How long should a training session be?

Deuce's file carries the only stated schedule in the corpus and the clearest argument for it. Fifteen-minute bursts, three sessions a day, chosen so that the time is quality time. A session is then judged against what it set out to do rather than against the clock: three goals named before the work, all three ticked off, about twenty to twenty-five minutes, and the conclusion that a session does not have to be forever if it accomplished something. The ceiling is on intensity as well as length - an hour at that pitch would have cost him the dog's goodwill. And the threshold is read at its very light end: mid-session he says he worked slightly too hard, that the dog is showing very light signs of being drilled too fast, and that the fix next time is more space between repetitions rather than fewer of them. Two weeks later, with both owners handling, the same pacing rule is given to them as an instruction - stop commanding for a minute or two and just walk.

Kodak's file supplies the number that changes what a session looks like. Not how long, but how much of it is commands: ninety per cent walking to ten per cent commands early on, sometimes ninety-five to five, moving toward eighty-twenty as the dog gets comfortable, with seventy-thirty named as a really intense session. The reason is given and it is mechanical rather than sentimental - if you do not leave that much space, the freedom stops being a reward and the dog is on a recall simulator, which means the repetitions have stopped being repetitions of anything. The same session says that being overwhelmed, scared or anxious shortens a training session as surely as heat does. A second Kodak file gives the session-scale version of the same instinct: the dog is flat, so end it, put him in a kennel, and run another later to find out whether it was the dog or the day - with company floated as part of the answer.

Luna is the dog with a number of her own, and it is said out loud four times across five sessions in figures that agree with each other. She taps out at about twenty-five minutes; fifteen to twenty is the trainer's normal unit; on a hot day she gave out at nine, and the answer was to split the day rather than push the session - a long one earlier and a five-or-ten-minute catch-up later. On another walk the reason for going home is given as the weather and the same fifteen-to-twenty figure. The handover lesson then gives the family the general form of it: all three dogs he has at the moment give him about fifteen minutes, so he breaks the sessions up across the day - and the warning that comes with it, that a family with a newly trained dog will want to drill everything because it is exciting, and drilling is what shuts the dog down.

Roxy's household gives the ratio in the form an owner can actually picture: ten minutes of breaks, ten minutes of playing, against an hour of repeated commands - and the claim is not that the breaks are kinder but that they are how the dog learns better. On a field trip the same handler stops a good stretch on purpose at ten minutes and turns the dog loose to wander. Late in the program, working Ramsey from the same household, a trainer states the cost directly: overworking will not make the work hard, it will make the performance drop. The same file produces the one measurable program variable in this concept - the warm-up a dog needs before it can work shrinks week by week and has practically disappeared by go-home. And on a hot day, the session is designed around cooling rather than shortened: pools are put out, the dog is in and out of one, and the sit work starts after she has cooled down.

Lola is where the dose is rationed hardest, because the exercise being dosed is the most effective one in her program. No more than two bolt-and-recall events in a session, and the failure mode is named precisely: not that the exercise stops working, but that a dog who comes to expect four of them every outing finds going out intense, and the trainer will not spend that. The closing rule follows from the same logic - reset in the same place, send her back out to sniff, take an ordinary repetition, and only then finish, because ending on the pressure or in the truck teaches the dog where relief lives. On her intake day the general version is stated without the exercise attached: long decompression gaps, because throwing a dog at a thing over and over is the mechanism of repression rather than merely inefficient. And on a hot day in a treeless park, the stopping point is set not at distress but a couple of minutes before it, at what he calls the threshold of smooth brain.

Dolce's protection work is where the ending criterion gets said most cleanly, because on a tug a dog's willingness and the quality of the work come apart. She still wants to play; the grips have stopped being worth having; the session ends on the second fact and not the first. The house word for it appears in two separate files - cooked. A first repetition of a new high-arousal exercise is planned as one and done, stated before it happens rather than decided afterwards, and the sleeve introduction gives the general rule: a few of them and then make progress from there, because you cannot keep going on that exercise.

Sage explains why her videos look like people sitting around. With a sensitive dog the repetition count is not a target: a rep is asked only when she is back at baseline, and between repetitions there may be ten or fifteen minutes of nothing in particular. The trade is stated to her owners in the lesson - a longer program with fewer total repetitions, which is acceptable because what makes a repetition count is how clear it was, not how many there were. The rule is then given in its blunt form: fewer reps with her in a good emotional state, rather than pushing through, because pushing through does not do much good. Sessions have a named last step - walk around, recover, all pressure off - rather than simply stopping. And the go-home lesson adds the environmental cap: her outdoor sessions rarely ran past thirty minutes, because of the heat, and towards the end of each one she would be done.

Bullet's file is the one where the dosing judgement is shown from both ends and where the trainer twice names his own error. On day one, at a park, forty-one minutes is called near the limit and the session is wound up with the reason attached - going too long can actually be counterproductive on a first day. Two sessions are checked against the clock mid-session. But the other direction happens too: on an indoor place hold the dog begins to settle, and the trainer extends the exercise rather than ending it, because the duration is the experience worth having. Later, pushing a place criterion, he reads the dog as having had too much and backs off; and on an off-leash morning at the same park he loses a two-minute hold at about ninety seconds and says plainly whose fault that was - I was going for two minutes, got greedy.

Bo contributes the three words the whole idea travels on. On a public trail, after a heel block that has already loaded the dog up, a planned down repetition is cancelled - not because it went badly but because it went well enough to have cost something - and the reason given is slow is fast. The remaining time goes to decompression. It is the pacing philosophy stated as a claim about the whole program rather than about one session: doing less today is the faster route through. Bo's go-home lesson supplies a different figure entirely, and the distinction matters - the thirty-minutes-to-three-hours range there is the length of the client handover, not of a training session.

Whether the dose is a fixed number or a reading of the dog.

Nobody on camera sets these against each other, and Luna's file is the closest thing to a reconciliation: a per-dog figure is named in advance - twenty-five minutes, or fifteen to twenty - and then overridden on the day she gives out at nine. Read that way the number is a plan and the state is the veto. But the two are genuinely different instruments: a number lets a handler schedule a day with several dogs in it, which is what a board-and-train needs, and a reading does not. Which one a given session ran on is usually not recoverable from the footage.

What to do at the moment the dog goes quiet: stop, or extend.

Both positions come from Bullet's own file, ten days apart, and one of them is later marked by the trainer as his own error on a different exercise ('got greedy'). The distinction that would resolve it - extend a settle, do not extend a worked repetition - is never stated by anyone on camera, and an owner given only one half will either quit a place hold at the useful moment or grind a heel past it. See `the-settle-criterion` and `one-good-rep-then-stop`.

How long should I train my dog for each day?

The one stated schedule in this corpus is about fifteen minutes at a time, up to three times a day - and the reason given is not that more would be unkind but that it would stop being quality time. Other files put the same shape differently: about twenty to twenty-five minutes when the session had three specific goals in it, a dog who taps out at twenty-five, a handler who says all the dogs he has right now give him about fifteen minutes so he splits the day up. Nobody on camera defends a long session. The nearest thing to one is a first day at a park that ran forty-one minutes, and the trainer stopped it there saying going too long can be counterproductive.

How much of that fifteen minutes should actually be commands?

Much less than you would guess. Working a nervous dog on a trail, a trainer put a number on it: early on, ninety per cent of the session is just walking and ten per cent is commands, and sometimes it is ninety-five to five. As the dog gets comfortable that moves toward eighty-twenty, and seventy-thirty is described as a really intense session. Another handler gives it as ten minutes of work to ten minutes of play and says outright that this is how they learn better. The point is not to be gentle; it is that the freedom is what you are paying the dog with, and if there is not enough of it the dog is on what he calls a recall simulator and the repetitions stop meaning anything.

