Skip to content
NatureHQ

Ecologyecological relationship

Predation

Where hunts have actually been counted, most of them fail. The famous capabilities — the speed, the venom, the bite — are rarely what decides it.

A hunt is a sequence: find, approach, attack, subdue, keep. It can fail at any stage, and where success has actually been measured, most attempts fail. The capabilities predators are famous for are rarely what decides the outcome.

The useful way to think about predation is as a chain of separate problems, because each link fails independently and a predator is only as good as the weakest one. Finding prey costs time and gives away position. Approaching means getting close without triggering an escape, which is why so much of predation is about concealment rather than performance. The attack itself is usually brief. Subduing is where a surprising number of hunts are lost, and it is where the risk to the predator sits: a hunting animal can be injured, and an injured predator may not eat again. And then there is keeping the meal, which for many predators is a real problem — a hyena or a lion or another group taking the kill is not a rare misfortune but a routine cost. Two things follow. First, the numbers are lower than people expect. Where hunts have been counted properly the success rate is usually well under half, and often far under. Second, and less obvious, most of a predator’s effect on prey does not run through the animals it catches. Prey adjust where they feed, how much they look up, when they move and how densely they group, and those adjustments cost them whether or not any attack happens. A predator that catches very little can still reshape where its prey can afford to live. NatureHQ publishes success rates with the predator, the prey, the habitat and the study attached, and does not rank predators against each other. A rate measured on one prey species in open country tells you almost nothing about the same animal somewhere else, and a league table built from such numbers would be comparing things that were never measured the same way.

Developed coverage · 79% complete · reviewed 2026-09-03

What this page covers

Predation occurs in every animal group and in fungi, plants and single-celled organisms. The quantified hunt data are concentrated in large mammals and birds, because those are the hunts an observer can watch from start to finish.

Often confused with: Scavenging, where the animal was already dead and something else killed it; Parasitism, where the host is normally kept alive; Herbivory, which is eating a living thing that does not usually die of it

Quick facts

A hunt is a sequence
Find, approach, attack, subdue, keep — each can fail alone
Most attempts fail
Quantified success rates are usually well under half
Handling costs time
Eating one thing is time not spent finding the next
The bigger effect is fear
Prey pay for risk continuously, whether or not anything is caught

Where this appears

Every organism below has been linked to this page because the evidence links them. Each one carries its own evidence, and its own limits.

Five problems, not one

And a predator is only as good as whichever link is weakest.

A hunt is a sequence — detect, approach, attack, subdue, keep — and it can fail at every stage. Where success has been measured properly, most attempts end without a kill.

Established

Specialists would state this without hedging. Multiple independent lines of evidence agree.

Predation proceeds through separable stages of search, encounter, attack, capture and handling, each with its own probability of failure and its own time cost. Measured per-attempt success rates in quantified field studies are typically well below fifty per cent and vary with predator, prey, habitat and social context.

Who this applies to
A general description of how predation works, supported by quantified hunt series in carnivores, birds and invertebrates.
Studied in
Animalia
Why we rate it this way, and what the caveats are
EstablishedHigh confidence

The stage framework is standard and the failure rates are directly observed. What varies between studies is the exact rate, not whether most attempts fail.

How far it can be extended

The stage structure was established across independent systems, and per-attempt success has been quantified in many predators.

Caveats

  • Success rates are not comparable between species without their conditions. A rate measured on one prey species in open habitat says little about the same predator elsewhere.
  • Observed hunts are biased towards those an observer could see, which in closed habitats means the record is incomplete.

Still unanswered

  • How much of the variation in success between individuals is skill acquired with age rather than circumstance.

Last reviewed 2026-09-03

The evidence (3 studies)
The stages of a hunt, and what goes wrong at each
StageThe problemHow it typically fails
FindLocating prey without wasting the dayNothing is there; searching costs more than the meal returns
ApproachClosing distance without being detectedPrey sees, hears or smells the predator and leaves early
AttackCommitting at the right momentLaunched too far out, or the prey turns inside the strike
SubdueHolding something that is fighting backPrey escapes the grip — and the predator can be injured here
KeepNot losing the meal to something elseScavengers and rivals take the kill

The fourth row is the one most often left out, and it is where the asymmetry of the whole business shows. Prey that fails is dead; a predator that fails is hungry. But a predator that fails badly can be injured by prey defending itself, and for a solitary hunter an injury is frequently fatal at one remove. That is why so many predators break off attacks that look winnable, and why hunting the largest available prey is not simply the best strategy.

Diagram

Five problems, and any one ends the hunt

Each stage fails independently.

Five problems, and any one of them ends the huntFindNothing thereApproachSeen too earlyAttackPrey turns inside itSubdueEscapes the gripKeepLoses it to a rivalWhere hunts have been counted, most attempts end at one of these.Subduing is where the predator itself can be injured — which is whybreaking off an attack is often the right decision, and counts as a failure.
The same explanation in words

Five boxes in sequence — find, approach, attack, subdue, keep — each with the way it typically fails written beneath it: nothing there; seen too early; prey turns inside it; escapes the grip; loses it to a rival. Where hunts have been counted, most attempts end at one of these. The subduing stage is marked as the one where the predator itself can be injured, which is why breaking off an attack is often the correct decision and is counted in the statistics as a failure.

How often it works, and why we will not rank them

The numbers exist. Comparing them across species mostly does not.

Quantified hunting success exists for a modest number of predators, and it is consistently lower than the reputation. That is worth stating plainly because the popular framing runs the other way: predators are described by what they can do at their best, and the best is rare. When five wild cheetahs were collared and 367 runs recorded, most of those runs ended without a capture — in the animal most often called the perfect hunter.

What NatureHQ will not do is turn those numbers into a league table. A success rate belongs to a predator hunting a particular prey species, in particular terrain, at a particular group size, recorded by a particular method — and studies differ on every one of those. Two rates measured that differently are not comparable, and a "most successful predator" ranking built from them would be an artefact of who studied what. Where this site gives a rate, it gives the conditions with it.

How we know

Measuring the whole hunt instead of the top speed

The cheetah’s reputation rests on a maximum speed measured in a handful of straight-line trials. What does the animal actually do when it hunts?

Five wild cheetahs were fitted with custom collars combining GPS with inertial sensors, recording speed, acceleration, deceleration and turning continuously. Three hundred and sixty-seven hunting runs were captured in the field, over natural terrain and against real prey.

What happened

Most runs peaked well below the species maximum. Deceleration exceeded acceleration in magnitude, and hunting success was associated with deceleration and turning performance rather than with peak speed.

What it shows

That cheetah hunting is a manoeuvring problem rather than a pure speed problem. The famous number is real and describes something the animal seldom does — which is a general lesson about capability figures measured under artificial conditions.

What it does not show

It does not lower the cheetah’s maximum speed, which is not in dispute. Five individuals in one population is a small sample, and how much of the result reflects this habitat’s terrain rather than the species is untested.

The controls — what makes this evidence rather than a story
  • Whole runs recorded rather than sampled segments, so a peak cannot be mistaken for a typical value.
  • Collar accuracy quantified independently and reported, since the entire result depends on the speed estimates.
  • Many runs per individual, separating what a cheetah can do from what it usually does.

From Locomotion dynamics of hunting in wild cheetahs

Diagram

What a run of hunting attempts looks like

Schematic, not a measured rate for any species.

What a run of hunting attempts usually looks likeFilled circles are successful attempts. Schematic, not a measured rate.Rates differ by predator, prey, habitat and group size.
The same explanation in words

A grid of forty circles representing hunting attempts, of which four are filled to mark successes and thirty-six are left open. The figure is schematic and carries no species or number: it illustrates that failure is the normal outcome rather than reporting a rate. Measured success rates differ by predator, prey, habitat and group size, and are not comparable between studies, which is why this site publishes rates with their conditions attached and does not rank predators against each other.

The part that happens with no attack at all

Prey pay for a predator continuously, and the bill arrives whether or not anything is caught.

Most of what a predator does to a prey population happens without a single kill. Animals feed less, move differently and stay more alert wherever risk is higher, and those changes have real costs.

Established

Specialists would state this without hedging. Multiple independent lines of evidence agree.

Predation risk induces measurable changes in prey habitat use, activity timing, group size, vigilance and foraging rate. These non-consumptive effects impose energetic and reproductive costs independent of mortality, and in several systems account for a substantial share of the predator’s total effect on prey populations.

Who this applies to
Documented across mammals, birds, fish and invertebrates.
Studied in
Animalia
Why we rate it this way, and what the caveats are
EstablishedHigh confidence

That prey change behaviour under risk is beyond dispute and experimentally demonstrated. The share of total population effect attributable to it is the part that varies by system and remains argued.

