Tug-of-War with Your Dog: Myths, Rules and What Studies Show

In brief

Few games have as bad a reputation as tug-of-war. For a long time, the advice was: if your dog wins, he’ll think he’s the boss. Research sees it differently. Four things matter. First: in an experiment with Golden Retrievers, winning or losing at tug did not change the measure that comes closest to the conventional idea of dominance. Second: after play, the dogs were more attentive and obedient, whatever the outcome. Third: in an observational study of 50 dog–owner pairs, winning or losing had no consistent effect; what stood out was who usually started the games. Fourth: tug can raise arousal substantially. That is why clear rules are useful, above all a start cue, a stop cue and an “out” built through trading rather than force.

This guide shows what the studies actually measured, how to use tug as a reward, and when to be careful.

1. What actually happens in a game of tug

1.1 A game between two partners

In tug-of-war, you and your dog hold the same toy and take turns pulling. Unlike fetch, it needs both sides: the person joins in, responds and gives way.

A review of play in dogs concluded that most of the evidence supports two functions: developing motor skills and social cohesion. Most forms of play appeared to improve the relationship between dog and human (Sommerville et al., 2017). That doesn’t prove tug in particular strengthens the bond, but it doesn’t fit the idea of play as a power struggle either.

👉 The neurobiology of play and development in dogs

1.2 Why so many dogs love tug

In practice: tug contains elements that come naturally to many dogs. Grabbing, holding, pulling and shaking are parts of predatory behavior that play out in a game with no real prey involved. Some dogs still have little interest in it, and that’s perfectly fine.

In practice: how your dog tugs says a lot about his mood. A relaxed dog often growls low and playfully, wags loosely, play-bows and comes back with the toy on his own after a break. That kind of growl is not a warning sign in itself. The overall picture matters more: if the body becomes markedly stiffer, the gaze changes or your dog becomes difficult to interrupt, the interaction may be tipping into conflict. Take a break; no single signal can reliably tell you on its own whether the dog is still playful or becoming uncomfortable.

2. The dominance myth

2.1 Where the worry comes from

The rule “never let your dog win at tug” dates from a time when dog behavior was explained almost entirely through rank. The idea: every game over an object is a contest for status, and a dog who wins regularly moves up and becomes harder to handle.

This view attributes an intention to the dog that nobody has measured, and it turns a game into a question of power.

👉 Why dominance fails as a scientific concept in dogs

2.2 What the 2002 experiment tested

Rooney and Bradshaw (2002) tested the question directly. Fourteen Golden Retrievers played several sessions of tug with a person who let them either win or lose. Their behavior was then assessed in a series of tests. The 52 individual measures were reduced to nine underlying factors.

The result: the factor “Confidence,” which comes closest to the conventional idea of dominance, was not changed by winning or losing. After play, the dogs showed more obedient attentiveness toward the person, regardless of whether they had won or lost. One finding stood out: the most playful dogs showed significantly more attention-seeking after winning sessions (Rooney & Bradshaw, 2002).

That’s closer to the opposite of what the myth predicts. The authors themselves note that play signals and the temperament of the individual dog may modify these effects (Rooney & Bradshaw, 2002).

2.3 What the 2003 observational study adds

A year later, Rooney and Bradshaw (2003) filmed 50 dog–owner pairs in short play sessions where the owners chose the games, then tested the dogs for an hour on behaviors commonly linked to “dominance” and “attachment.” Dogs that played tug and fetch scored high on a factor the authors called confident interactivity. Winning or losing had no consistent effect on the test scores.

One finding is interesting for everyday life: when the dog started the majority of games, he was significantly less amenable and more likely to show aggression. That’s an association, not a cause. The authors interpret their results to mean that how a dog plays reflects his temperament and the relationship more than it shapes them. They found no evidence that games play a major role in dominance (Rooney & Bradshaw, 2003).

2.4 What this means for you

You can let your dog win. For many dogs, that’s the best part of the game, and the evidence doesn’t suggest it makes him “move up.”

In practice: what matters is not who ends up holding the toy, but whether the game has rules your dog understands. A dog who wins and then brings the toy back to you is at least showing that the shared game is not over for him.

3. Tug as a reward in training

3.1 Play can support learning

The finding that dogs were more attentive after tug (Rooney & Bradshaw, 2002) isn’t the only hint that play and learning go well together. In a study of 16 Labrador Retrievers, the dogs first learned a discrimination task. Then half of them had half an hour of play, while the other half rested. Twenty-four hours later, the play group needed an average of 26 trials to master the task again, the resting group 43. Salivary cortisol dropped after play (Affenzeller et al., 2017). The sample is small, and the study tested play after learning, not tug specifically.

3.2 Food or toy?

Is a tug toy a better reward than food? A small study with 10 pet dogs compared the two. For each dog, the researchers first identified the favorite food and the favorite toy; the options included a ball and a tug toy. Then the dogs had to work for each reward. Nine out of ten dogs worked more for their favorite food than for their favorite toy (Lazaro et al., 2023).

That doesn’t mean toys are useless as rewards. Food and toys were never offered at the same time, and dogs with a history of toy-based training were not included (Lazaro et al., 2023). What it does mean: don’t assume your dog values a game of tug more highly than a good treat.

👉 Reward value in dogs

3.3 Toys wear out

In an experiment with 16 kennel-housed Labrador Retrievers, interest in a toy declined quickly with repeated presentation. A new toy that differed in color or odor brought the interest back (Pullen et al., 2012). The study mainly demonstrates rapid habituation during repeated brief presentations. It did not test whether a tug toy at home loses value simply because it is available all day. In practice, rotation can still be useful if it increases your individual dog’s interest.

In practice: you can reserve a particular tug toy for training or shared play and rotate between several toys. Whether this actually increases reinforcement value for your dog is something to observe individually.

3.4 Reward instead of confrontation

In a survey of 364 dog owners, rated obedience was associated with the number of tasks trained with reward, not with the use of punishment. More punishment, in contrast, went with more reported behavior problems (Hiby et al., 2004). That’s an association from owner reports, not an experiment.

In practice: tug works especially well as a reward for exercises where your dog is allowed to have energy, such as a fast recall. For very calm exercises, tug can be an awkward immediate reward if it sends your dog’s arousal sharply upward. Whether food or play fits better depends on the dog, the training goal and the reinforcement value in that situation.

4. Arousal: the real sticking point

4.1 Tug winds dogs up

In practice, one important variable in tug is arousal. Some dogs become increasingly frantic and may accidentally catch a hand. A single contact like that does not tell you the dog intended aggression; it can, however, be a sign that the game has become too fast, too close or too exciting.

A finding on self-control fits here: in a study of 106 dogs, increased arousal improved inhibitory control in calm assistance dogs but impaired it in pet dogs that were more excitable to begin with (Bray et al., 2015). The study didn’t look at tug. It does show that the same level of excitement doesn’t mean the same thing for every dog.

👉 Arousal regulation in dogs

4.2 A start cue

In practice: give the game its own start cue, such as “take it” or “tug.” Only on this word is your dog allowed to grab the toy. That lets you use the game deliberately as a reward. If your dog grabs before the cue, calmly move the toy away and hold it still for a moment. When he waits, the cue comes.

Dogs that started most of the games themselves were less amenable in the observational study (Rooney & Bradshaw, 2003). That is an association, not evidence of cause and effect, and it does not mean dogs should be prevented from initiating play.

In practice: a start cue is still useful because it creates clarity and helps prevent your dog from simply grabbing a moving toy. If your dog brings you the toy and invites play, you can respond and use your start cue to begin the actual tug game.

4.3 Building the “out” through trading

In practice: the most important cue in tug is letting go. You build a good “out” through trading, not force. Here’s how:

• Take the tension out. Stop pulling and hold the toy completely still, close to your body.
• Offer a trade. Hold a good treat right in front of your dog’s nose. When he lets go, say “out” at the moment the toy comes free, and give the treat.
• Give the game back. Right afterward, the start cue often follows and the game continues. That way your dog learns that letting go isn’t the end, just a pause.
• Move the cue earlier. Once this works reliably, say “out” before you show the treat. Later, the chance to keep playing can itself be the reward for letting go.

Why not just take the toy out of his mouth? In a survey of 140 dog owners at a veterinary behavior referral service, each of the confrontational techniques, including forcibly removing items from the mouth, drew an aggressive response from at least a quarter of the dogs on which it was used (Herron et al., 2009). These were retrospective owner reports from dogs already presented for behavior problems. The study does not prove that trading prevents such responses, but it supports particular caution about confrontational removal in dogs prone to conflict.

4.4 Building in breaks

In practice: interrupt the game regularly, not just when it tips over. After a few pulls, ask for the “out,” then take a short break. Only when your dog has taken a breath does the next start cue come. At the end of the session, give a clear finish cue, such as “all done,” followed by a calm wind-down, for example with a chew or a small search game.

5. The rules at a glance

In practice: these rules come from experience, not from studies.

