Running Contacts Madness Continues

We haven’t been training much, but Java was hitting her mat like a pro. Any distance, any place. True, we only used racing to a food bowl as a reward, but it looked awesome. Until she got back and tummy problems and it all went downhill. 😦

We didn’t do any RC training for 3 weeks. Even though Java seemed fine in the evenings I didn’t want to risk making her uncomfortable and getting that associated with training. Then after three weeks we tried again and – BAM – Java hit perhaps 50% of the time. Say whaaaaaaaat? And it wasn’t just the first session. It looked like she wasn’t even trying.

Concluding that the dog isn’t trying is a very risky thing – it could be very unfair to them if they actually WERE trying and they can’t tell you you’re wrong. So I brushed it aside. Surely she just forgot how to adjust (as weird as that seemed to me). She used to do best if we did one session every day, we’ll do that for a week and she’ll be fine.

Then I got a little lazy busy and we only did two or three sessions that week. Not much improvement.

Luckily our weekend getaway place had a perfect lawn with a perfect little slope so I could even start doing hill RC work if hits got close to 100% again.

Unfortunately, we had trouble on the road and arrived after dark. Next day, it rained all day, but cleared at night. I don’t know what got into me, but I decided to do our first RC training in this new place in the dark, on a slope next to the cottage, lit only by the porch lights. Java did BEAUTIFUL hits! She was excited, she was fast, covered the mat nicely with her stride and only leaped over it once. I was feeling very excited about our next training!

Next day we woke up into a sunny morning. The lawn was waiting for us. Java took food in the morning, didn’t seem nauseous, we were good to go. First try, poor speed, leaped over. Next try, leaped over. I dragged the mat further down so it was on the flat. I got some hits, but also some leaps. Disappointed.

I got Ruby out, did some sends to the mat, impulse control, threw toys for him and had a blast. He is such a fun dog to train 🙂 Then I thought what the heck, this RC training is going so poorly I might as well break the rules and reward with toys even though she’s likely to look back at me IF she’s going to hit at all.

First try, great speed, but leaps over, looking back at me. Next try, hit, possibly looks back a bit while moving forward, gets toy thrown. Next try, another beautiful hit, focusing forward more. And so on, about 8 hits in 10 tries. 😀

I didn’t get it. What happened here? Food is not working anymore, but toys are? And why did food work great on Friday evening?

Here’s my sister’s theory. If a dog is nauseous, even a little, food will have a lower value, won’t it? And if this behavior of hitting the mat is mentally difficult for Java, then she will need a really good incentive to think hard. So it’s possible that on Friday evening food had a higher value because Java felt 100% OK (she was also fast, which is another indicator of both food value and how she felt). The next morning, however, food had lower value. Remember, Java’s nausea always came on in the morning. Java was slower. She didn’t try as hard. But toys were still high value and it was worth it to try hard to hit.

I don’t know if the above theory is true. It could be. Or it could just be that random breakdown of behavior that happens so often with RC. Three weeks without training can change the behavior.

Just in case I did stop training RC in the mornings for a week and got a better % of hits. Then we started some hill work*. First try was a leap, the rest were hits, and in the end she got three beautiful jackpots. 😀 Of course I got all excited about this, but the next session was on the flat with a different setup and it didn’t look half as good. I really thought that once the dog got the idea of hitting the mat and learned how to adjust stride to hit, the performance wouldn’t randomly change like it does with running the plank. Boy, was I wrong. I must be crazy to be training this. Well, at least it gives me something to complain write about, right? 😉

* hill work: putting the mat on a gentle slope (currently much less than a dog walk). I put it in the middle of a slope, not at the bottom of the hill (my slope doesn’t have a well defined “bottom” anyway). There is no reason to leap over anything except when she wants to avoid the mat. The plan is to work up to a steeper slope (probably a bit steeper than a dog walk) and then move to a hill that will have a similar contact to ground as the dog walk so that Java will have to learn to run all the way down to hit the mat.

Then There Were Turns

Here’s where we are right now with running contacts, not training as often as we should, but still making progress. We started doing some 90 degree turns. I decided to put props at the end of the mat to make it very clear what is good vs bad execution. She can still do a “bad” turn by jumping over them, so I’m hoping that will work in my favor and they will be easier to fade. Plus, they stack 🙂

Aaaaand we finally took the show outside which means we have more room to practice from different starting spots! There was some confusion with 2o2o in the beginning, but turns are helping with that 😉 Then for a while she was quite prone to jumping over the mat, probably from the added excitement of training outside. Now she seems to be back to hitting almost 100% of the time so I can be picky about which hits I like best. That would be:

1. two front feet
2. a combination of front and rear feet
3. just one front foot, but in the middle of the target

I don’t like to see just rear feet or one front foot on the edge of the target.

