Finding The Optimal Training Zone
|Self researcher(s)||Ralph Pethica|
|Related tools||Strava, GeneTrainer|
|Related topics||Sports and fitness, Heart rate, Activity tracking, Personal microbiome, Mood and emotion|
Builds on project(s)
|Show and Tell Talk Infobox|
|Event name||2018 QS Global Conference|
|This content was automatically imported. See here how to improve it if any information is missing or out outdated.|
Finding The Optimal Training Zone is a Show & Tell talk by Ralph Pethica that has been imported from the Quantified Self Show & Tell library.The talk was given on 2018/09/22 and is about Sports and fitness, Heart rate, Activity tracking, Personal microbiome, and Mood and emotion.
Description[edit | edit source]
A description of this project as introduced by Quantified Self follows:
Ralph Pethica has been combining fitness tracking and subjective data with genetics, using techniques from his work with professional athletes to help find the optimal way for him to train.
Video and transcript[edit | edit source]
Finding the Optimal Training Zone So this is for people that haven’t seen me talk before. This is another installment in I guess what has been a few year’s quest into connecting genetics with sport data. And also kind of getting people to think beyond one silver bullet right, like the whole industry is around, ‘Do this and it’s going to change your life.’ But we’re moving towards something where we understand that we’re complicated, and you’ve got to take into account a lot of signals. So in doing this I ended up working with athletes, so I’ve been now in the professional space for about five years, building platforms for athletes to use. So we’ve got a few professional teams in Europe now that use our software. And I guess today is really about how can I apply this to myself, right. We’re not athletes but how can we use this to improve our fitness. And so one technique that I want to talk about today is this idea of finding the optimal zone. And so some people who have been doing a lot of stuff in sport before will probably know a little bit about this, but i need a few slides just to show you anyway. So yeah, there he is. So this is one of the guy’s I work with. That’s Semi Radradra, who’s about 220 pounds, six foot three and you definitely don’t want to get in the way of him when he’s got a rugby ball. And on his back, you can’t really see it on here, but there’s a little bump and that’s a catapult tracker, which is tracking basically his movements and accelerometer etc. And this is one of the ways that they quantify athletes when they’re on the field and of course behind that there are other ways you can do things. So I like to break these down into kind of five things, I don’t know if you can see this alright, but there are things that you quantify and there are things that you use to calibrate. The things that you use to quantify are things like subjective data, ‘How do you feel today? How difficult was that session?’ Calibration is more like, how much does the guy weigh, what’s his heart rate, what’s his you know, resting pulse, what’s his heart rate variability. And then there’s things like the gym and the muscles, and then this whole space of genetics. We know people are different, how do we use it. And when you’re increasing fitness, you do what we call progressive overload. So you can just start by wanting to run a marathon by running a marathon. You’ve got to do it in little phases and you’ve got to build up to it. But the big question is, how fast do you build up to it, how much time do you give in between each session in order to recover, what’s that kind of optimum ramp up there? And a baseline, something that we’ve created would look something a bit like this, right. These little spikes that you see are me running or doing a cycle or something with heart rate. And it’s just looking at the heart rate versus the time, right. But you can do this in other ways. And over time it figures out what I’m doing today, and what I did in the past in the last few weeks, what I’ve done today and then how the two change. And of course, it’s like what they always say about cycling, you know it doesn’t get easier, you just get faster. And so these models, the zones there basically adjust on you being able to do more and more. So the idea is to keep progressing right. With athletes, when you find yourself in the red zone, the orange and red zones at the top there, injury risk is really really high. But actually, if you find yourself in the low zone, the injury risk is high as well because you’re insufficiently prepared for the sport you’re trying to do. And of course, we can personalize them a little bit by adding things like genetics and age and anything else to the equation. So those models assume the same recovery time for everyone which is incorrect. So this is me with a subtle I’m slightly better than average at recovering from certain types of exercise, so I can take that into account. Now, it might seem like a tiny detail there, but if you’re looking at that first spike that might be the difference between injury or not, right on an athlete scale. So that’s the science. But this is about me. People that have seen me talk before know how I like surfing and I used to be a break dancer and I broke my neck in 2014. And since breaking my neck I’ve done about five or six Olympic triathlons and a bunch of other stuff. This is me training for the triathlon this year, between January and May. I normally do a bunch of training in there, so you can