How do I stop my dog bolting or escaping?

Finn is the case the concept is named from and the one it is most complete in. The equipment spec arrives in the first ten seconds of his first trip out - major flight risk, he will bolt, backup collar on, e-collar in front so the collar cannot slide off, flat collar relatively tight - and the same routine recurs weeks later as an unremarked double hook set up before leaving the vehicle. He is kept indoors around the clock because his handler does not trust him in a pen and thinks he would get out and never be seen again, and the family add, in the same exchange, that he has run out the door a couple of times. His handler wrote flight risk on the board on his first days and told the staff not to touch him. The honest limit is stated plainly: in a real bolt the job is to catch the fifty-foot line, because he is not catching the dog. And then the ranking: without a recall the program is woefully incomplete, he will try a heel but it is not the thing that will change this dog's life. What actually put him right, by the handler's own account, was not enforced at all - the check-in grew out of the long-line work and the dog worked it out himself.

Lola's file holds the behavioural half in its fullest form, and it is the only place in the corpus where the bolt is deliberately allowed to happen. Behaviour is modelled as a ranked list of options the dog reaches for, and training as moving one option down the list rather than deleting it: bolting is her go-to, so it has to produce nothing but pressure, and returning has to produce a long stretch of relief - after which, in her own terms, if I feel panic, don't bolt, come back. The repetition is found rather than manufactured: walk on a long line, watch her face, because the thought is visible before the movement. And the exercise has a ceiling - no more than two of those events in a session, because it is intense, and a dog who comes to expect four of them starts to dread going out. Her handlers also price her containment honestly: if she bolts at full speed nobody on staff could catch her, including the fastest trainer, and at the handover the moment the family are warned about is when she looks behind her and realises nobody is holding the line.

Sage supplies the design principle: build the recall the dog actually needs. Her presenting problem is stated by the trainer as random fear of loud noises with no recall once she is frightened, and he names it as the thing most likely to cause her harm. So the recall is trained for the spooked state rather than the excited one - the goal being that 'Sage, dude, like come' snaps her out of it and she learns that coming makes her safe. The same reasoning is applied to a second dog in her first session, where it is stated as a general rule: he does not need a recall for excitement, he needs to be trained that when he panics he comes back. Her file also holds the least sentimental line about what happens after the line comes off: if she runs, it is over, she is gone, and he is relying on the recall.

Kodak is where the containment is fitted fastest. Within four minutes of starting his first outing a slip lead goes on underneath the working collar, so that pulling out of the first still leaves him attached, and the handler says why in the same breath: the flightiness and the fearfulness he is showing right now, and he could easily panic. He adds that he does not do this with every dog any more, which makes it a judgement rather than a habit, and attaches a handling rule - never drop an unlocked retractable with a dog like this. A later session gives the field version: a long line even on a fresh dog, with a backup slip, a hand kept on the line, and the claim that this is what makes working a dog loose in an open field safe at all. And his file supplies the failure nobody warns owners about, which is the line itself bunching up into a ten-footer with a dog running away on the end of it.

Luna's is the case where the question is named and not answered on camera. Her defining risk is stated as flight rather than fear or defiance - her big thing is that she wants to take off and bolt - and at the second owner lesson, at the family's house, the handler says out loud that he cannot send a dog home if he is worried she will get skittish off leash and run, that he does not know which way it will go, and that he will find out the next day. The family answer in the same conversation that they hope she stays, because chasing her last time was not fun. Four days later he gauges it at ninety per cent safe, fetches the long line for the first time, and watches a bolt form in her eyes without happening. His policy, stated in that session, is that dogs only bolt a couple of times and that punishing each one extinguishes it - and no bolt is ever punished on camera, because none ever happens. He is also explicit about the limit of the tool: no amount of e-collar will override the flight impulse, and the level goes up because her nerves are higher rather than because she is bolting.

Roxy's handover holds the moment before a dog becomes a flight risk, described as something the trainer sees all the time: the look in a dog's eye the first time she considers that she could go somewhere. He says he usually sees it on a fifty-foot long line rather than on a retractable, and that a dog needs that learning experience - the one where she tries it and finds out it did not work at all. Her own episode is reported with the yelp left in. He also explains why households are blindsided: a puppy's recall looks reliable because until about five or six months she was not confident enough to test it, and then one day she is. Her case is also the one where the diagnosis is not fear at all - she arrived with no sense of context, a dog for whom any moment could be the moment to go and have fun, which is a different route to the same risk.

Louie's file gives the argument for why the recall has to be unconditional, and it is not about obedience. The scenarios named are physical and concrete - a dog chasing a bird even on a retractable and hitting the end of it too hard and hurting himself, or pulling hard enough on the other end to injure the person holding it - so the standard is that no matter what is going on, he recalls when he is asked. His placement then shows the containment version for a dog with no recall yet: the owner names the risk himself, that if he lets him go he will run away, and the answer is a tether at first so that leaving is not an option at all, with a remote collar available, while the dog learns where he lives. The freedom comes after the boundary walk rather than before it.

Zeus contributes the one piece of the equipment logic that is stated as a general rule rather than as a dog's spec: collar tightness only matters if the dog spooks easily, and it matters then because panic is how a dog gets out of a collar and disappears. The same handover argues against a harness on two grounds - dogs slip out of most of them, and you cannot steer a dog with one - and for a retractable plus a trained dog, because then the handler stops being a restrainer. It also supplies the reason knowing the commands is not the same as being safe: reliability comes from obligating them, and the household's second dog is ruled out of off-leash freedom by his own owner for exactly that reason.

