Why Do Ducks Fly in a V? What the Research Actually Shows

The usual answer is that the birds behind get a free ride in the air pushed up by the bird in front, and that the leader drops back when it tires. Most of that is probably right. Very little of it has actually been measured in ducks, and one of the most repeated figures — the claim that flying in formation saves birds seventy per cent of their energy — comes from a theoretical model published in 1970, not from watching real birds.

Here is what the evidence supports, what it does not, and which ducks fly in a V at all.

The aerodynamics, in plain terms

A flapping wing does not push air straight down. It leaves a pair of spinning vortices trailing from the wingtips, and while the air immediately behind the bird is pushed downwards, the air just outside each wingtip is swept upwards. A bird positioned in that upwash is flying through air that is already rising, so it needs less of its own effort to stay level. A bird sitting directly behind another is in the downwash instead, which is worse than flying alone.

That geometry is why the formation is a V and not a line. The following bird has to be offset sideways to reach the rising air.

The study everyone cites — and the species it was actually done on

In 2014, Nature published a study by Steven Portugal and colleagues that measured formation flight in extraordinary detail. The finding was twofold: the birds positioned themselves in “aerodynamically optimum positions, in that they agree with theoretical aerodynamic predictions,” and they also synchronised their wingbeats, flapping spatially in phase so that upwash capture was maximised through the whole flap cycle. When a bird did end up directly behind another, the phasing reversed into anti-phase — apparently to reduce the downwash penalty.

It is a beautiful result. But the birds were northern bald ibises, not ducks and not geese. And the study measured position and wingbeat timing; it did not measure energy use at all. No energy-saving percentage comes from that paper, however often one is attached to it.

The earlier landmark study, Weimerskirch and colleagues in Nature in 2001, did measure something physiological — heart rate in great white pelicans trained to fly in formation behind a boat — and reported that the birds “save a significant amount of energy by flying in formation.” The specific percentages circulating from that work come to us second-hand through later review papers, and those reviews do not agree with each other on the numbers. We are not going to print a figure we could only read in a summary of a summary.

The best direct measurement is recent, and it is about starlings

The most useful study on this question was published in PNAS in 2024, and it opens by stating the problem outright: metabolic flight costs “have not been directly measured in any of these group flight contexts.” The team flew European starlings in a wind tunnel in groups of two and three and measured carbon dioxide production directly.

The birds adopted a V motif on average, and trailing birds’ flight costs were significantly lower than their own solo benchmarks. The best-case figure was a 25% reduction in flight cost for a follower that held the best observed alignment for an entire flight.

And then the finding that never makes the headlines: followers with more positional variability “did less well, even increasing their costs above solo flight.” Formation flying is not a free gift for being behind someone. Hold the position well and you save a quarter of your effort; hold it badly and you would have been better off flying alone.

Field observations support that it is hard. One review notes that a significant number of birds do not occupy the theoretically optimal position — only 17% of greylag geese in one analysis matched it, in contrast to Portugal’s ibises, which flew almost exactly as the simulations predicted. An analysis of Canada goose formations concluded that actual efficiency improvements were only about 20% of the theoretical maximum, with the best gain at a wingtip spacing of about 0.16 metres.

Where the “71%” comes from, and what it actually measured

The very large figures trace back to a 1970 paper in Science by Lissaman and Shollenberger, and they are usually quoted wrongly. That paper estimated that a formation of about twenty-five birds could increase its range by roughly 71%. A range increase is not an energy saving of the same size, and the number came from an aerodynamic calculation of an idealised twenty-five-bird formation rather than from any measurement of birds.

So the familiar line that “birds save 71% of their energy by flying in a V” mangles the original three times over: it turns a range figure into an energy figure, it drops the flock size the calculation assumed, and it presents a theoretical ceiling as an observed result. Real flocks — flapping, drifting and constantly correcting — do not reach it.

The other explanation: they can see each other

Energy is not the only candidate for why the shape exists. A staggered formation also gives every bird an unobstructed view forward and a clear line of sight to its neighbours, which helps a flock stay together and navigate as a unit.

An engineering analysis of bird formations notes that formation geometry correlates with retinal features and the position of the eye on the head, and that observations of pink-footed geese favour the visual communication hypothesis. Both explanations require the same behaviour — tracking the bird in front — which makes them hard to separate in the field.

Bruce Batt, chief biologist at Ducks Unlimited, sets out both and declines to choose: one purpose is conserving energy in the upwash, the other is orientation and communication, and “these explanations are not mutually exclusive, and both have been backed by a variety of studies.” Collision avoidance and improved vision are also cited in the aerodynamic literature as contributing factors.

Which ducks actually fly in a V — and many do not

This is where duck-specific evidence finally exists, and it comes from the US Fish and Wildlife Service identification guide Ducks at a Distance, which teaches hunters and birdwatchers to name a species by the shape of its flock. Its summary is worth quoting in full: “Mallards, pintails, and wigeon form loose groups; teal and shovelers flash by in small, compact bunches; at a distance, canvasbacks shift from waving lines to temporary V’s.”

Species by species, from the same guide:

• Canvasback — migrates “in lines and irregular V’s”
• Redhead — “Migratory flocks travel in V’s; move in irregular formations over feeding areas”
• Gadwall — “Small, compact flocks fly swiftly, usually in a direct line”
• Scaup — “Flock movements are rapid, often erratic, usually in compact groups”
• Goldeneye — fly “singly or in small flocks, often high in the air”
• Brant — “Flight is swift, in irregular and changing flock patterns”

So the V is a diving-duck and goose habit more than a dabbler habit. The mallard — the duck most people are picturing when they ask this question — travels in loose groups, not in a neat V. The distinction between dabbling and diving ducks turns out to show up in the sky as well as on the water.

There is also a shape correction. Batt notes that linear flight formations are called echelons, that the V and the J are the most recognisable, and that “studies of several species have shown that a true V-shaped echelon is, in fact, less common than a J formation is.” Much of what people call a V is a lopsided J with one long arm.

Does the leader really take turns?

In one species, yes, and it is documented carefully. A 2015 PNAS study followed 14 juvenile northern bald ibises on a human-guided migration and framed the problem exactly as you would expect: the lead bird gets no upwash, so who flies in front? The birds spent 32% of flight time in another bird’s wake, and the correlation between how long an individual led and how long it followed was 0.78. Pairs swapped positions directly, with a median gap of one second. The authors concluded the ibises “cooperate by directly taking turns in leading a formation.”

For ducks and geese, the answer is less tidy. Batt, writing for Ducks Unlimited, states it plainly: “The leaders of formations change from time to time, but the causes, frequency and characteristics of these changes have not yet been determined.” Leader changes are seen. Whether they are fatigue-driven turn-taking, as the popular version has it, has not been established in waterfowl.

What to take from all this

The upwash explanation is sound physics and is supported by direct measurement in at least one species. The savings are real but modest and conditional — around a quarter at best, and negative if you fly the position sloppily. The enormous percentages come from a fifty-year-old model. The turn-taking story is documented in ibises and merely observed in geese. And a great many ducks do not fly in a V at all.

Which is a more interesting set of facts than the tidy version, and it fits the pattern of everything else in duck migration: the closer anyone looks, the more the simple story turns into a real one.