see the ration is going up and down. And you can also see something really weird is starting to happen in April, certainly before that. And of course, knowing what your zones are doing, going up and down, you can also monitor your fitness right. Your fitness measured by your ability to do low looks a bit like this. So people who’ve used Strava often see models like this, where you’ve got fitness going up over time. And so you can see there’s this dip in that in April, which for me is basically is just a lingering back problem. Like nothing, nothing complicated. So I started quantifying that as well, and I also went to the doctor and I got some pretty good meds, which apparently loads of athletes are on as well. And with that, that’s actually what’s allowed me to make this massive spike, right. And if you look into the zone you can that I basically went a bit mad on the right. So that’s basically me being like, yeah, I can do anything. So that kind of goes down over time, and you can see that the fitness goes up as the back pain goes down which is kind of unsurprising. So a few correlations and things that I’ve understood from this. If you do it subjectively or with heart rate, you can still get the same pattern. So this is me just saying how difficult each session was and multiplying it by the time, and you can overlay that with heart rate. Now what’s interesting in the professional sports world, is that athletes get better at this with age. So the 18 year old’s who come in have no idea how difficult a session is. But guys who are a bit older, and I’m now 37, so maybe I’m getting better at his, they can you know say, you know that actually was about x, or 10 hard or whatever, yeah. The other thing that I looked at as well, so I weigh myself and you get the fat percentage out and this is not really rocket science. But you can find this nice correlation between the fat percentage and the fitness which is pretty cool. In fact, my last fat percentage coincides with my trip to the USA, so it went back down when I went back to Europe. And another one that I thought was kind of interesting was resting pulse over time, right. So this is measured, I measured this particular one with HRV for training which some of you guys might use. So this is taken in the morning, and you can really see that over the years since about 2014/15, there’s been this net drop in resting pulse, which is a pretty good indicator of fitness. And you can also see it changing, especially from January to May, this spike in there. Whenever I trained for a triathlon it seems to go down a fair bit. And the goal for me is really is like is have a really high output in terms of fitness but very little work, right. So what you’re looking at here, these are charts of how far I ran, right. And this is basically to say how little I ran more than anything else, so I know you guys don’t think in kilometers, but we’re talking of a maximum of about 50 to 60 kilometers a month. So what I’m saying, like is strategically using this, getting the zones right allows you to do way less work and get the same fitness output, which is kind of a goal. So what I learnt from this is really that you, you know it’s working and you get fitter. The calibration definitely helps. It works for amateurs as long as you train regularly enough. I was saying to someone before that age is definitely a thing and I think you feel that in your 30’s and I feel that at any time, but I found that getting these zones right has become more important with age. Because little things like lingering injuries and such become more frequent like that. And very quickly there are too many things to correlate. I looked at this data before and I was like, there are all kinds of things in there. I think, you know one of the things we need to work on in general is to be able to automatically correlate things that are happening, right. So this is something else that you can do. And in the professional world we’ve created these ratios of absolutely everything that an athlete might want to measure, and that’s very revealing. So if you’re measuring like the number of shocks on a rugby player, or the number of kilometers, you get these ratios out and you very quickly see when something goes completely off. And when that goes off you go and look into it, there’s always something behind it. So knowing where your baseline is and knowing where you are today turns out to be pretty useful for lots and lots of different things. So next, I just want to talk about some of the things we’re doing on this. We implemented the algorithm to do this very very fast, so we can calculate a baseline for a lifetime of data in a couple of milliseconds. And that allows us to make a parallel version of all these things that you can run through whenever you want. And we’ve also on the genetics side and this is for another day, but we’ve added a bunch of new interesting genes the we isolated through another means and added them to the story and tested this on a couple of teams and athletes, and we found that that’s increase the reliability of our models as well.
As I said before, I think we need to get better at automatically correlating stuff. So, we need more things out there that say, ‘Hey, did you know that on Tuesday’s when you do this, or when you get this amount of sleep this goes up, this goes down’, right. So that’s maybe food for thought for the next phase.
About the presenter[edit | edit source]
Ralph Pethica gave this talk. The Show & Tell library lists the following links: