic worse. NYC traffic is infinitely better.

    Thumb up 0 Thumb down 0