• Teeth on skin interrupts the game. If your dog catches your hand, even by accident, calmly say “oops,” let go of the toy or take it away, and take a short break. Don’t scold. If it happens often, arousal is too high or the toy is too short.
• A long toy. Use a tug toy long enough to leave plenty of space between your dog’s mouth and your hand.
• Pull sideways, not sharply upward. Move the toy horizontally and from side to side rather than yanking it up.
• Your dog may win. Let go now and then. If your dog comes back with the toy, you’ve done everything right.
• Short rounds. Several short sessions with breaks are better than one long one that keeps getting wilder.
• Stop before it tips over. Ideally the game ends while your dog is still easy to reach.

6. When to be careful

6.1 Puppies and teething

In practice: puppies can learn to tug gently early on. During the change of teeth, which in many dogs falls roughly between the fourth and seventh month, the gums are often sensitive, and loose baby teeth are no longer firmly in place. During this time, let your puppy hold and pull lightly rather than pulling back hard yourself. A soft, giving toy is better than a hard rope.

6.2 Neck and spine

In practice, vigorous tugging creates mechanical loads on the jaw, neck and the rest of the musculoskeletal system. We found no studies establishing the forces produced in ordinary tug or a threshold at which they become injurious. Avoid extreme lifting by the toy, abrupt direction changes and forceful swinging or jerking. For dogs with neck, back or joint problems, older dogs and dogs recovering from surgery, talk to your veterinary practice first.

6.3 Children and tug

In practice: children often pull jerkily, let go unpredictably and move quickly. That drives the dog’s arousal up, and small hands are close to the dog’s mouth. If at all, then only with older children who know the rules, supervised by an adult, with a long toy and a dog who reliably knows start, out and pause. For young children, calmer games such as search games are the better choice.

This matters most with dogs that guard things. In a clinical case series of 111 dogs that had bitten a child, food guarding was the most common circumstance for bites to familiar children, at 42 percent (Reisner et al., 2007). This was a clinic sample, and it was about food, not toys. But it shows that resource guarding around children needs to be taken seriously.

👉 Children and dogs: safety and respect

6.4 Dogs that guard things

In practice: if your dog freezes over a toy, growls seriously when you approach, or runs off with the toy and stops responding, he may be guarding a resource. Then tug is not the right setting. Practice trading without competition and get professional support.

👉 Resource guarding in dogs

6.5 Dogs with arousal problems

In practice: some dogs find it hard to come down from high arousal. They keep panting for a long time, grab at the leash or clothing, or react more strongly to other dogs afterward. In that case, keep tug very short and calm for now, or replace it with nose work.

6.6 The right toy

In practice: a good tug toy is long, sturdy and easy to grip, for you and for your dog. Throw out frayed ropes or toys with loose parts, because swallowed fibers can be dangerous. After the game, put the toy away.

👉 Choosing safe dog toys

👉 Ball games with dogs

7. What to avoid

In practice: some advice recommends correcting or punishing a dog who gets too wild during tug. That can stop a behavior in the moment. The question is what your dog learns from it, and at what cost. Confrontational corrections can add threat and conflict to the situation. Whether they damage the relationship or trust in an individual case cannot be established from the studies cited here. In practice, it is more useful to de-escalate early: pause, trade and make the rounds shorter.

The confrontational techniques in the survey by Herron et al. (2009) drew aggressive responses from a substantial share of the dogs.

Drop the rule “the human must always win,” too. It rests on the dominance idea, for which the tug studies provided no evidence (Rooney & Bradshaw, 2002, 2003).

8. A note on the evidence

Research on tug-of-war is thin and fairly old. The winning-and-losing experiment (Rooney & Bradshaw, 2002) included only 14 Golden Retrievers and captured short-term effects; whether the results apply to other breeds, to dogs with behavior problems or over months remains open. The study of 50 dog–owner pairs (Rooney & Bradshaw, 2003) is observational and shows associations, not causes. The review on the functions of play (Sommerville et al., 2017) covers play in general, not tug specifically.

The studies on reward and learning are small: 16 Labradors in Affenzeller et al. (2017) and in Pullen et al. (2012), the latter kennel-housed, and 10 pet dogs in Lazaro et al. (2023). The study on arousal and inhibitory control (Bray et al., 2015) did not examine tug. The surveys (Hiby et al., 2004; Herron et al., 2009) rely on owner reports, and the case series on bites to children (Reisner et al., 2007) comes from a clinic seeing dogs already referred for aggression. The game rules, the approach to the “out” and the tips on teething, neck and spine, children and toys are practical experience and are marked as such.