She is running toward a food bowl for now, but at some point I will have to replace it with static toy I think.

We’ve Got Stride Adjustment :)

I hope you like running contacts because you might be reading a lot about them in near future. I *tried* coming up with a different topic for this blog post, I really did, but the thoughts about RC just wouldn’t go away.

We hit a very important milestone last week: Java is adjusting her stride to hit the target!!!

I suspected it before, but I couldn’t be sure because there isn’t enough space in my living room to vary the distance much. But last week I was able to train her at a bigger distance and she was hitting the same way no matter where she started. Not only there were no misses, but majority of the hits were very precise: first front paw hit the edge of the target, followed by the other three paws neatly spread out across a 30x80cm target. I’m over the moon that she adjusted her stride to give me the best possible hit. 🙂

We’re training on three different targets:

• Non slip bath mat, white, 30x75cm, very thin – gives me the best hits
• Non slip mat for washing machine, black, 30x60cm, 1cm thick – gives good hits, but I think it might look confusing on my carpet. It’s also harder to see black feet on a black mat…
• Dog pillow, blue, 50x70cm, 5cm thick – this one is very tempting to jump over, which is great for teaching Java a difference between good and bad tries

One of Susan Garret’s Puppy Peaks videos shows Swagger running across a low wobble board as preparation for his running contacts and now I’m thinking that it would make sense to make such a board for Java as well. I’m guessing that a dog who can confidently stride through a wobble board won’t have so many problems if the training plank will wobble a little bit. I already have a wobble board in my “dog gym” (my living room 🙂 ), but its fulcrum is way too high to run through. I need something lower.

Along with working different targets it’s also time to pay more attention to turns. I like the idea of working on these from the ground up, so that the dog experiences the differences in contact hits when going straight vs when turning. We have only done a few sessions on turns so far, so it’s high time to get more serious about it.

I have to admit I’m a little unsure about what I like vs what I dislike which is why I have been avoiding them. What do I do when front feet hit the target, but one of the rear feet hits next to the target during a turn? Do I reward that? On a dog walk that rear foot would hit thin air, so I would be rewarding an impossible turn execution.

We’re training turns using a cone/pole to wrap around, so we don’t have problems with missing the target. I did try one session without a pole and Java started just skimming the target, but we soon came to an understanding that she needs to actually hit it. Still, those hits were quite high on the target and I would prefer her to cover the target well and then turn, so I went back to using a pole. Turning high on the contact might be OK on a full dog walk (haven’t decided yet), but while the target is still on the ground I see no reason to let her do it that way.

Then there is also a lurking struggle with stopping vs running. Sometimes Java decides that the training setup looks like 2o2o and offers stopping on the target, especially when the reward is in front of her, because that’s how we proofed 2o2o and she was very good at it. I can usually break her out of it by changing the setup, but it would be nice if we got to a point where I could use a static reward without confusion.

So obviously plenty of things to work on, but for now my heart is singing because Java is adjusting her stride to hit 🙂 Surely everything will be easy now!

Yeah, right.

Missing That Running Contact Feeling

When I decided to stop training running contacts, a friend (who obviously knows me better than I know myself) laughed and said I won’t be able to stay away from them for long. She was right. I love everything about running contacts… the fluidity for performance, the rush of running alongside the dog, the thrill when you see your dog run a full height dog walk for the very first time after months of training. Everything except a whippet falling off the dog walk because they were too crazy and misstepped and couldn’t hang on. That part is no fun at all.

[Just to clarify, I’m only training RC to Java, Ruby will continue to perform 2o2o as he goes crazy when I include dog walk in a sequence and I need to babysit how he runs up so he doesn’t hurt himself.]

A search for a slower performance

Many moons ago

So I was thinking… what if Java wouldn’t run full speed? I’m not talking about trotting across the dog walk, but a bit less than crazy full speed running would be nice. And especially, running in a bit calmer state of mind. Could I train a slightly slower and safer dog walk performance? I heard the warning many times that if you don’t train with maximum speed then you might never get a maximum performance of that obstacle, and that would actually work in my favor here. But can I teach the dog to consistently hit the contact while NOT running full speed?