How do I know when to stop?

By the quality of the work, not by the clock and not by whether the dog still wants to carry on. The clearest version is said over a tug: she still wants to play, but he is not getting anything and it is not as solid as it was, so it ends there. The same call is made on obedience - the answers get slower, the dog is showing very light signs of being drilled too fast, he is panting and slowing down in the heat. With a sensitive dog the rule is put the other way round: fewer repetitions with the dog in a good state beats pushing through, because pushing through does not do much good. If you are still unsure, the useful question is whether the next repetition is likely to be as good as the last one. If it is not, you have already had the last one.

He has gone flat halfway through. Should I push on and finish properly?

No. Two different answers in this corpus, and both of them are to stop. Outdoors, the answer is to lower the criteria, take a shorter and easier repetition, and give the rest of the time to decompression - slow is fast. Indoors, the staff answer is more mechanical: end the session, put him in a kennel so he can rest, and run another one later to find out whether the flatness was the dog or the day, with company considered as part of the answer. What nobody does is add repetitions to a dog who has stopped being present for them.

My dog just came home from a board-and-train. How much should I practise?

Less than you will want to. The handover lessons say this explicitly, and they say it because it is what families do: the commands are new and exciting, so you run them again and again, and drilling them over and over is what makes the dog shut down. The advice in the same breath is to break the work up across the day rather than to do it all at once, because these dogs give about fifteen minutes each. There is an alternative to practice sessions altogether, and it is the better answer for most households - use the commands when there is an actual reason to, rather than setting aside a slot for drilling them.

It is hot. Should I skip training?

Not skip it - redesign it. On camera the heat is treated as a fact about the dog's biology that shortens the session, not as an excuse being made for him, and the responses are practical and varied: put big pools out and let her cool down before the sit work starts, loop the walk back into the shade rather than carry on, move a flat-faced breed's session earlier in the day, and expect the outdoor work to end at around thirty minutes because the dog will be done. One trainer names the general principle: being overwhelmed, scared or anxious shortens a training session in the same way the environment does.

Is there anything I should only do once or twice in a session?

Yes, and the trainers are stricter about it than about anything else. Anything with real intensity in it gets a hard ceiling. On Lola's bolt-and-recall work the limit is two per session, and the reason is not that the third would fail but that a dog who comes to expect four of those every time she goes out finds going out intense. In the protection work a first repetition of a new exercise is planned as one and done before it happens, and the sleeve introduction is given the same rule - a few of them, then move on, because you cannot keep going on that exercise. Ordinary obedience is dosed more loosely, but the same principle applies at the end: do not finish on the hard thing, and do not finish by putting the dog straight in the car.

Drilling the new commands when the dog comes home, because it is exciting and it works - which is precisely what shuts the dog down.

Treating an hour of repeated commands as more training than twenty minutes with breaks in it.

Filling the whole walk with commands. If the dog never gets long enough to reset, the freedom stops paying for anything and the repetitions stop being repetitions.

Reading the dog's enthusiasm as capacity. She may still want to play long after the work has stopped being worth anything.

Repeating the hard repetition instead of lowering the criteria and shortening the next one.

Ending a hard session by putting the dog straight in the car, which teaches him that the car is where the relief is.

Working through the heat rather than designing the session around it - moving it earlier, looping back into shade, or putting the dog in water first.

Your training sessions end because you have run out of things to practise, not because the dog has run out.

He was good for the first ten minutes and then got sloppy, and you kept going to fix the sloppiness.

The walk has become a series of commands, and he no longer seems pleased to be let off them.

He has started to be reluctant about going out at all, and the outings have had a lot of hard moments in them lately.

The dog came home from a program doing everything, and a fortnight of enthusiastic practice later he is doing less of it.

In summer he stops working after a few minutes and you have read it as stubbornness.