Selfish-routing network paradox
Braess's Paradox
Adding a low-cost link to a congested network can worsen equilibrium travel time when individually rational route choices impose congestion on one another.
more capacity can raise Wardrop-equilibrium cost
The paradox compares user equilibrium before and after a network change. It is not universal: demand, cost functions, information, control, and network topology determine whether the added link helps.
Vehicles repeatedly choose the currently faster route. Toggle the shortcut to see decentralized choices converge to a different network equilibrium.
(minutes)
The primary slider and this instrument share one state.
- CHANGE
- Travel demand
- WATCH
- route use and equilibrium time
- MEANING
- Vehicles repeatedly choose the currently faster route. Toggle the shortcut to see decentralized choices converge to a different network equilibrium.
A shortcut can become a trap everyone chooses.
Route colors show load and travel time while moving vehicles reveal the migration toward equilibrium.
What it actually says
Braess's Paradox arises from interaction between congestion and decentralized route choice. A new edge changes incentives, and the individually attractive deviation can concentrate traffic on shared bottlenecks.
A system optimum and a user equilibrium are different objects. Pricing, routing guidance, reservations, or capacity control can align them, but information alone can also redirect everyone toward the same congested path.
"A useful law compresses a pattern. It does not erase the conditions that make the pattern true."
How the idea developed
The modern form emerged through observation, argument, and later refinement. The timeline separates the first insight from the version now used in textbooks and practice.[1]
Dietrich Braess publishes the paradox in German.
Congestion-game theory connects it to Wardrop equilibrium.
An English translation appears in Transportation Science.
Transportation, communication, and power networks study capacity paradoxes.
How the pattern works
The relation becomes useful only when its mechanism, measurement process, and operating range are visible.
Each traveler minimizes personal cost given current loads.
A route choice slows others on shared edges.
The shortcut creates a strategy that appears individually dominant.
Repeated switching can settle at a higher common cost.
The paradox compares user equilibrium before and after a network change. It is not universal: demand, cost functions, information, control, and network topology determine whether the added link helps.
Where it earns its keep
Applications are strongest when the law changes a decision, measurement, model, or experiment rather than merely providing an analogy.
Test road projects as network games
ApplicationSimulation can compare equilibrium and system-optimal outcomes.
Use calibrated demand and induced-demand scenarios.
Design routing incentives
ApplicationLatency-sensitive flows can overload attractive links.
Protocol dynamics and stochastic demand matter.
Evaluate pricing or closures
ApplicationTolls and selective restrictions can reduce total delay.
Distributional access and diversion effects require review.
Where it stops working
Not every added link produces the paradox, and observed before-after changes can mix construction, demand, incidents, and adaptation.
Static deterministic equilibria omit departure time, heterogeneous preferences, uncertainty, learning, and induced demand.
"Closing roads always improves traffic"
Better: Only particular network-demand-cost structures do."The shortcut itself is slow"
Better: Its incentive effect on shared bottlenecks creates the paradox."Drivers behave irrationally"
Better: Each route choice can be individually rational."Traffic information guarantees efficiency"
Better: Common information can synchronize harmful choices.Sources and further reading
Original publications and serious secondary scholarship are prioritized over summaries.
- Braess et al. - On a Paradox of Traffic PlanningEnglish translation of the original result.https://doi.org/10.1287/trsc.1050.0127
- Roughgarden and Tardos - How Bad Is Selfish Routing?Foundational price-of-anarchy analysis.https://doi.org/10.1145/502090.502094
- MIT OpenCourseWare - Transportation Systems AnalysisNetwork equilibrium course materials.https://ocw.mit.edu/courses/1-201j-transportation-systems-analysis-demand-and-economics-fall-2008/
- Transportation Science - Braess ParadoxPublisher record and historical context.https://pubsonline.informs.org/doi/10.1287/trsc.1050.0127