I first learned about running contacts from Silvia Trkman and her method requires that the dog runs full out. The story goes that if the dog isn’t running full-out in training, they certainly WILL in trials and then they will leap over the contact because they will either:

• Not care about hitting the contact when they’re in such a hurry to get to the next obstacle, or
• Won’t know HOW to hit it at that speed

Daisy Peel adapted this method so that it can be trained at lower speeds as well and it seems that she is able to train dogs to a decent level of performance without the necessity of running full tilt from the start. So that myth is busted. She trains using a Manners Minder, which for many dogs automatically results in less speed. So this method would be a natural choice for me, right?

Well, not so fast

I would also like to change another part of the method: run the dog over a plank at ever increasing heights and click when she is running nicely (without leaping) AND hitting the contact. That’s simple for the trainer (once they learn to see the footfalls), but if I had you run for 10m and click for quality of one of your footfalls, how long would it take you to realize exactly which footfall I’m clicking for and which quality earned you that click?

Take into consideration that my timing would be way off because a series of events would have to happen:

1. I would first have to see the footfall,
2. then I would have to decide whether it fits the criteria,
3. then the signal to move the muscles would have to travel down to my thumb
4. which would then press the button on the clicker.

By the time you would hear the click you would be at least one stride further down the path. The only way to have good timing is to start *predicting* whether the dog is going to hit and clicking in advance. Otherwise you’re clicking on her first step OFF the dog walk – which is just about the worst timing possible, as you don’t want the dog to focus on how to get to the ground as soon as possible.

Some dogs have nice even strides that make this prediction easier. Other dogs soon realize that you’re always late with your clicks and figure out that the click wasn’t about reaching the grass at all, but about what they did on the plank. Java doesn’t seem to belong to either of these categories. I think she believed that my clicks and verbal markers were for running on the ground, not on the plank, so it “worked” for a while, until she found a shortcut – just jump straight to the ground 😛

Well, Ruby also didn’t belong to either of those groups of dogs, I just persisted long enough that he somehow figured it out. But there has to be a better way, right?

There has to be a better way to teach that it’s all about the last 60 or 45cm of a 400cm long plank

Devorah Sperber of Art And Dog Blog got me thinking about building a Clicker Board, which is a great idea since it’s cheap and ensures perfect timing of the click. I even bought a board and a bunch of clickers to test it 🙂 Unfortunately I don’t have my own dog walk to incorporate the clicker board onto, so I would have to use it on top of the club dog walk and I felt it was too bulky for whippet sensitivities.

For a while I was tinkering with building an electronic contact board which would be much slimmer. (I know I could buy a commercial version, but they seemed a wee bit expensive to me, especially when including shipment to Slovenia.) I got it partly working, but not well enough for training. Maybe some day when I get around to it 🙂

Enter foot targets

There used to be a widespread belief that foot targets were an ineffective way of training RC, but these days I see more dogs with beautiful running contacts that were trained that way. Another myth busted. Of course a foot target needs to be faded in the end. Still, if it helps to bring clarity I’m all for it.

A year ago I tried Dawn Weaver’s method because her dogs seem to have a very controlled style of running the dog walk, no craziness (yay!). I am not at liberty to discuss the details of her method, but will say that her dogs have both stop and a run and her method reflects this. Unfortunately Java became quite conflicted over when she should stop and when she should run. Then she got injured in April and I had a lot of time to think.

I was crushed. I thought what’s the use in training a really really difficult behavior when the dog could get injured at any time (not necessarily during agility, but even out on a walk like Java). Even the big names of agility world take two years or longer to train all possible entries and exits to a running dog walk in all sorts of situations. Running contacts might *look* trained once they start using them in competition, but there is a lot to be done still. Wouldn’t it be wiser to just go with 2o2o like I first planned so we could finally start running agility courses?

But 7 months (April-November) was a long time to wait, and I was still curious about running contacts. I started looking at other methods that use foot targets at least in part – whatever I could find out about Susan Garrett’s, Martina Klimesova’s and Laura Chudleigh’s method. Of course, after learning about different approaches, I couldn’t just sit on it and not try it out. Watching my friends training running contacts with their young dogs might have had something to do with it as well 😛

This is probably crazy, but three weeks ago Java and I started playing with foot targets. At this point she is basically going through retraining, which is generally a very bad idea. The best way is to train ONE type of performance and stick with it. Java has done some Silvia Trkman’s method, some Dawn Weaver’s method and some stopped contacts – enough to confuse anyone 🙂

Wish us luck, we’ll need it 😀

No More Running Contacts

I have been thinking for quite some time that I should switch Ruby’s dog walk performance to stopped contacts. He gets all crazy and wild-eyed when running over the dog walk and it’s not safe to be in that state of mind at that elevation and that speed. I love running contacts so this was quite a hard decision to make and I was postponing it… Until last week when Ruby fell off the dog walk.

