In a road bicycle race, the peloton (, originally meaning ) is the main group or pack of riders. Riders in a group save energy by riding behind or close to other riders (drafting, or slipstreaming). The reduction in drag is dramatic; riding in the middle of a well-developed group, drag can be reduced by as much as 95%. Exploitation of this potential energy saving leads to complex cooperative and competitive interactions between riders and teams in race tactics. The term is also used to refer to the community of professional cyclists in general.
Definition More formally, a peloton is defined as "two or more cyclists riding in sufficiently close proximity to be located either in one of two basic positions: (1) behind cyclists in zones of reduced air pressure, referred to as ‘drafting’, or (2) in non-drafting positions where air pressure is highest. Cyclists in drafting zones expend less energy than in front positions."Trenchard, Hugh, 2013. "Peloton phase oscillations," Chaos, Solitons & Fractals, Elsevier, vol. 56(C), pages 194–201. A peloton has similarly been defined "as a group of cyclists that are coupled together through the mutual energy benefits of drafting, whereby cyclists follow others in zones of reduced air resistance."Trenchard, Hugh & Richardson, Ashlin & Ratamero, Erick & Perc, Matjaž, 2014. "Collective behavior and the identification of phases in bicycle pelotons," Physica A: Statistical Mechanics and its Applications, Elsevier, vol. 405(C), pages 92–103. A peloton is a complex system, which means that collective behavior emerges from simple rules of cyclists' interactions.
Formations
Pelotons are typically observed during bicycle races in which drafting is permitted, although pelotons also form from cyclist commuter traffic.Alfredo Garcia, Carlos Llorca & Jorge Serra-Planelles (2020) Influence of peloton configuration on the interaction between sport cyclists and motor vehicles on two-lane rural roads, Journal of Transportation Safety & Security, 12:1, 136-150, DOI: 10.1080/19439962.2019.1591557Li, M., Chen, T., Du, H., Ma, N., & Xi, X. (2022). The speed and configuration of cyclist social groups: A field study. Physica A: Statistical Mechanics and its Applications, 592, 126849. Pelotons travel as an integrated unit in which each rider makes positional adjustments in response to the movements of adjacent riders and those ahead. Riders at the front are fully exposed to wind resistance, hence they experience higher fatigue loads than riders in drafting positions. After a period of time in front, leading riders maneuver farther back in the peloton to a drafting position to recover. Thus the peloton is in fluid motion as a continuous rotation of riders push from the back through to the leading edge, then falling away. Like bird flocks, peloton-like behavior that involves drafting or similar energy-saving mechanisms has been identified in a variety of biological systems.Trenchard, Hugh and Matjaž Perc. 2016. “Energy saving mechanisms, collective behavior and the variation range hypothesis in biological systems: A review.” Bio Systems 147, 40–66 .Trenchard, Hugh, Carlton E. Brett and Matjaž Perc 2017. “Trilobite ‘pelotons’: possible hydrodynamic drag effects between leading and following trilobites in trilobite queues.” Palaeontology 60, 557–569.Trenchard, Hugh. “Cell pelotons: A model of early evolutionary cell sorting, with application to slime mold Dictyostelium discoideum.” Journal of Theoretical Biology 469 (2019): 75–95 .Trenchard, H. 2012. “American coot collective on-water dynamics.” Nonlinear Dynamics, Psychology, and Life Sciences, 17, 183–203.
The shape or formation of the peloton changes according to multiple factors. Comparatively high power output efforts due to high-speeds on flat topography, a strong headwind or inclines (hills) tends to spread out or lengthen the formation, often into single file. A slow pace or brisk tailwind in which cyclists' power outputs are low result in compact formations such that riders ride side-by-side, often filling roads from one side to the other. When two or more groups of riders have reason to contest control of the peloton, several lines may form, each seeking to impose debilitating fatigue on the other teams. Fatigue is a decisive factor in the outcome of every race. Cyclists' range of peripheral vision is a significant factor in peloton formation.Belden, J., Mansoor, M.M., Hellum, A., Rahman, S.R., Meyer, A., Pease, C., Pacheco, J., Koziol, S. and Truscott, T.T., 2019. "How vision governs the collective behaviour of dense cycling pelotons." Journal of the Royal Society Interface, 16(156), p.20190197.
Thus these formations comprise two main phases of behavior: a compact, low-speed formation, and a single-file, high-speed formation. Peloton phases are indicated by thresholds in collective output that can be modeled mathematically and computationally.Trenchard, Hugh & Ratamero, Erick & Richardson, Ashlin & Perc, Matjaž, 2015. "A deceleration model for bicycle peloton dynamics and group sorting," Applied Mathematics and Computation, Elsevier, vol. 251(C), pages 24-3 The principles of phase behavior identified by Trenchard et al. have been applied to optimize engineering problems.Poitras, G., Cormier, G. and Nobelle, AS. 2018. "Novel Optimization Algorithm for Composite Steel Deck Floor Systems: Peloton Dynamics Optimization (PDO)." Building Tomorrow's Society, Canadian Society for Civil Engineering annual conference, June 13–16, 2018
Similarly, these thresholds in peloton formations define transitions between peloton cooperative behavior and free-riding behavior.Trenchard, H. 2015. "The peloton superorganism and protocooperative behavior," Applied Mathematics and Computation, Elsevier, vol. 270(C), pages 179–192 Cooperation and free-riding in pelotons have been studied using game theory and as a social dilemma,Brouwer, T., & Potters, J. 2019. "Friends for (almost) a day: Studying breakaways in cycling races." Journal of Economic Psychology, 75(part B), [102092]. https://doi.org/10.1016/j.joep.2018.08.001Hoenigman, Rhonda, Elizabeth Bradley, and Allen Lim. "Cooperation in bike racing—when to work together and when to go it alone." Complexity 17, no. 2 (2011): 39–44.Mignot, J.F., 2016. "Strategic behavior in road cycling competitions." In: The economics of professional road cycling (pp. 207–231). Springer, Cham. and have also been considered in terms of equivalencies to aspects of economic theory.Trenchard, Hugh, and Matjaz Perc. "Equivalences in biological and economical systems: Peloton dynamics and the rebound effect." PLOS ONE 11, no. 5 (2016): e0155395.
Basic peloton behaviors have also been modelled with robots,Bedruz, Rhen Anjerome, Argel A. Bandala, Ryan Rhay Vicerra, Ronnie Concepcion, and Elmer Dadios. "Design of a Robot Controller for Peloton Formation Using Fuzzy Logic." In 2019 7th International Conference on Robot Intelligence Technology and Applications (RiTA), pp. 83-88. IEEE, 2019.Bedruz, Rhen Anjerome R., Jose Martin Z. Maningo, Arvin H. Fernando, Argel A. Bandala, Ryan Rhay P. Vicerra, and Elmer P. Dadios. "Dynamic Peloton Formation Configuration Algorithm of Swarm Robots for Aerodynamic Effects Optimization." In 2019 7th International Conference on Robot Intelligence Technology and Applications (RiTA), pp. 264–267. IEEE, 2019. and principles of peloton behavior have also been considered in relation to the future of collective robot behavior.Trenchard, H., 2018. "When Robots Get Bored and Invent Team Sports: A More Suitable Test than the Turing Test?." Information, 9(5), p.118.
Models and simulations
Olds' model
Olds' analysis involved peloton breakaway and chasing groups. He identified the factors involved in determining the likelihood that a breakaway group would succeed in reaching the finish ahead of chasing groups. He identified the following critical factors: distance remaining in the race, the speed of the breakaway group, the number of riders in both the breakaway and chasing groups, how closely riders draft each other, course gradient and roughness, and headwinds and crosswinds (referred to as "demand" factors). Introducing riders' physiological variables including metabolic power production and time to exhausion ("supply" factors), Olds' presents an iterative algorithm for determining the mean power of each group and their relative times to exhaustion, thus determining whether the chasers will catch the breakaway.Olds, T. 1998. "The mathematics of breaking away and chasing in cycling". Eur J App Phsiol, 77, 492–497
Olds' key findings include that group mean velocity increases rapidly as a function of group size up to five or six riders, and then continues to increase but only gradually up to about 20 cyclists; wheel spacing is a significant determinant of group speed due to drafting advantages; mean velocity falls as a function of distance remaining; the required lead time for a breakaway group falls rapidly as the number in the breakaway group increases up to about 10 riders, but flattens as the number of riders in the breakaway group approaches the number of riders in the chasing group. Similarly, Olds' observed that if the chase-group size is less than the size of the breakaway group and the wheel spacing among the chasers is greater than 3 meters, a chasing group will never catch a lead group, assuming other factors remain constant between the groups.
Hoenigman et al.'s model
Agent-based computer models allow for any number of independent "agents" with assigned attributes to interact according to programmed rules of behavior. In this way, simulated global behaviors emerge which can be studied for their properties and compared with actual systems.Wooldridge, M. An Introduction to Multi-Agent Systems; Wiley: United Kingdom, 2009. For their cyclist agents, Hoenigman et al.Hoenigman, R., Bradley, E., Lim, A.2011. "Cooperation in bike racing when to work together and when to go it alone." Complexity 17(2) 39–44 assigned individual maximum-power-outputs over a heterogeneous range among peloton cyclists and individual and team cooperative attributes in which agents share the most costly front position, or defect by seeking lower-cost drafting positions within the peloton, both according to some probabality. Hoenigman et al. introduced power equations from the literature for non-drafting and drafting positions, an approximate anaerobic threshold as a percentage of cyclists' maximum power when traveling alone without drafting, and a time-to-exhaustion parameter. The authors also introduced a "breakaway" state in which defecting riders increase their speeds to a higher threshold either to breakaway or to catch a group ahead.
The authors performed experiments by varying the noted parameters over a simulated flat road race containing 15 teams of 10 riders. Cooperators (those willing to take the most costly front position) spend 5 minutes at the front, then rotate to the back of the pack. Defectors spend only one minute at the front. As the race approaches the end, strategies change such that each agent increases their output incrementally based on their remaining energy up to 100% of their maximum power output. Results of the model shows that weaker riders are better off defecting, while cooperation is a good strategy for stronger riders. The results are realistic when compared with real-world competitive cycling and demonstrate the effectiveness of this kind of agent-based model which facilitates accurate identification and analysis of underlying principles of system (in this case, peloton) behavior.
Erick Ratamero's simulation