9. Conclusion

The direct studies do not support the common claim that winning at tug makes a dog more “dominant” toward a person. In the small experimental study, winning or losing did not change the measured dominance-related dimension of the dog–human relationship. After play, the dogs showed more obedient attentiveness. In practice, arousal is an important variable, and clear rules can make the game easier to manage.

The three things that make the difference:

Let your dog win without guilt. If he brings the toy back, the shared game can simply continue.

Practice start, out and pause. The “out” comes from trading, not force.

Keep an eye on arousal. Short rounds, teeth on skin interrupts the game, and extra care with puppies, children and dogs that guard things.

Key takeaways on tug-of-war with your dog

No measured dominance effect of winning. In 14 Golden Retrievers, winning or losing at tug did not change the Confidence factor, which in that study came closest to the conventional dominance concept (Rooney & Bradshaw, 2002).

More attentive after play. The dogs showed more obedient attentiveness after tug, regardless of the outcome (Rooney & Bradshaw, 2002).

Who starts the game stands out. Dogs that started most games themselves were less amenable; winning or losing had no consistent effect (Rooney & Bradshaw, 2003).

Food often beats toys. Nine out of ten pet dogs worked more for their favorite food than for their favorite toy (Lazaro et al., 2023).

Toys lose their appeal. Interest in a toy declined quickly with repetition and returned with a new toy (Pullen et al., 2012).

Arousal works differently in different dogs. More arousal helped calm dogs with inhibitory control and hurt excitable ones (Bray et al., 2015).

Working on a behavior problem with your dog? Our step-by-step dog behavior workbooks cover leash reactivity, being left alone, resource guarding, two-dog households, barking at passersby and noise fear.

References

• Affenzeller, N., Palme, R., & Zulch, H. (2017). Playful activity post-learning improves training performance in Labrador Retriever dogs (Canis lupus familiaris). Physiology & Behavior, 168, 62–73. https://doi.org/10.1016/j.physbeh.2016.10.014
• Bray, E. E., MacLean, E. L., & Hare, B. A. (2015). Increasing arousal enhances inhibitory control in calm but not excitable dogs. Animal Cognition, 18(6), 1317–1329. https://doi.org/10.1007/s10071-015-0901-1
• Herron, M. E., Shofer, F. S., & Reisner, I. R. (2009). Survey of the use and outcome of confrontational and non-confrontational training methods in client-owned dogs showing undesired behaviors. Applied Animal Behaviour Science, 117(1–2), 47–54. https://doi.org/10.1016/j.applanim.2008.12.011
• Hiby, E. F., Rooney, N. J., & Bradshaw, J. W. S. (2004). Dog training methods: Their use, effectiveness and interaction with behaviour and welfare. Animal Welfare, 13(1), 63–69. https://doi.org/10.1017/S0962728600026683
• Lazaro, X. A., Winter, J. M., Fernand, J. K., Cox, D. J., & Dorey, N. R. (2023). Efficacy of edible and leisure reinforcers with domestic dogs. Animals, 13(19), Article 3073. https://doi.org/10.3390/ani13193073
• Pullen, A. J., Merrill, R. J. N., & Bradshaw, J. W. S. (2012). Habituation and dishabituation during object play in kennel-housed dogs. Animal Cognition, 15(6), 1143–1150. https://doi.org/10.1007/s10071-012-0538-2
• Reisner, I. R., Shofer, F. S., & Nance, M. L. (2007). Behavioral assessment of child-directed canine aggression. Injury Prevention, 13(5), 348–351. https://doi.org/10.1136/ip.2007.015396
• Rooney, N. J., & Bradshaw, J. W. S. (2002). An experimental study of the effects of play upon the dog–human relationship. Applied Animal Behaviour Science, 75(2), 161–176. https://doi.org/10.1016/S0168-1591(01)00192-7
• Rooney, N. J., & Bradshaw, J. W. S. (2003). Links between play and dominance and attachment dimensions of dog–human relationships. Journal of Applied Animal Welfare Science, 6(2), 67–94. https://doi.org/10.1207/S15327604JAWS0602_01
• Sommerville, R., O’Connor, E. A., & Asher, L. (2017). Why do dogs play? Function and welfare implications of play in the domestic dog. Applied Animal Behaviour Science, 197, 1–8. https://doi.org/10.1016/j.applanim.2017.09.007

Note: This guide does not replace individual advice. If your dog seriously threatens during play, guards toys from people or has trouble settling after play, get professional support. If he shows pain in the mouth, neck or back, have him examined by a veterinarian.

Frequently asked questions about tug-of-war with your dog

Dominance, rules and safety