The answer
The trainers' first test is simple: did the dog care? Josi's trainer: "if you see she doesn't care well then you didn't correct her it's just that simple" - the same as a reward the dog does not want not being a reward. Roxy's go-home lesson gives the family one factor to judge by: "Does it look like it?" If the dog is still pulling while you say 'don't you ever', it did not work. Kelton, with Astro's owners, sets the bar for serious behaviour higher still: if a dog who was lunging just stops but is still amped up, that is not good enough - you want to see something like 'I'm sorry, I did it wrong'. Sage's owner is told the same: you have to see it in the dog. Two later lessons put the test in terms of what happens next. Kelton, to Beo's owner: 'if you don't see a reduction in the behavior was your punishment effective no'. Kelton (Reaper's trainer): a penalty that 'temporarily stops the behavior at the moment' but does not 'lower the frequency of the behavior' is 'an ineffective thing'.
In the September lessons it is the answer given, and given openly. With Beo, Kelton says of the collar he had just used that Beo was still fixating and barking, so 'I need to increase the level of the penalty'. Hattie's adopters describe the setting they had been using (15 on their own collar) as 'intermittently effective', failing when she is really worked up, and her trainer tells them 'clearly it was underpowered' and to 'make the penalty harsher'. Reaper's trainer adds the rule behind it: the higher the stakes, the firmer the penalty needs to be. Check the equipment: Roxy's handler discovers the collar was not on; King's trainer wonders whether the collar is too loose. Check the timing: King's trainer reviewed his own footage and found he was not leaving enough latency between cue and correction, and Major's owner is told that a spray two seconds after the barking makes no connection (watch 03:56). Check surprise: Kodak's staff notice that a correction of similar strength he had ignored before landed when it caught him by surprise. Check what you are training: Roman's trainer admits his fence correction was not the full event he wanted - Roman went back to it - and says it probably needs to be novel, and that he would normally teach 'nope' first inside a game. Change the handler rather than the dog: with Rocko forging, the trainer simply walks very slowly.
Two cautions recur. A verbal 'nope' only keeps its power while it is backed up; Astro's owner says he is now verbally responsive, and Kelton answers that is the conditioned punisher, "You've got to have it." And turning the level up does not produce a command the dog does not yet understand: Finn's trainer declines to raise it for that reason.
Said on film: all of the above. Inferred: that these are one troubleshooting habit; nobody gives them as an ordered checklist.
Angier K9 is a family-owned, play-based Board & Train serving Angier, Raleigh and the Triangle, with three full owner lessons.
What to do
Judge a correction by the dog, not by what you pressed: if he did not care, it did not happen. Before going harder, check the collar is on and fitted, check your timing - cue, a beat, then the consequence - and check whether he understood what was asked. For something serious, a dog who stops but stays amped has not taken it in. Back up your 'nope' or it stops meaning anything. Sometimes the fix is yours: walk slower, change the setup. And watch what comes next - fixing one fault can create its opposite.
What the trainers say
had to do until i saw that she cared right because that's all it is is if if you see she doesn't care
well then you didn't correct her it's just that simple it's just like if you if you give them a reward and they don't want it you didn't actually give them a reward at all
What you don't want to end up having is that down means getting popped
Um, essentially, if you want to know, all right, was the dog penalized or not, you need to think uh about only one factor. Does it look like it?
If it does, success. If it does not, it it didn't work. Right. So, if you go, don't you ever do and and Roxy's like, right? while you're holding her by the collar. Yeah, it didn't work at all.
he did the wrong thing if he just kind of stops doing it if it was really serious barking lunging
not good enough no okay not good enough if he's still amped up and like still like
It's called a condition punisher. You've got to have it.
and you see it you need to see it if she's still like it didn't affect you it didn't work at all no punishment happened right you have to see it in the dog
that one it was like the event wasn't like the full punishment that I want because he did go back to it a little bit at first and so probably need to be novel
um so teaching a conditioned punisher like that correctly with the right amount of like
he's either much tougher with the e-collar than I initially considered, or it's too loose
I was reviewing some of my own video footage and I found that I was not giving enough latency
That has to be charged on a regular basis. However, the how regular of a basis is decided by by him by him. If he starts blowing off stuff, then it's got to get charged, right?
But then yeah, that one correction caught him by surprise.
He's really intent on forging. I'm going to walk really slow. Kind of force the issue here.
So there I'm just penalizing forging.
because he knows the heel but he's uh if i speed up faster he um he falls behind he's not forging
so I don't I don't want to turn up the collar too high because if you don't get a command you're turning it up higher is not gonna get you what you want
We'll get over it. I won't press on that command again to prove to you that it's perfect, but he is doing well with it.
an effective punishment is one that is uh it works right and so if you don't see a reduction in the behavior was your punishment effective no
then i used the collar this last time yes but did he stop fixating or barking no he's still doing it right
um so then that should be an indicator right well that means that I need to increase the level of the penalty right so you have to penalize more severely for it to work
clearly it was underpowered it's just under power right yeah and so it sounds like you're trying to do the right things at the right times
right times and it's just like you almost kind of need somebody like to tell you like hey like up the pressure there like make the penalty harsher
one quality is that it needs to be effective right so that just means that um if it doesn't lower the frequency of the behavior even if it kind of temporarily stops the behavior at the moment it's an ineffective thing
the higher the stakes are the more uh firm your penalty needs to be the harsher your penalty needs to be
i want to avoid it because i think this is like enough for him to be like oh yeah i'll stop for a second but not enough for him to be like oh i don't want that to happen again
and over and over again every day forever if you do it like low quality
How it looks, dog by dog
Josi
The effect-based definition, said about Josi's attention-barking: keep going until you see she cares; if she doesn't care, you didn't correct her - just as a reward she doesn't want isn't a reward. The same logic explains why treats did not work for her later.
On film
had to do until i saw that she cared right because that's all it is is if if you see she doesn't care
well then you didn't correct her it's just that simple it's just like if you if you give them a reward and they don't want it you didn't actually give them a reward at all
Milo
Milo's file adds the sequencing that keeps a correction meaningful: the verbal 'nope' comes first, once, so it carries the correction - and the cue itself must never come to mean a pop.
On film
What you don't want to end up having is that down means getting popped
Astro
The standard for serious behaviour, set at an owner visit: if a dog who was lunging just stops but is still amped up, not good enough - that can become a habit loop, because if he enjoyed it he will do it again. What you want to see is something like 'I'm sorry, I did it wrong'. At the go-home the owner says he is now verbally responsive, which he was not before, and Kelton names why: the conditioned punisher, which you have to have and which stays strong only while it is backed up.
On film
he did the wrong thing if he just kind of stops doing it if it was really serious barking lunging
not good enough no okay not good enough if he's still amped up and like still like
It's called a condition punisher. You've got to have it.
Roman
The failure admitted on camera. Walking Roman along a fence toward another dog, the trainer uses 'nope' and a correction, and then says it was not the full punishment he wanted - Roman went back to it - and that it probably needs to be novel, delivered fresh before the dog settles into it. He adds that normally he would teach the conditioned punisher inside a game and obedience first, then use it where it needs to generalise, not introduce it in the triggering situation.
On film
that one it was like the event wasn't like the full punishment that I want because he did go back to it a little bit at first and so probably need to be novel
um so teaching a conditioned punisher like that correctly with the right amount of like
Roxy
The owner's one-factor test, from the go-home lesson: does it look like it? If the dog is still straining while you hold her collar and scold her, it did not work. The trainer adds that treats used to fail him the same way when a dog did not value them.
On film
Um, essentially, if you want to know, all right, was the dog penalized or not, you need to think uh about only one factor. Does it look like it?
If it does, success. If it does not, it it didn't work. Right. So, if you go, don't you ever do and and Roxy's like, right? while you're holding her by the collar. Yeah, it didn't work at all.
Kodak
Two corrections of similar strength, two outcomes: the one he expected he ignored, the one that caught him by surprise reorganised his attention. Surprise, not magnitude, is the observation - made in passing by staff and never developed.
On film
But then yeah, that one correction caught him by surprise.
Rocko
The handler-side fix. Rocko forges; instead of correcting harder, the trainer walks really slowly and forces the issue.
On film
He's really intent on forging. I'm going to walk really slow. Kind of force the issue here.
Bullet
The correction that worked and cost something. Forging is penalised in one session; in a later assessment pushing ahead, crowding and drifting are gone - and he falls behind when the handler speeds up. The trainer states both ends; nobody on film connects them.
On film
So there I'm just penalizing forging.
because he knows the heel but he's uh if i speed up faster he um he falls behind he's not forging
anymore like he used to he's not pushing ahead and he's not going out to the sides or crowding
King
The self-audit. When King seemed not to feel the collar, the trainer considered two explanations - tougher than he thought, or the collar too loose - and later found in his own footage that he was not leaving enough latency between cue and correction. At the owner lesson he explains why a verbal warning dies without follow-through, with the parent's-look analogy, and says how often the collar has to be 'charged' is decided by the dog.
On film
he's either much tougher with the e-collar than I initially considered, or it's too loose
I was reviewing some of my own video footage and I found that I was not giving enough latency
So, like if you tell your kids like one, two, three, or you just give them a look and then they go back and do it again and nothing happens, right?
Finn, Sage, Bo
Finn's trainer will not turn the collar higher to get a command the dog does not yet give. Sage's owner is told you have to see it in the dog. Bo's trainer, when a command looks off on a public trip, declines to repeat it just to prove it to the camera.
On film
so I don't I don't want to turn up the collar too high because if you don't get a command you're turning it up higher is not gonna get you what you want
and you see it you need to see it if she's still like it didn't affect you it didn't work at all no punishment happened right you have to see it in the dog
We'll get over it. I won't press on that command again to prove to you that it's perfect, but he is doing well with it.
Beo
The failure said plainly in front of the owner. Defining punishment for Beo's owner as anything that makes a behaviour less likely, Kelton gives the test - if you don't see a reduction, it was not effective - and applies it at once to the lesson they are in: he used the collar on Beo's fixating and barking, and Beo is still doing it. His conclusion is to increase the level of the penalty. He then weighs it by the total outcome: a momentary bad feeling for the dog against walks that are more fun, more places and more freedom.
On film
an effective punishment is one that is uh it works right and so if you don't see a reduction in the behavior was your punishment effective no
then i used the collar this last time yes but did he stop fixating or barking no he's still doing it right
um so then that should be an indicator right well that means that I need to increase the level of the penalty right so you have to penalize more severely for it to work
but I always look at like penalties as outcome-based but I look at the total outcome
Hattie
The family's collar setting (15 on their own collar) had been, in their word, 'intermittently effective': it worked on some things and not when she was really worked up. Their account includes zoomies with nipping at heels and ankles, where they used the collar; speaker turns are merged on the recording, but the reply is plain - 'clearly it was underpowered' - and the trainer tells them they need someone to say 'up the pressure there like make the penalty harsher'. The collar is not reported as used in either incident with the household cat that prompted the visit.
On film
And do you notice that it's effective sometimes and other times not effective at all?
Intermittently effective.
clearly it was underpowered it's just under power right yeah and so it sounds like you're trying to do the right things at the right times
right times and it's just like you almost kind of need somebody like to tell you like hey like up the pressure there like make the penalty harsher
Reaper
After a correction for Reaper starting a confrontation with a passer-by, the trainer explains to camera what makes a penalty work. First, it has to be effective, judged over time: if it does not lower the frequency of the behaviour, even if it stops it for the moment, it was 'not enough to move the needle'. Second, severity follows the stakes. He also says there is no need to get emotional, angry or loud to do it.
On film
one quality is that it needs to be effective right so that just means that um if it doesn't lower the frequency of the behavior even if it kind of temporarily stops the behavior at the moment it's an ineffective thing
it was not aversive at all like it wasn't enough to move the needle
the higher the stakes are the more uh firm your penalty needs to be the harsher your penalty needs to be
Bane
In protection training the trainer reads Bane's response to a collar correction as not enough: it made him stop for a second, not decide he never wanted it to happen again. He tells the owner it needs to be a big enough correction for that, and that it may help with his recall too.
On film
enough for this particular application we need to be able to make it a big enough correction where
i want to avoid it because i think this is like enough for him to be like oh yeah i'll stop for
a second but not enough for him to be like oh i don't want that to happen again that was not good
Dolce
On a walk (published 17 Jul 2026) Dolce leaves the heel to look at passers-by. The first time the trainer used only the conditioned punisher; when she did it again he backed it with the collar, because she needed a little more reinforcement there.
On film
those people. Alright let's go come on. Yeah because I I just did the conditioned punisher
Um, this time I got her for it because she needed, um, she needed to get a little bit more reinforcement there.
Dutchess
Three versions on Dutchess's film. A level that isn't getting a response is named as not high enough for her and changed rather than repeated (published 4 Aug 2026). After a down that needed reinforcing more often than it should for a trained dog, the trainer sets the next objective rather than calling it done. And at an owner lesson the test is applied to a success: she came over to a scuffle and backed off at 'nope' and the collar, where before it took grabbing her collar and hauling her away.
On film
she'll get the next one will be the e collar oh it's not high enough for her so dutch hey
Yeah. So, I had to reinforce that more than I
yeah okay so that was that was an improvement okay good good oh like i would have to before
i would have to grab her collar and like pull her away and they would still be going at it
Major
Major's owner had tried 'no', 'leave it', a collar and a spray. The trainer explains two reasons: the beeper bothered Major less than barking at passers-by pleased him, so after a while 'that don't do nothing to me'; and a spray that arrives a couple of seconds after the barking, when the delivery driver has already gone, makes no connection.
On film
getting all crazy you just nip it in the bud like i've tried everything no leave it the collar
care about this no you're gonna run for this right and so after a while he's like man i don't care
about that that don't do nothing to me i'm gonna go run that so now the punishment has to be
spraying him yeah yeah if you didn't catch him right on time when that happens then it then if
you spray him before like after way after like two seconds after even know what he won't make
Peque (PK)
Peque's handler reports the verdict on herself: Peque still sometimes reacts to dogs, and she has been told she needs a firmer punishment than she has been giving. In a park session a 'nope' that didn't produce the recall is backed with a tap.
On film
we're working on that gotta give her a firmer um uh punishment apparently i'm not doing it as firm
and I did tap it because I said nope but she didn't come so I gave it a tap
Sputnik
At a private lesson for Sputnik's household (published 16 Jan 2026) Kelton gives the cost of the weak version: fewer reps are needed if they are higher quality, but a low-quality, barely aversive penalty has to be repeated every day, forever - 'like the ten dollars', a fine too small to change anything.
On film
A lot of it will depend on the quality.
and over and over again every day forever if you do it like low quality
just keep doing it cuz they just they just it's not a big enough penalty it's
Why it works
Stated: a consequence is defined by its effect on the dog, not by the handler's intention or the dial (watch 09:40; watch 11:33). If the dog stops but stays aroused, the behaviour was still rewarding and will recur (watch 05:26). A predictable correction is absorbed; a novel or surprising one lands (watch 14:41; watch 05:16). Timing matters: too little latency between cue and consequence blurs which is which (watch 05:03). Pressure cannot supply understanding the dog lacks (watch 15:32). Inferred: that correcting one end of a behaviour can produce the opposite end (Bullet) - the sequence is filmed, the cause is not stated. Severity is tied to the stakes (watch 07:52).
Where the trainers differ
When a correction does not work, turn it up or change something else?
One view
Adjust pressure or reward: two dials, more pressure to do it, or a bigger reward afterwards.
On film
clearly knows what to do he's not doing it right so then you have two dials that you can turn one dial that you can turn is you can increase the amount of pressure that you put on him to do it and the other dial is you can make the reward bigger afterwards
um so then that should be an indicator right well that means that I need to increase the level of the penalty right so you have to penalize more severely for it to work
right times and it's just like you almost kind of need somebody like to tell you like hey like up the pressure there like make the penalty harsher
Another view
Do not turn it up for a command the dog does not yet understand; fix timing, fit, novelty or pace.
On film
so I don't I don't want to turn up the collar too high because if you don't get a command you're turning it up higher is not gonna get you what you want
He's really intent on forging. I'm going to walk really slow. Kind of force the issue here.
How to read it
The Zeus 'two dials' applies when the dog 'clearly knows what to do'; Finn's applies when he does not. Likely compatible, but never put side by side. Laid out, not resolved.
Is intensity or surprise the active ingredient?
One view
Surprise/novelty.
On film
that one it was like the event wasn't like the full punishment that I want because he did go back to it a little bit at first and so probably need to be novel
Another view
Intensity matched to the dog, charged as often as the dog requires.
On film
That has to be charged on a regular basis. However, the how regular of a basis is decided by by him by him. If he starts blowing off stuff, then it's got to get charged, right?
Common owner mistakes
- Judging the correction by what you did rather than by the dog.
On film
If it does, success. If it does not, it it didn't work.
Roxy · Watch at 11:46 ↗ - Scolding while holding the dog by the collar as she keeps straining.
On film
If it does, success. If it does not, it it didn't work. Right. So, if you go, don't you ever do and and Roxy's like, right? while you're holding her by the collar. Yeah, it didn't work at all.
Roxy · Watch at 11:46 ↗ - Repeating warnings with nothing behind them until they mean nothing.
On film
So, like if you tell your kids like one, two, three, or you just give them a look and then they go back and do it again and nothing happens, right?
King · Watch at 28:41 ↗ - Accepting a stop while the dog is still wound up after serious behaviour.
On film
not good enough no okay not good enough if he's still amped up and like still like
Astro · Watch at 5:26 ↗ - Turning the level up to get a command the dog does not understand.
On film
so I don't I don't want to turn up the collar too high because if you don't get a command you're turning it up higher is not gonna get you what you want
Finn · Watch at 15:32 ↗ - Letting the cue itself become the correction.
On film
What you don't want to end up having is that down means getting popped
Milo · Watch at 3:56 ↗ - Keeping a collar setting that only works some of the time. Hattie's adopters called theirs 'intermittently effective'; her trainer's verdict was that it was underpowered.
On film
Intermittently effective.
Watch at 12:40 ↗ clearly it was underpowered it's just under power right yeah and so it sounds like you're trying to do the right things at the right times
Watch at 24:08 ↗
Questions owners ask
How do I know if a correction worked?
Look at the dog. Roxy's trainer: one factor - does it look like it? Josi's: if she doesn't care, you didn't correct her. For serious behaviour, Kelton wants more than a stop: a dog still amped up has not taken it in.
On film
Um, essentially, if you want to know, all right, was the dog penalized or not, you need to think uh about only one factor. Does it look like it?
well then you didn't correct her it's just that simple it's just like if you if you give them a reward and they don't want it you didn't actually give them a reward at all
Should I just turn the collar up?
Not first. Trainers on film check fit and whether the collar is on, check their own timing, and ask whether the dog understands. Finn's trainer says a higher level will not get a command the dog does not yet give.
On film
he's either much tougher with the e-collar than I initially considered, or it's too loose
so I don't I don't want to turn up the collar too high because if you don't get a command you're turning it up higher is not gonna get you what you want
Why does 'no' work some days and not others?
The trainers point to two things: a warning with nothing behind it stops meaning anything, and a predictable correction is absorbed while a surprising one lands.
On film
So, like if you tell your kids like one, two, three, or you just give them a look and then they go back and do it again and nothing happens, right?
But then yeah, that one correction caught him by surprise.
My dog stopped pulling ahead and now lags. What happened?
It is on film with Bullet: forging penalised, then later lagging when the handler speeds up.
On film
because he knows the heel but he's uh if i speed up faster he um he falls behind he's not forging
The correction stops him for a moment and he does it again later. Did it work?
Not by the trainers' test. Kelton (Reaper's trainer) says a penalty that stops the behaviour at the moment but does not lower how often it happens is ineffective, and Kelton tells Beo's owner the same: no reduction, not effective. In both September lessons the next step named is a stronger penalty, weighed against the stakes.
On film
one quality is that it needs to be effective right so that just means that um if it doesn't lower the frequency of the behavior even if it kind of temporarily stops the behavior at the moment it's an ineffective thing
an effective punishment is one that is uh it works right and so if you don't see a reduction in the behavior was your punishment effective no
um so then that should be an indicator right well that means that I need to increase the level of the penalty right so you have to penalize more severely for it to work
the higher the stakes are the more uh firm your penalty needs to be the harsher your penalty needs to be