King's file gives the underlying logic in its general form, and it is the one sentence that governs everything above: the cost of a single failure is what dictates the equipment, not the dog's average reliability. One moment of hesitation or fright and the dog is under a bus - which is why a professional who could take a client dog off leash in a leashed area does not. He quotes his own dog at 99.99 per cent with the squirrel, the cat and the bird named as the exception, and adds that training decays, using his own dog as the example. It is the counterweight to every off-leash claim in his own playlist.

Deuce is the thinnest entry here and belongs for one observation. His handler names the signal he looks for as the sign a dog will not take off - the moment dogs start checking up on where he is - which is the positive, equipment-free version of the whole concept. His file otherwise contributes the radius mechanics that the containment eventually hands over to, and the reminder that the hard part of a handover is physical coordination rather than understanding: getting to the collar in time.

What equipment does a dog who bolts actually need?

Redundancy, on the assumption that one thing will fail. On one dog's first outing that is a backup collar plus the e-collar worn in front, so the collar cannot slide off, with the flat collar snug. On another it is a slip lead fitted underneath the working collar, so that pulling out of the first still leaves him attached - and the handler adds a handling rule, never drop an unlocked retractable with a dog like this. A second attachment point goes on before leaving the vehicle. Fit matters specifically because of panic: one handler says he only worries about collar tightness for a dog who spooks easily, because that is the dog who can back out of a collar and be gone, and another rejects a collar whose adjustment creeps looser when you pull on it. A harness is argued against on two grounds: most are easy to slip, and you cannot steer a dog with one.

Which commands should I actually work on?

The recall, and then the recall. The trainers rank the set for the individual dog rather than selling it as a package, and for a dog who leaves they say so out loud: without a recall the program is woefully incomplete, and the heel - which they will still attempt - is not the thing that will make the difference to his life. The other half of the ranking is not a command at all: getting the dog's mindset to change about being out, seeing people, seeing other dogs, without freaking out. If you have limited time and a dog who bolts, that is where it goes.

His recall is great at home. Why does it vanish when he is frightened?

Because it was trained in the wrong state, and the trainers treat that as a design error rather than a failure of the dog. One puts it as a general rule while working a different dog: he does not need a recall for excitement, he needs to be trained that when he panics he comes back. The goal is a word that reaches a frightened dog and, in the handler's own framing of the dog's side, means coming makes me safe. So the practice has to include the state you are worried about - which is why one program deliberately makes loud noises and walks the dog past things that have startled him before, to see what his reaction is when the pressure is on.

Can you actually train a dog out of bolting?

The model the trainers use is not deletion but ranking. Bolting is a nervous dog's first reach, so the work is to move it down her list of options: make running produce nothing but pressure, make the return end the pressure and be followed by a long stretch of relief, and the dog's own conclusion becomes - in the handler's words for her - if I feel panic, don't bolt, come back. One handler adds a claim about how fast it goes: dogs only bolt a few times, and if each one is answered they stop, because it stops working. The honest caveat is in the same corpus: that claim is never demonstrated on camera, and the handler who runs the exercise caps it at two events per session because it is intense and a dog who expects several per outing starts to dread going out.

If he runs, what do I actually do?

You catch the line. Two handlers, about two different dogs, say plainly that they cannot catch the dog - one notes he is quite fast and not that fast, another that neither he nor the fastest trainer on staff could catch this dog at full speed - so the fifty-foot line exists to be stepped on rather than to hold. That is also the argument for line length: long enough to be reachable when the dog is already going. The handling that goes with it is unglamorous: lay the line out untangled before you use it and stay on top of it, because a bunched fifty-footer is a ten-footer at the worst possible moment. And when he does get back to you, whatever he did first, the arrival has to be worth having - that rule is on its own page and it matters most for exactly this dog.

Why bother, if he mostly stays close?

Because the trainers price this decision against the worst case rather than the likely one, and the worst case is not a scolding. One says that all it takes is one moment - the dog does not listen straight away, or something frightens him - and he is under a bus, which is why he does not take client dogs off leash in leash-required places however reliable they are. Another gives the smaller, more everyday version: a dog chasing a bird even on a retractable can hit the end of it hard enough to hurt himself, or pull hard enough to injure the person holding it. On that reading the recall is a safety device, and the standard is unconditional - no matter what is going on, he comes when he is asked.

Trusting a single point of attachment on a dog who panics.

Leaving the collar loose, or using one whose adjustment creeps looser when it is pulled.

Walking a flight risk on a harness, which most dogs can slip and which cannot steer them.

Planning to grab the dog. On a fit dog you are catching the line, and probably not that.

Spending the program on the heel, on a dog whose problem is that he leaves.

Practising the recall only when the dog is happy and interested, so the version that exists is the wrong one.

Running the bolt-and-recall exercise repeatedly, until going out itself becomes something the dog dreads.

Dropping an unlocked retractable, or letting the long line bunch up in your hand.

Reading a dog who has never run as a dog who will not, especially a puppy.

When something frightens him he runs rather than freezes - and he only freezes when he is cornered or on a leash.

He gets out of doors. Not once - a couple of times, and he tries to break through anything he can.

You have taken the leash off once or twice, it went badly, and you stopped trying.

He has slipped a collar, backed out of one, or jumped a fence.

You can see him thinking about it - something changes in his face before he moves.

His recall is excellent when he is happy and gone entirely when he is frightened.

He is faster and fitter than everyone in the household, and the plan has always been to grab him.