Friday, February 25, 2011

Split legs and the running dogwalk

Lots of people training a running dogwalk are under the impression that if the dog's back legs are split apart that they can't jump and they use this for criteria.  Take a look at this video at around 27-28 seconds.

I can't speak for all dogs but my dog can take a nice big leap with his rear legs apart.  And good luck spotting that while you're training.  You can check for it on video after the fact to decide if you've been rewarding properly but it's not practical criteria for the training field.  And as Strummer proves, this criteria doesn't work for all dogs.  As always it's important to look at your dog and adjust your training accordingly.  I gave up on that split leg theory ages ago because I could see it didn't apply to Strummer.


  1. Yeah, the split criteria doesn't hold for Ruby, either. In fact, for a long long time all his jumps were done with nice hind feet separation (and there were many more jumps than running reps). So Silvia told me to look for even striding, both in length and height of the stride. Good luck with seeing THAT while running :)

    BTW, my opinion is that by that definition Strummer's performance on 0:27 wasn't a jump, just a missed contact ;)

    Fortunately, Ruby never jumps if he lands on or near the contact zone, so I'm cheating and only checking where his feet land.

  2. Well I'm glad I'm not the only one and that I'm not crazy. I like to see even striding but Strummer very rarely has it and he can hit the contact without even striding. I've seen video of Daisy's Solar doing it too. On the one hand it's not a bad criteria to consider but on the other I fear it might be too strict for Strummer to physically manage and also as you say, good luck seeing that while running. I can think about listening for footfalls though.

    Interesting notion that that jump on the video at 0:27 wasn't technically a jump. I guess it boils down to how we define a jump and if you use the split leg/rotary gait criteria to define what is a jump then you're right, it wasn't a jump. So the problem is more that Strummer can miss the contact without jumping. Or the way we're defining a jump is too strict because I can still have a jump and a missed contact. Either way, the criteria clearly doesn't work.

    Usually what I look for is the front feet hitting the down ramp twice and no 'jump' in the sense of a lot of vertical motion when he leaves the dogwalk. I check where his feet land too.

  3. As far as I understand Silvia's RC plan has two stages:

    1) Hitting the contact isn't important, only running is. That is, all strides should have equal length and height, and should be real running, not "bouncy" running that is closer to trot. One stays with this criteria until plank forms quite some angle with the ground so the dog can see where the plank ends and ground starts.

    2) In second stage the dog is required to take the last stride from lower and lower position on the plank and is jackpotted for extra low hits, which in time brings him into the contact zone and eventually into the lower 2/3 of the contact.
    When training reaches second stage the dog must figure out by himself how to adjust striding to get the reward, so I think it's only normal that strides aren't equal anymore. But just as you wrote, last stride shouldn't be higher than the rest.

    Of course, these are not Silvia's words, just my understanding :)

  4. Thanks for the explanations. I'll keep an eye for these things on my videos. I think we did spend plenty of time at the early stage with the plank working on getting a good 'rhythm' with his running but getting it to translate to the full dogwalk has been a challenge. Even now I think he's doing well with the dogwalk in isolation but in the middle of a course it's another story. I think the big challenge for now is getting him to focus ahead and not look to me. When I look at his misses in competition and in practice course work it's largely due to him looking at me. I don't think I did a good job of fading the treat gizmo and transferring the focus to the obstacles/course so that's what we'll work on next for a bit.

    Thanks so much for all the info. It's great to hear from someone who trains with Silvia! I wish I could somehow return the favor of being helpful but for now I'm still learning myself.

  5. I'm seeing a combination of striding. In the first slow motion he has a jump just before the downplank. On others his feet are together and then they separate on different locations on the plank. He is not doing a consistent running sride from beginning to end on the entire plank - his stride changes.

  6. Well sure there's lots of stuff going on in that video but the gist of it is that the rotary gait/split leg criteria is not a guarantee of a contact hit and not useful criteria to look for, at least for Strummer and at least one other dog and who knows how many others. So many people seem to be latching on to that split leg thing and it may end up not being useful criteria.

  7. Well I'm not an expert and still learning but my understanding is that split leg criteria was indicative of a running stride across the entire plank not just at the end. So if a dog does not have a consistent running stride, even if they change gait at the end it still may end up with a jump. Did you see the video someone posted of their dog running a plank in what appeared to be split back legs at the end of the contact and DP said the dog was jumping not running and that it's legs weren't split far enough apart. It was on the RC group list and generated a discussion. Have you thought of posting Strummer's video on that list for feedback. It would be interesting to hear what the experts say about it.

  8. If you look at only the rep that starts at around :22 you can see a nice big split in his legs across the entire plank and yet he can still stride right over the yellow. I have lots of other more subtle examples but this was a pretty obvious one. I should have edited the video down to that one rep. since that's the only one I meant to reference to show my point.

    I did post this video and this issue to the RC group and no one responded. I didn't ask a question about it though so I wasn't expecting an answer, was just pointing out an observation and people could do with it what they wanted. If you have a question about it for them you're welcome to reference my video in a post.

    I didn't see any recent discussion where Daisy Peel (I assume that's who you meant my DP?) provided an answer so I'm not sure which video you're referencing. I do question how many 'experts' are responding to things on that list though. There are some good handlers but most of them don't seem to be very far along the learning curve.

  9. Hmmm...
    So what is your new criteria?

  10. My criteria is the same as it ever was, I realized that the rotary gallop/split leg thing was leading me down a path long ago so I didn't adopt it as part of my criteria or I should say I like to see a rotary gallop but it's not enough criteria on it's own.