How far it can be extended

Risk-induced behavioural change has been measured independently in many taxa and habitats, including under experimental manipulation of perceived risk.

Caveats

  • How much of a predator’s population-level effect runs through behaviour rather than mortality is genuinely contested and differs between systems.
  • Perceived risk and actual risk can diverge, and most experiments manipulate the former.

Still unanswered

  • Whether prey correctly calibrate risk to actual danger, or systematically over- or under-respond to particular predator cues.

Last reviewed 2026-09-03

The evidence (2 studies)

Where predation connects

What is not settled

  • How much of the difference in hunting success between individuals is skill?

    Why it matters: If success is largely learned, then an experienced predator is a different ecological actor from a young one, and populations losing older individuals lose more than their number suggests.

    What would settle it: Following identified individuals across years with enough observed hunts each to separate improvement from circumstance — which very few studies have managed.

  • How often do predators break off attacks they would have won?

    Why it matters: Abandoned attacks are counted as failures, but a predator avoiding injury is making a different decision from one that was outrun. Conflating them inflates apparent failure and hides how much risk shapes hunting.

Claims about this, checked

Things people have heard, and what the evidence actually supports.

The research behind this page

35 studies, newest first. Each one has a page explaining what it found and what it could not show.

2018Nature

Biomechanics of predator–prey arms race in lion, zebra, cheetah and impala

Predators had higher maximum speed, acceleration and deceleration capacity than their prey, but hunts were usually won or lost on manoeuvre rather than on raw performance, and prey succeeded often by turning inside a faster pursuer.

2017eLife

Nematophagous fungus Arthrobotrys oligospora mimics olfactory cues of sex and food to lure its nematode prey

Trap-bearing cultures released volatiles that attracted nematodes, including compounds resembling nematode sex pheromone and food-associated odours.

2016PLOS ONE

Zebra stripes through the eyes of their predators, zebra and humans

Stripes are indistinguishable to lions and hyenas beyond roughly fifty metres in daylight and at much shorter distances at twilight or night, by which point the predator has almost certainly detected the zebra by other means.

2015Journal of Experimental Biology

Snake constriction rapidly induces circulatory arrest in rats

Arterial pressure halved within about six seconds of constriction beginning, while venous pressure rose sharply.

2014Nature Communications

The function of zebra stripes

Striping was consistently and strongly associated only with the distribution of biting flies.

2013Nature

Locomotion dynamics of hunting in wild cheetahs

Cheetahs rarely ran near their maximum speed.

2013Biology Letters

Cheetahs do not abandon hunts because they overheat

Body temperature rose only modestly during hunting and did not reach the levels the overheating hypothesis requires.

2013Current Biology

Nematode-trapping fungi eavesdrop on nematode pheromones

The fungi produced traps in response to the pheromones alone, without any nematode present.

2012Behavioral Ecology

Nonlinear effects of group size on the success of wolves hunting elk

Hunting success rose sharply from one to about four wolves and then levelled off, with additional wolves adding little.

2012Biology Letters

Snake modulates constriction in response to prey’s heartbeat

Snakes constricted longer and applied more pressure when a heartbeat was present, and reduced their effort shortly after the simulated heartbeat stopped.

2012Science

Predatory fish select for coordinated collective motion in virtual prey

Selection by real predators drove the virtual prey towards greater attraction and alignment, producing coordinated collective motion.

2010Ecology

Are wolves saving Yellowstone’s aspen? A landscape-level test of a behaviorally mediated trophic cascade

Aspen were not recovering preferentially in high-risk areas.

2009Philosophical Transactions of the Royal Society B

Animal camouflage: current issues and new perspectives

Camouflage mechanisms differ in which stage of visual processing they defeat.

2009Annual Review of Entomology

Manipulation of host behavior by parasitic insects and insect parasites

Behavioural manipulation is achieved by several distinct routes — direct injection into the nervous system, secretion of neuroactive compounds, and alteration of host physiology — and the neural mechanism is well characterised in very few systems.

2004Mycologist

Morphogenesis in the nematode-trapping fungus Arthrobotrys oligospora: an extensive plasticity of infection structures

Traps are formed in response to the presence of nematodes rather than produced continuously.

2004Journal of Theoretical Biology

Obligate vertebrate scavengers must be large soaring fliers

Living entirely on carrion is energetically viable only for animals that can cover large areas at very low cost — that is, large soaring fliers.

1999BioScience

Predator-prey arms races: asymmetrical selection on predators and prey may be reduced when prey are dangerous

Where prey are dangerous, a failed attack can be fatal for the predator too, and the life–dinner asymmetry weakens.

1996BioScience

Challenges in the quest for keystones

The term had been applied to species of every abundance and to effects of every size, including cases where the species was simply dominant.

1994Animal Behaviour

Cooperative hunting in wild chimpanzees

Taï chimpanzees hunted in complementary roles, success increased with the number of hunters well beyond the point seen in other populations, and meat was preferentially shared with those who participated.

1994Princeton University Press

Parasitoids: Behavioral and Evolutionary Ecology

Parasitoids occupy a distinct category between predators and parasites: a single host is consumed, as in predation, but slowly and from within, as in parasitism.

1992Behavioral Ecology and Sociobiology

Cooperative hunting in lions: the role of the individual

Individual lionesses occupied consistent positions across hunts — some habitually took wing positions and drove prey, others habitually took centre positions and waited — and hunts in which individuals occupied their preferred positions were more successful.

1991The American Naturalist

Complex trophic interactions in deserts: an empirical critique of food-web theory

The real web was vastly more species-rich and more connected than published webs suggested, with widespread omnivory, cannibalism, feeding relationships that change with life stage, and species occupying several trophic levels at once.

1990Canadian Journal of Zoology

Behavioral decisions made under the risk of predation: a review and prospectus

Prey animals adjust habitat use, group size, vigilance, activity timing and feeding rate in response to predation risk, and these adjustments have measurable costs even when no predator ever attacks.

1988The American Naturalist

The evolution of cooperative hunting

Group hunting is frequently stable not because the group is more efficient but because a non-participant would still obtain a share, and cooperation in the strict sense — where individuals do better by coordinating than by hunting alone — is harder to demonstrate than group hunting itself.

1986Animal Behaviour

Oddity and the confusion effect in predation

Bass capture success fell sharply as shoal size increased, and rose again when odd-looking individuals were present.

1979Proceedings of the Royal Society of London B

Arms races between and within species

Selection on prey to escape is typically stronger than selection on predators to catch, because the two sides are not playing for the same stakes — the life–dinner principle.

1974Science

Sea otters: their role in structuring nearshore communities

Islands with otters had few and small urchins and extensive kelp.

1973American Journal of Physiology

Temperature regulation and heat balance in running cheetahs: a strategy for sprinters?

The treadmill-running cheetahs stored heat rather than dissipating it, and refused to continue at around 40.5 °C, which the authors proposed as a limit on sprint duration.

1972University of Chicago Press

The Spotted Hyena: A Study of Predation and Social Behavior

Spotted hyenas hunted the majority of what they ate, and lions frequently appropriated hyena kills.

1971Journal of Theoretical Biology

Geometry for the selfish herd

Individuals reducing their personal risk by moving towards others produce an aggregated group as a by-product.

1969The American Naturalist

A note on trophic complexity and community stability

Some species have effects on community structure disproportionate to their numbers or biomass, so that their removal changes the community more than their abundance would suggest.

1966The American Naturalist

Food web complexity and species diversity

On the removal plot, mussels expanded and progressively monopolised the rock.

1964Science

Nematode-trapping fungi

Several unrelated soil fungi produce specialised structures for capturing nematodes.

1959The Canadian Entomologist

The components of predation as revealed by a study of small-mammal predation of the European pine sawfly

The number of prey taken rises with prey density and then levels off, because every capture consumes handling time that cannot be spent searching.

1927Sidgwick & Jackson

Animal Ecology

Animal communities can be described by who eats whom, with characteristic patterns in the numbers and sizes of organisms at successive feeding levels.

This page is a stop on a longer route

A guided journey reads several subjects in a deliberate order, with an argument for why one follows another. You can join in the middle.

Where to go from here

Each of these follows from something on this page — a relationship in the evidence, a claim people ask about, or the next mechanism along.

How complete this page is, and what it is still missing

NatureHQ publishes its own gaps. This page is at 79% completeness against what we would call a finished subject, and was last reviewed on 2026-09-03. It carries 17 claims and answers 3 mapped search questions.

  • no research from the last few years is attached — check for newer work
  • Invertebrate and marine predation are under-represented here relative to their share of the phenomenon.
  • Scavenging is distinguished from predation but is not covered in its own right.
  • Quantified success rates exist for far fewer predators than readers expect, and this page cannot fix that.