Quite often in road racing cyclists will create a C-shape curve on the road, following the path taken by the peloton leader.

A C-shape path during a Giro stage

source: youtu.be/61LHrIa-NnU?t=82

The peloton leader seems to create this curve because they're looking behind them (for some challenger they're tracking in one point classification or another, perhaps—could there be any other reason why the leader creates this path?).

But then the rest of the peloton follow the track. I have trouble imagining that the benefit of drafting is so high it outweighs the advantage of cutting it short and continuing straight. This is not quite a circle, but if it is, the extra distance is as much as 3.14/2, or 57% longer.

Why do racers follow the leader in a curve rather than take the shortest path?

Speculation #1: Might this be a sudden side-wind draft that forces everyone to lean (to the right in this case) and then turn to accommodate the leaning? If that's the case, why do they not stay (on the right, in this case)? Why do they return to the left of the road? (The road is closed for them.)

Speculation #2: Might there be an imperfection on the road that they can see, but that we can't see in recordings?

Speculation #3: "Everyone arriving within 3 seconds of another cyclist is awarded the same finish time." This rule was introduced (in 1953?) to avoid an entire peloton fighting for a slight improvement in ranking at the finish and taking undue risks of crashing the peloton. Might this rule also act here to eliminate the incentive of improving ranks?

4 Answers 4


I have trouble imagining that the benefit of drafting is so high it outweighs the advantage of cutting it short and continuing straight.

Believe it. Drafting can reduce the power needed to maintain speed by close to 50%, so there is considerable reason to continue drafting even if the lead rider takes a circuitous course.

Note also that the image you posted is foreshortened from being taken from far away with a telephoto lens. That 5m or so difference in placement from right-to-left-and-back on that road probably takes place over more than hundred meters, if not longer.

  • Good point about the foreshortening. It's definitely much more ellipse than circle in that picture. Can get a rough estimate from counting bike lengths
    – Andy P
    Commented May 23, 2022 at 19:02
  • 5
    There are maybe 21 riders. If each bike is 1.8 meter long and they leave, say, zero distance (ok, ok, Caleb Ewan proved that the distance needs to be strictly greater than zero), the total is around 38 meters. If each lane in this road is (guessing) five meters, the ratio is still about 4:1. Right then, it's definitely not a half-circle.
    – Sam7919
    Commented May 23, 2022 at 20:38
  • @Sam Road lanes are not that wide. Typical rural road lanes have about 1m on top of the maximum allowable vehicle width, which is usually 2.5m. That's only 3.5m. Autobahn lanes are a bit wider, perhaps 4m to 4.5m, so 5m is definitely about the upper bound of the widest lanes in existence. And if you look at the tire tracks on the right lane, you see that its more on the lower end of the spectrum. I'd say that 3.5m is quite a good guess. Commented May 26, 2022 at 8:47
  • @cmaster-reinstatemonica Right.. I was eyeballing. You're probably right, though it could be a street with the dreaded (for cyclists) space for parking. Either way, the point is that it takes an overwhelming increase in power for a cyclist 20-30 bike lengths behind to overtake the peloton—an amount of power that each cyclist keeps for the very last sprint, one that, if used at anytime before that, will mean that the cyclist has used up their peak energy and may be done for the day.
    – Sam7919
    Commented May 26, 2022 at 15:58
  • 1
    @Sam I was not contradicting, just remarking that 5m per lane is a significant over-estimate. An over-estimate that causes your ratio to be an under-estimate. The correct ratio would be somewhere around 6:1. Commented May 26, 2022 at 16:40

The 'leader' creates this path because they don't want other people drafting them. They want to either break the draft and create a gap (probably to attack) OR have others come around them and take the front.

The followers follow them because they don't want to give them a gap which they can use to attack with no-one in their wheel, and they don't want to overtake and be pushing the wind themselves.

In some cases it may be possible that people much further back could cut across that curve for less effort, but then they would also have given up their spot in a draft and have to 'fight' with someone further up the line to get back in the wheels.

  • 9
    Finally, you are right, people much further back could probably cut across that curve for less effort Not at the speeds pro riders move at. Riders in the pro peloton might have to almost double their power output should they take a "shortcut" and have to fight the wind themselves. Pro riders on level ground are almost certainly going over 50 kph. Drafting at that speed provides a huge advantage, which is why they just follow the leader. Good point about losing the rider behind for attacks. Commented May 23, 2022 at 19:04
  • Good point about the very high speeds involved. Although in a situation like in the picture I suspect the draft is less than ideal and somewhat less. I've edited the last paragraph a bit to reflect your input
    – Andy P
    Commented May 23, 2022 at 19:11
  • 2
    "wants to either break the draft" Neat. This may indeed be a signal that he's exhausted and is asking for a teammate to take the lead. "don't want to give him a gap" Very intriguing! They may indeed be punishing him ("oh, you wanted to lead? Sure, lead. If you no longer want to lead, go to the back.")
    – Sam7919
    Commented May 23, 2022 at 20:32
  • "_ spot in a draft and have to 'fight'_". Interesting. The etiquette, or maybe the rule, is that one cannot squeeze oneself in a peloton, even if there is a semblance of a gap. But what if a teammate is in the line-up? Couldn't that teammate slow down (safely) a tiny bit to let them in the line-up?
    – Sam7919
    Commented May 23, 2022 at 22:36
  • 2
    @Sam yes, a team mate could give up the wheel to let you in
    – Andy P
    Commented May 24, 2022 at 7:43

The shortest path isn't all that much shorter. Your original calculation of 57% more distance in the curve assumes that the lead rider does semicircle loops, actually moving perpendicular to the road at the start/end points, rather than heading toward the finish. A more modest but still somewhat extreme curve would be a quarter turn, which would still have the riders at a 45 degree angle to the direct path, riding as much across the road as down it. In the quarter turn case, the looping path is only 11% longer than the direct path. If you actually were to imagine a full circle of riders in the picture from the question, I'd estimate this is even less than a quarter turn, so the distance savings will be even less than 10%.

A good draft can easily save 30% or more in energy costs, so it's a net energy savings to take a slightly longer path with a draft, compared to a shorter path without a draft. It probably wouldn't be efficient to draft a leader doing half-circles down the road, but the actual curves are quite a bit shallower and don't add very much distance.


Not answer; only a counterpoint.

No one "took the bait" so to speak, in the question's example.

Here is an example where the two followers did decide to lead.

strategy in Giro d'Italia 2022


Gijs Leemreize (yellow jersey) is surely aware that to maintain his lead he needs to take a steep left turn, switching from the right to the left of the road.

He doesn't. He goes straight and lets Stefano Oldani (sage jersey) and Lorenzo Rota (white jersey) overtake him.

But then, in what makes his abandonment of the lead look like a poker bluff, he immediately attacks. Had he attacked while he's leading, they would have seen him rise from the saddle and would have anticipated the attack. They would have stayed on his tail, even benefiting from his draft for longer (they're still ~1 km away).

The bluff worked. Stefano Oldani seems surprised to see Gijs Leemreize take off. But Oldani does not immediately counter-attack. He only catches up and bides his time, using again another draft from Leemreize, and ultimately wins the stage.

Essentially it appears that Gijs Leemreize overestimated the energy he has left on that day. By the time Oldani catches up with him (second 42 in this video), he holds back, keeping what he has left for their ultimate sprint.

This amount of strategy in road cycling clearly makes it much more than just a "who got the most power?" game.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.