    I remember now that there's a running contact group on Facebook. Try this link, if it doesn't work you'll have to search for the group on Facebook, join, go to the Discussions tab then the Running Contact Methodology discussion. Daisy Peel has a bunch of stuff there about the rotary gallop. In fact she says she doesn't specifically select for the rotary gallop but rather she selects for 'don't leave the board with your front feet in a way that looks like a jump'. This was from years ago so it's possible she's changed her ideas, I don't know, but it's something to consider.

  11. Back when Daisy came up with these ideas they were considered interesting ideas to think about and now they've turned into gospel and people are all 'oh woe is me, how will I train my Corgi if I can't see the split in the back legs'. I'll bet if you asked the people on that Yahoo group to explain why a dog supposedly can't jump while in a rotary gallop/rear legs split that most couldn't do it, they simply accept it without understanding the theory behind it because someone on the internet told them so. Daisy first presented these ideas as observations/theories and even admits that she's not sure they're keys to success in using Silvia's method (at least that's what she said back then). They're interesting ideas and it took some creativity and innovation to come up with them but they're unproven in large samples. They're certainly useful ideas to some people, probably even most people but there are going to be outliers (Strummer and Andreja's whippet) and dogs for whom these ideas aren't terribly useful (Corgis) and I'm merely pointing that out.

  12. I'm the one who posted to the RC yahoo list about my corgi. My question was specifically in reference to the Trkman method as there is SO much emphasis on split back legs by both Silvia and the group. I am fully aware the dogs who don't offer a lot of split legs can be taught running contacts I was just wanting to know if dogs have been taught using her method instead of others out there.

    Even with my toller there are times where he does have split legs but easily strides over the yellow. your dog is just a much clearer example!

    Anyways I have always enjoyed reading your blog and this post especially!

  13. Yeah, that's my point exactly, everyone's so obsessed with this split leg thing and it's not 100% accurate or useful. Silvia didn't always have that stuff about the split legs on her website, I'm guessing she picked it up from Daisy and incorporated it in later. So there were plenty of dogs trained with her method before she adopted these ideas (I think it was sometime in 2008 that these ideas started floating around). And again, I can't speak for Daisy, but the way she initially presented the info. was more as an observation, something interesting you can look for but not necessarily the be all and end all of criteria.

    Absolutely you can train your Corgi using Silvia's method, you just have to have a different way of deciding your criteria for a jump. And you have the advantage of having worked with your other dog and probably having obsessed over boatloads of video so I'll bet in the back of your mind you have a good idea of what you're looking for, you just have to do some reps, shoot some video, trust your gut and see what you come up with. The basic concepts are the same for both dogs, it's the details and execution that will be different.

  14. And Laura, I wasn't meaning to pick on you and your Corgi in particular. I had no specific discussion in mind when I posted my comment, in fact I've heard that lament numerous times on other lists over the years. I think the whole beauty of the method is that most any dog can learn it but trainers have to adapt the details of it to fit the particular dog.

  15. You're welcome Elayne :) I can only train contacts a few times per week so I get my daily fix by reading about what others are doing ;)

    Of course, both jumping and missing contacts while running is a problem, but they are approached in a different way.

    Let's say that the dog can perform running contacts on home dogwalk very reliably, but is less reliable on unfamilliar dogwalks. Silvia seems to believe that if the dog is merely missing the contact on a new dogwalk, the problem will go away by itself as he gains more confidence. But if the dog is jumping, he doesn't understand his job well enough and this problem needs to be solved through training.

    Of course, while the dog is still learning running contacts on home equipment, every miss is a miss, but I would much rather have a running miss then a jumping one :)

  16. I did feel a little singled out when you mentioned corgi discussion but no worries! I've enjoyed reading your blog for awhile and of course with anything on the internet you never know where the person is coming from.

    Thank you for reminding me that her method originally did not focus on the split legs. It really is all that is emphasized now but clearly cannot be ones sole criteria.

  17. Oh my memory is not that good, I don't remember particular internet discussions. There's a good chance I never even read it in the first place since I only sort of skimmed through some posts when I joined. I went through some of the archives and parts of the rotary gallop discussion but just sort of skimmed through. I don't get email from the group, I have to remember to go to the web to see the posts and I try not to spend a whole lot of time on those groups for various reasons.

  18. @Andreja-that dogwalk is the dogwalk Strum learned on, I never had one at home, just a plank on a table. He's been on that dogwalk for a couple of years now. I do think, despite all the training, that he's still not sure of how to control his body at the speed he travels at.

  19. The first time I heard the split legs description was at a DP seminar last year. I think you're right, I don't think ST described it much in that way in the beginning. I think her only concept was to get the dog running across a board at top speed. If they are running - they are not jumping (her words).

    This whole discussion about strides and such (on the yahoo group not your blog) makes me want to train Soleil 20/20 on the DW.

  20. Oh Soleil is doing great so far. Don't let a yahoo group discourage you. Or maybe try training both and see which one you like. Then you would at least get the benefit of the running A-frame without having to train a different method for that.

    I have a hunch that ST incorporated the split leg thing into her website because it's a very easy, discrete way to explain to somebody what she means by 'running' and 'jumping'. And it is a great way to help people understand as long as it applies. But it's important to look at the notion with a critical eye and not get too caught up in it if it doesn't work for a particular dog.

    Some people like to over complicate things. All that other stuff is interesting and if you're a geek like me you like to think about it but I don't let it complicate my training. Dog runs through the contact and doesn't jump, dog gets reward. Easy peasy. Unless you're a short middle aged lady with a bum knee chasing a crazy assed rocket collie but that's a whole 'nuther issue.