A lot of owners of adolescent dogs have quietly stopped playing. It got too rough, or too loud, or it always ended with teeth on a wrist, so the toys went in a box.
That is worth getting back. Play is where an adolescent dog will do hard things for you, and where stopping can be practised.
Play is trained, the same as anything else
Reward-based training is associated with higher reported responsiveness and fewer reported problem behaviours. A game you start and end is one of the strongest rewards you have.
Responsiveness to your cues also dips around the onset of adolescence, and returns later in the stage. That dip is easier to work with inside a game than inside a drill.
Rules of tug
Keep one toy that is only for tug, out of reach between games. A toy that lives on the floor is not much of a prize.
Start on a word, every time. Hold the toy still until your dog is still, then say it and move.
Keep the toy low and let your dog do the pulling, rather than lifting them off their front feet.
Teeth on the toy, and the game carries on. Teeth on skin or clothes, and you stand up, put the toy behind your back and count to 10.
Let your dog win, often, and let them carry it around. Winning is what makes them play again tomorrow.
You decide when it ends. Finish with the toy put away and a chew handed over, rather than when somebody has had enough.
Adults run tug, not children.
If the problem is that your dog will not let go, that is its own piece of work, and there is an article for it. Do not turn the release into a wrestling match.
The two-toy game
Two identical toys, one in each hand, in a hallway or a small garden. It builds a dog who brings things back without ever teaching them that being chased is fun.
Throw toy 1 a short distance and let your dog pick it up.
Ignore what they do with it. Bring out toy 2, make it move along the ground, and wait.
The moment your dog drops toy 1 and comes toward you, throw toy 2 the other way.
Collect toy 1 while they are gone, and repeat. 6 throws, then stop.
Nobody says a cue, nobody takes anything out of a mouth, and nothing gets chased. Dropping the first toy becomes the thing that starts the next throw.
Play breaks, where the off switch gets built
Relaxation is trained as a behaviour in its own right, in short repeatable sets below your dog's threshold. A break inside a game is where that gets practised for real.
Play for 20 to 30 seconds, then stop and stand still with the toy behind you.
Wait for your dog's feet to go still, feed 1 treat on the floor, and restart.
Aim for 4 or 5 breaks in a 5 minute game. The breaks are the training and the game is the pay.
End on a break rather than on a burst, and hand over a chew.
Working below the threshold at which your dog reacts is the condition for learning. In play that means pausing while your dog can still hear you.
Why this beats another lap of the park
You can start it and stop it, which you cannot do with another dog.
It is 5 minutes in a hallway, so it survives rain, a bad week and a dog who cannot be off-lead yet.
It gives a dog with no off switch a place to practise having one.
It is something you both enjoy, in a stage where much of the day is management.
Behaviour change is built from management plus reinforcement together, and reward-based methods are what the veterinary behaviour guidance recommends. Play is one of the few parts of the day that is reinforcement almost all the way through.
Safety note
Get help now if play produces hard grabbing at your arms, or teeth on skin you cannot interrupt. The same applies to growling that does not let up when you step away. Stop games that end that way, and speak to your veterinarian and a qualified behaviour professional who uses reward-based methods. Pain frequently presents as a change in behaviour rather than as obvious lameness, and sudden behaviour change warrants that examination.
Today's version, in one line: 5 minutes of tug tonight, started on your word, with 4 stops in it.