Loose Leash Walking

When I first trained Gus to behave politely on the leash, I trained him not to pull by popping the leash gently when he pulled. It was really just a jangle of his tags to remind him to pay attention, not a nasty yank or anything particularly painful or intimidating, and it worked. He became a dutiful non-puller on the leash. I thought I had to teach him not to pull.

When he and I went to CGC classes that were much more geared around motivating a dog with rewards and catching him doing the right thing, we went from a strong team to a happier, stronger team. He was a great teacher. He taught me that I could train him the old fashioned way, and he'd come through for me bigtime, but he also taught me that he'd do even better if I stopped worrying about showing him he was "wrong" and focused on being super clear and motivating him when he was right. It was one of my first lessons in catching my dog doing something right.

I learned that, rather than teaching a dog not to pull, I could teach him to offer a loose leash; that was a real awakening for me. That's the mindset that this process is based on: you build a connection with your dog, and you extend that connection so that she learns to pay attention to you and offer slack on the leash. You can use a modified version of this to teach your dog truly precise heeling, but this article is focused on teaching dogs, even young puppies, to stay with you and behave nicely on the leash.

Have your dog sit facing you (on leash), show her a treat, and step backwards saying "OK, follow me!" or whatever "let's go" command comes naturally to you. Walk a few steps backwards, and then give a treat as the dog follows you, marking the behavior each time with a clicker or with a happy "yes!" At first, you need a high rate of rewards so your dog realizes that it's a fun game you're playing, and it's really important keep moving backwards as you reward, rather than stopping as you give treats. The behavior you're looking to mark and reward is staying connected with you and moving at the same time. It may take you even more practice than it takes the dog to get the hang of this, but that's just fine. One mistake to avoid here is luring. If you need to show your dog a treat at first in order to get her attention, that's OK, but as soon as possible, you want to teach your dog to follow you without having to see the treat. Hold the treats in a closed hand, reach out to the dog to give one, and then bring your hand back in to your chest.

Once you can walk backwards, mark, and reward smoothly, and your dog knows to prance along and watch you, it's time for the second phase. Some dogs are ready after a few trial runs with the backwards walking, but some may require a few sessions of walking backwards with high rates of rewards before they're ready to move to the second phase. If your dog doesn't keep attention on you when you try this step, go back to the previous step and practice it a few more times with nice high rates of reward so your dog really has it down.

For the second phase, put the leash, your right hand and the treats in your left, and start walking backwards, marking and treating as before. When you have a really good connection going, pivot clockwise and reward her with the treat when she reaches desired position at your left side. When she hits the spot right at your hip, mark it with your click or "yes!" and treat while still moving. It really, really helps to make sure she gets the treat when she's in the right spot. Remember that you're trying to teach her that being by your hip is fun and rewarding, so be sure to mark and reward her when she's there. Drop your hand down right along the seam of your pants and give the treat from a backwards-facing hand. Always reward along that pant seam and with the hand that's on the same side as the dog. If you reach forward to reward her, that's where she'll learn to walk. If you reach across your body to reward, you'll teach her to forge forward and come in front of you.

Peacock and her handler learn the pivot at Paws N' Effect in 2013.

So remember, for the pivot, you need to:

  • have the leash in the hand on the opposite side from where you'll want the dog.
  • have the treats in the hand on the side the dog is going to end up.
  • mark with a "yes!" (or a click if you're coordinated enough to have a clicker in your leash hand) when the dog is in the right position.
  • reward with your palm facing backwards.
  • reward at the outside seam of your pants.
  • reward while still moving.

It will take you several tries—or more—to get the hang of pivoting and rewarding. And you're going to need the hang of it because your dog is probably going to forge ahead of you right after getting the treat. If she does, change directions, say "let's go" (or whatever you've been saying), and walk backwards. This way, you're triggering her to remember the game you already worked hard to ingrain with her as super fun. Then, when you have her attention and have rewarded her a couple of times while walking backwards, you can try the pivot again.

You can also try this work indoors and take the leash out of the equation entirely. If you're the most interesting thing in the room, you can work on walking backwards, pivoting, and rewarding without worrying about having a leash in your offside hand. You can also throw the leash over your shoulder so it's out of the way as you work. The idea is to build a relationship and a connection to keep your dog around you, so you don't need to be pulling on the leash unless you are in a larger space and you can't keep the dog with you.

Play this game a couple of times a day for about five minutes, and remember that it's a fun game you play together, not rigid training to induce compliance. If she gets bored and it's hard to get her attention, you're playing for too long. You want to end games while she still wants to play, not after she's lost interest. And if you're having trouble coordinating your feet, your leash, and your treats, you'll get good at it faster if you practice in short sessions that are less likely to induce frustrations.

Don't forget to switch sides. In competition obedience, dogs heel on the left, but in the real world, you want your dog to learn to walk on both sides. If you only practice with your dog on the left, you'll teach her to try to get over there, even when you actually want her on the right. If you plan on competition obedience some day, you can use different words for each side, for example, "heel" on the left and "side" on the right.

Also, avoid the classic mistakes that folks make with leash skills. If your dog pulls, you must show her that it does not work. If she pulls towards a person she wants to greet, you cannot let her drag you over there at all. If she pulls and gets to greet, you just told her that pulling is what you want her to do when she wants to greet somebody. If, however, she pulls and you stop and wait her out and only let her move forward when she offers loose leash, you're teaching her that offering loose leash gets her where she wants to go.

Another classic mistake is luring for too long. You can lure a little when you're first walking backwards and teaching the game in the early stages, but when the dog is learning to walk by your side, you need to avoid using the treat as a magnet. Hold treats up by your chest or in the opposite hand and then pass them to the dog-side hand and drop that dog-side hand down to the seam of your pants to reward. Then bring it right back up out of her face.

After a few weeks of short sessions, you should have a dog who thinks it's super-duper fun to prance alongside you because when she does, you become a happy treat-dispensing machine. Once she starts to get the hang of it, you slowly phase the treats by asking for a longer interval between each, but always stay generous with your praise. Make sure to practice indoors, in the backyard, and out on the street, so the sense of the game carries from place to place and new distractions don't get in the way.

Sign up with your email address to receive an e-mail notification when there is a new entry in the Journal.

* indicates required
Advertisement: