Saturday, December 24, 2011

Speed, Control and Perfection

First of all I really hope you weren't expecting a xmassy type post from me.  I am not a very festive person, yeah yeah I can hear the chants of Scrooge and Grinch, whatever...  So at the risk of incurring the wrath of many bloggers and dog-people in general, I don't 'get' dressing dogs up in Christmas paraphernalia, taking pretty pictures and sending them to all your friends with festive messages.  Well this might partly be due to the fact that I don't own anything christmassy, it might also have to do with the fact that if I tried to dress up my dogs they would think I have gone insane or it might have to do with the fact that I am pretty sure The Nerd would finally have me institutionalised if I do something like that...  Regardless, I hope all of you are having a jolly time over the holidays.  Me?  I would much rather talk about agility.

Let's look at the equation:

Each dog is an individual, with a temperament and structure unique to them.  So let's call dogs X, where X = infinity .

Each person is an individual, there for we can derive that each agility handler/trainer is an individual.  So let's call handler/trainer Y, where Y = infinity.

The number of training methods available to train one behaviour already are at least double figures, and every day new training methods and tools are discovered.  Add to that the fact that there are endless behaviours that will aid our dogs in agility, whether it will be directly or indirectly.  So for the purpose of my argument let's simplify Training to:  Training (Z) = behaviours + methods, there for Z = infinity.

So the end product is  ways to get it right in Agility (well HOPEFULLY getting it right in Agility) and let's call that P (for perfection).

(X) x (Y) x (Z) = P

And seeing that X=infinity, Y=infinity, Z=infinity then P=infinity³

No wonder agility is so bloody addictive.

Anyhow, within all these infinites we can (and sometimes have to) generalise.

In the last while I  have been thinking a lot about some of these generalisations.

Now before I go on, I will state categorically state that I am not saying there is a wrong or a right and success can be reached either way, this is just my opinion.

I think it is fair to say that we all aim to train for perfection.  And by that I am not implying that we want to replace our dog's brain with a circuit board and get a remote.  I am just saying that obviously we want to get it right.  While our ideas of perfection might differ, I think somewhere in all of our definitions we will find the words 'fast' and 'reliable'.  How we reach these goals differ.  Specifically I have come across people that put the emphasis on perfection, right from the word go.  People that will not reward unless the behaviour is performed 100% correctly.  That is what they aim for.

I am very different on this subject, enthusiasm and speed is a lot more important to me and I will rather build towards that perfection.  To use a recent example, when Volt started doing his first full dog walks, I was not fussed at all that he overshot some of them.  He was happy and enthusiastic and doing them with speed (well enough speed for me to be happy, considering it was the first time he did them).  My foundation was 100% and he knew his job, if he overshot because he was a little over-enthusiastic, cool.  If he overshot because he had no clue what I was asking, it would have been a different story.  I am the same with turns.  Yes my foundation, with flatwork or just on wings I train 100%, but when I start stringing things together, I don't expect perfection immediately.  If my young dogs take a few wide turns, but they are running full speed and with enthusiasm, cool.  Personally I feel that if you only reward for perfection, you will sacrifice speed and enthusiasm.  It is at this point, I suppose that I have to duck for bullets.

Some will argue that if you don't expect perfection from the word go, you will never be able to achieve it.  I beg to differ.  As long as your foundation is in place, you can definitely achieve perfection.  Those that saw Chaos when he first started, will remember the wide turns on our first courses.  Now I can turn him as tight as I want.  The most important part is that he does it with enthusiasm and never slows down.  I have just seen too many dogs lose some of there core speed, because they are too busy thinking of perfection.  Not that these dogs slow down to a snail's pace, not at all.  But some of that pure instinctual 'drive forward' speed is lost forever.  I know of a lot of dogs trained according to this 'perfection' recipe that have done very well, so once again, it is not the WRONG way necessarily.

Personally I also think this takes away some of the value of whatever reward you are using.  I am not a huge fan of food based training, yes I use it for some initial behaviours etc, but I much prefer toy training.  And I play very rough with my dogs, from the word go.  We have wrestling matches, I push them away from there toys and encourage them to come back with gusto.  Not only to create a higher value reward, but also to encourage confidence and prepare them for the 'big loud outside world'.  My dogs KNOW that they will get rewarded at some point.  When training for perfection, you inevitably lower your behaviour to reward ratio, since your dog is bound to 'get it almost right' a lot more than 'getting it perfectly right'.

How many discussions have there been about 'high value vs. low value' rewards.  The advantage of the way I do things, is that the value of the reward is NOT up to the toy that I use, it is up to me.  I can 'just throw the ball' and my dogs will be happy, or 'just tug' and this counts as low value.  The more involved I (ME, as a trainer and handler) get involved, the higher the value of the reward.  Of course this means that the neighbours, the family and The Nerd think I have lost my mind, since I get very involved when playing with my dogs and I am pretty sure I must look like a clown on happy pills, but so be it.  I can do this with any toy.  I did a little experiment yesterday, I alternated between two kinds of tug-toys, two squeaky toys and two different kinds of balls.  The three dogs I was training (Chaos, Quake and Volt), did not bat an eyelid when I switched toys and each toy was met with equal enthusiasm.  I honestly believe my 'forgiving' way of training youngsters have something to do with this.

A huge part of perfection training of course is handler mistakes.  When training for perfection you are not compensating for your mistakes.  Let's say my young dogs pull of a jump, because I got too far behind, or I tapered, I will still reward or continue the exercise.  The perfectionist trainers, will re-do it.  Obviously, I start expecting more from my older dogs, I slowly but surely build the criteria and teach them to recognise lines.  I compensate for my own weaknesses when training my young dogs and over time I will teach THEM to compensate for my weaknesses.  Sigh, yes, I have many of those.  And let's face it, we ALL forget the challenges of baby dogs, we get into a comfort zone with our older, reliable, well-trained dogs and by the time a new pup comes along... eish.

Okay well the whole 'handler mistakes' issue is a different story on its own, I am thinking of 'The Blamers' of course.  You know the ones I am talking about, it is ALWAYS the dog's fault... even when you show them a freeze frame of them clearly showing the dog the off-course obstacle there is STILL a 'yeah, but...'.  I am pretty sure you are all thinking of names now.  Anyhow that is a different blog post on it's own.

Just on a different (but actually slightly related) subject.  I could never train one of my dogs with the others running loose.  I would say 'tunnel' and have five dogs competing to go in there first.  My dogs get... erm, very enthusiastic while I am running another dog, I love this.  The Nerd doesn't agree with me on this, since most of this enthusiasm is very vocal.  I know of people that do full training sessions with their other dogs in down stays.  And while I have to admire the self-control of the poor dogs, I just cannot agree with it.  I want my dogs to be TOO excited to exercise THAT amount of self-control.  By this I am NOT saying that I don't proof my own stays and expect self-control from my dogs, but only when it is THEIR time with me.  I love it when my dogs go nutter-butters for me, that is part of the game isn't it?

Anyhow, I will now await the rain of criticism... which by the way is not a bad thing.  A lot can be learnt from constructive criticism.  For now I will smile contently and be happy about the fact that my dogs (if only them) at least think I am the super-most exciting thing in all of everness.



  1. I love this post. Can you talk more about how and what you train your dogs that helps them compensate for your weakness. thanks.

  2. Thanks Diana! Oh yes indeed, thanks for giving me ideas for blog posts!

  3. Having had nutso dogs, I don't always ask for crazy speed as the 1st thing on my list, but I do a lot of "driving back to me for a fun chase/tug game" with my foundation work, so the speed seems to arrive anyway.

    I just wanted to say - don't worry, my dogs absolutely could NOT stay while I worked the others. They can barely wait in line for their turn at the trials... and I don't care;)

  4. I have the same philosophy as you describe here when living with and training my dogs - but I confess I DO the Christmas picture. Did it one year several years ago just for something different and had so much appreciation from all ages that I kept going...