Sometimes You’re Up, Sometimes You’re Down

When running contacts still included… you know, running

Java’s running contact training has been going really well when we were able to train a few times per week. We even had a 10-day long pause one time and when we started again the dog walk was a bit higher than before and Java was still running like a champ. But we were only able to squeeze in a few trainings before life interfered again and this time we were unable to practice for three weeks straight. And then it all went downhill.

The Drama Of Running Contacts

For some of us there is something addictive about watching a dog run full speed over a dog walk and hit the contact with ease like they were born knowing how to do it. It looks so effortless and so “right”.

Training it, however, can be a different story. It often resembles living a soap opera: Julia gets engaged to be married, Julia gets hit by a car, Julia is in a coma, Julia wakes up, but then her heart stops, Julia makes miraculous recovery, Julia gets out of the hospital, finds out that Ramone has been cheating on her and tries to commit suicide…

Full of DRAMA! If you always wanted to know how manic depression feels try training running contacts, it will get you pretty close 😉

Ruby’s Running Contacts – 2 Years Later

I know I promised to tell you what Java learned in Agility Foundations class, but I’m so excited, this just couldn’t wait!

Just Another Training Day

Ruby’s back was fine after 5 minutes of agility on Thursday, so I decided to do a little bit on Friday as well, but this time something went wrong. He looked uncomfortable walking up the stairs when we came home. Luckily after a little massage he looked OK again, but I don’t know what was bothering him. His toe? Biceps? Back? Time to call the physio to make sure everything is allright…

We ran a sequence from homework 4 of Agility Foundations since it was already set up, this time with low bars. It’s amazing how well he remembers some things that we haven’t trained in two years, like the cue for the back side of the jump. I am able to cue it and pretty much continue running while he does his job. I love this dog!

Awesomeness

Had an extra short session of running contacts with Java. 5 perfect hits in 5 trials.

Awesomeness.

 

Training Running Contacts

I train running contacts using Silvia Trkman’s method which is in a nutshell “make the dog run full speed over a plank on the ground, gradually angle it more and more and mark when dog hits the contact”.

When training RC with Ruby there was no special class for it. Every two weeks we would go to LoLaBu Land for our Agility Foundations lesson and I would pester Silvia with questions about RC. We trained them at home and it seems like I made every possible mistake.

I thought that for sure before we can start focusing on hitting the contact the dog should know that he needs to run over the plank. At first I tried to keep him on by clicking for walking on it, then I put thin poles at the end of the plank to make it a bit more obvious. My plank was of the same width as dog walk (30cm) and the sand on it (to prevent skidding) was actually small pebbles. Ruby did not like the feeling of those pebbles under his feet so keeping him on the plank was quite a project. This is one of our early sessions:

As you can see he was quite bouncy and I spent all my energy on keeping him on the plank instead of getting him to run fast. I was even clicking his bounces because I didn’t see them from my vantage point… I only saw them on video. So for a whole month I was teaching him how to leap instead of how to run.

Compare this to Java’s training. This is her second session:

She is running on a carpet that is 1m wide – that is 3x the width of Ruby’s plank. I was using a carpet because I wanted to minimize the risk that she would be leaping to get off the plank. As you can see she prefers to run on grass, but we handle this in a way that still allows her to run fast and it’s not a big problem because running on a carpet is not that different from running on the grass.

We did a few sessions on the carpet and Java was not leaping at all. Ever. Just happy, fast runs. So now came the time when we should transfer this to a plank, but I wanted to have a GREAT plank for learning RC. A plank that would be 50cm wide so she would feel she has enough space that she won’t fall off, 4m long so that she can fit in two strides and as thin as possible so that she won’t hurt her toes if she steps on the edge. Getting such a plank was quite a project so our RC training was put on hold.

When my SuperPlank was ready for a test Java ran beautifully. She tried running beside the plank a few times, but then we agreed on running on the plank and again there was no leaping, just fast running.
We did about five sessions on this setup, not because she would have problems, but to get many successful repetitions before I make things more difficult for her.

Today I raised the SuperPlank for the first time and Java didn’t even seem to notice! Sure it was a small raise (about 7cm), but I just can’t believe how EASY it is to train running contacts with the right equipment and with the right training mechanics (though my ball throwing skills still have a lot to be desired)! In fact it’s so easy that I was wondering whether I should blog about training RC at all. I just expect it to work and it does. Magic.

Silvia has a wonderful DVD that details all aspects of her method including turns after dog walk / A frame that you can get here. You can also enroll in her online class on running contacts at LoLaBuLand site.