Structure and Obedience

How to Teach a Rock-Solid Down-Stay (Duration That Actually Holds)

A stay is worthless if it falls apart the moment you walk away or the doorbell rings. Here is how to build one that holds in the real world.

Christopher Lee Moran · Dog Trainer, Instinctual Balance · Updated June 3, 2026 · 9 min read

What a real down-stay actually is

A down-stay is a dog parked in a calm down until you release it. Not until it gets bored. Not until the cat walks by. Until you say the word. That gap between the dog’s idea of “enough” and yours is the entire skill, and it is one of the load-bearing pieces of a structured home. The bigger picture lives in our structure and house rules guide, and a stay is where most owners feel the payoff first.

Here is the honest part. A stay is not one trick. It is three separate skills wearing the same name: holding for time, holding while you move away, and holding while something interesting happens. Train them as one blob and the whole thing collapses. Train them one at a time and they stack.

Calm is earned in this exercise more than any other. A dog that can lie still and do nothing while the world moves is a dog that has learned to settle its own engine. That is worth more than a hundred party tricks.

How do I teach a dog to stay, step by step?

You need a hungry dog, a handful of small treats, and five quiet minutes. No doorbells, no kids, no other dogs. Start where it is boring on purpose, because boring is easy, and easy is where reliability gets built.

The six steps

1. Lock in the down. Ask for a calm down and wait for the dog to fully settle, hips rolled, not a tense sphinx ready to launch. If the down is shaky, fix that first.
2. Pay the hold. The second the dog is down, mark with your word (“yes”) and drop a treat between its front paws. Feeding in position teaches the dog that staying down is what pays, not popping up to you.
3. Add duration. Stretch the hold a few seconds at a time. One second, then three, then eight, then fifteen. Build to a calm one to two minutes at your side before you move a single step.
4. Add distance. Take one step back, step right back in, reward. Grow distance slowly, and drop your duration back to a couple seconds whenever you add steps. Then rebuild the time at the new range.
5. Add distraction. Drop a toy, have someone walk past, bounce a ball, start easy and close. Reward heavily for the dog that holds through it.
6. Release on purpose. End every single rep with one word, like “break” or “free.” The stay is over when you say it, never when the dog decides.

Keep each session to five minutes and quit while the dog is winning. Ten good reps beat thirty sloppy ones. The American Kennel Club’s overview of teaching the stay command follows the same duration-distance-distraction order, which is no accident: it is just how dogs learn.

Why does my dog break the stay when I walk away?

This is the most common wall owners hit, and the cause is almost always the same: distance got added too fast. To a dog, you walking away looks like an invitation to come along. Leaving is the hardest variable in the whole exercise, harder than time, harder than distractions.

The fix is not more nagging. It is smaller steps. Literally. Go back to one step away, reward, and return. Then a half step further. A dog that holds at twenty feet was built one step at a time, not in one heroic walk across the yard.

Duration versus distance, side by side

People lump these together and wonder why progress stalls. They are different problems with different fixes. The table below shows how to read which one is breaking.

One more honest note on distance. If you plan to layer in off-leash reliability later, the foundation stay has to be bulletproof first. The how and when of that lives in our guide on introducing an e-collar, and the short version is: tools come after the dog already understands the behavior, never before.

How do I proof a stay against real distractions?

A stay that only works in a silent kitchen is a parlor trick. Real life has doorbells, squirrels, and dropped food. Proofing means deliberately rehearsing the stay against the stuff that normally breaks it, in tiny, controlled doses.

Start with weak distractions and grow them. A toy set on the floor before a toy bounced. A person standing before a person walking before a person jogging. Reward the dog generously every time it chooses to hold instead of chase. You are paying the right decision so it becomes the default.

Change the location too

Dogs do not generalize the way we assume. A stay learned in the living room genuinely feels like a new exercise in the front yard, the park, or a friend’s house. Stanford and other learning researchers call this context-dependent learning, and the practical takeaway is simple: reteach the stay, easier and shorter, in each new spot. Petting a dog through this stage instead of correcting it pays off, and resources like PetMD’s stay training overview walk through the same proofing logic.

What mistakes wreck a down-stay?

Most failed stays are not stubborn dogs. They are predictable handler errors, and they repeat across nearly every owner I work with. Knowing them in advance saves weeks.

The five that show up most

• No release word. If the dog never learns when the stay ends, it makes up its own ending. Pick one release word and use it every rep.
• Pushing all three variables at once. Far away, long duration, and a squirrel in the same rep is a setup to fail. Raise one thing at a time.
• Repeating the cue. Saying “stay, stay, staaay” teaches the dog the word is background noise. Say it once and back it up.
• Calling the dog to you to reward. That pays leaving the spot. Walk back and feed the dog where it is lying.
• Ending on a failure. If the dog breaks, make the next rep easy enough to win, then stop. Quit on a success, not a flop.

None of this is fast, and anyone promising a one-day stay is selling something. Reality over fantasy: most dogs reach a calm one to two minute stay within a week or two, and a distraction-proof version takes one to three months of short daily reps. The owners who stay consistent get there first, every time.