Skip to content
AI sportā // Edgars Bernāns, Tim Cusick episode artwork

Dec 8, 2024 2 hr 6 min

AI sportā // Edgars Bernāns, Tim Cusick

from Train2Win Endurance Coaching Podcast Hosted by Train2Win Endurance Coaching Podcast

Labdien Pasaule! Šodien baigā dziļā tēma un par cik mēs ar Kasparu neesam paši labākie eksperti šajā nišā, mēs pieaicinājām citus, zinošākus cilvēkus. Tāpēc mums šodien ciemā ir RSU vadošais pētnieks, slēpotājs, podkāsta autors Edgras Bernāns. Tā pat šī šķiet ir pirmā epizdoe kur mums paveicās nozagt mazliet viņa dārgo laiku un saņemt komentāru par šo tēmu no industrijas “smagsvara” Tim Cusick, kurš ir ne tikai pasaules kalibra treneris, bet arī WKO vectēvs. Daži pieturas punkti tiem no jums, kas viņa vārdu nav pazīstami. Data Analytics Accomplishment Highlights TrainingPeaks WKO product leader Codeveloper of WKO5 Instructor: Advanced Training with Data Coaching Accomplishment Highlights 3 world championships 18 national championships 2 world records/bests Numerous race victories in multiple disciplines Tā kā visu mēs nepaguvām izrunāt visu, šī būs arī pirmā epizode kurai es iekļauju visus savus pierakstus, lai TU redzētu ko mēs nēpārrunājām un cik daudz nianses ir tajā par to, ko tomēr paguvām izrunāt. AI trūkumi: Ja sportistam ir liels apjoms (daudz stundu) Sarežģīta plānošan - komandējumi/nometnes Nevar iepazīt sportistu. Vismaz ne pagaidām Tas vienmēr būs metric based - junk in/junk out. Piemēram sūdīgs PM vai HRV/HR no rokas nevis no jostas Izdara secinājumu drīzāk “kvantitatīvi” nevis “kvalitatīvi”. Lai arī teorijā to varētu izdarīt Nav anamnēzes. Traumu vēsture + prevencija, iedzimtības utt Ārpus cipariem zūd pielietojamība. Sāp pleci, nevar tehniski nobraukt sekcijas, taktikas nepilnības. Asinsanalīzes un vispār vispārēja/holistiska pieeja Nav multidisciplināru risinājumi, piemēram slēpošana, spēka, cores utt treniņi (nevar pēkšņi gribēt uzkāpt Everestā, tik līdz Parīzei vai Pekinai) Accountability! Neticu ka kāds sekos AI plānam tā kā trenera norādījumiem 🙂 (piemērs ar maniem treniņplāniem) Parasti nav ilgtermiņa risinājumi Trūkst individualizācijas - 20 gadīgam sportistam būs līdzīgi principā kā 60 gadīgam. Well, as far as I know Nav FAN Factor AI priekšrocības: Ievērojami zemākas izmaksas salīdzinājumā ar treneri Vienmēr laicīgi saplānot Precīzi cipari/apraksts/uzstādījumi Uzreiz iespēja redzēt teiksim mēnesi uz priekšu (ko treneris parasti nedara) Citi apspriežamie jautājumi: No “rodas” AI. Tam ir jāmieāca/jāiedod principi, kas būs nevis kaut kāda meta analīzes, bet ticāk (šobrīd) kāda trenera pārliecība/metodes Nutrition un recovery integrācija. Teorijā ir iespējams, bet šobrīd nav tāda risinājuma. Grūtie jautājumi (hard questions) par progresu vai tā neesamību par rezultātiem, stresa menedžmentu utt.. Kā zinātne nonāk līdz plaši pielietotai metodoloģijai? Laiks + bias (tas par tiem kāda cilvēka principiem). Pēdējais iespējams nav slikti, bet pirmajā gadījmā tas nekad nebūs cutting edge, ko parasti pielieto praksē. Citējot Kaupi - čujs, ņuhs un poņa ir laba trenera iezīmes. AI būs tikai poņa. Kuram ir atbildība? Ja kāds noliek ķelli, man ir apdrošināšana un kaut kādi papīri? Ko darīt ar AI platformām? Līdzīgi kā lidmašīnās ar pilotiem un Teslās ar autopilotu :) Kāpēc Pogijs vai Ingebrigtsen netrenējas ar AI? Nu retorisks jautājums. Ar kādiem datu kanāliem AI operē? GPS / PWR/ CAD / Run PWR / HR / HRV / Cal (in/out) / Run stride/gate Kļūdas datos? PM +/-1.5%, kas teorijā var būt 3% GPS pie lielākiem ātrumiem ir neprecīzs Head unit ierakstīšanas frekvence. Garmin devices typically record at 1 Hz (once per second) for most metrics like speed, power, cadence, and heart rate Leomo’s motion sensors (IMUs) sample at 100 Hz (100 times per second) for metrics like movement, posture, and pedaling dynamics. Tipiskais fit fails 2-3mb kur tāds pats 300-400mb HRV day-to-day. Longitudinal data prasti būs ok Miegs. Jo principā viss ir aprēķins nevis mērījums Ziemas FTP/zonu auto korekcija. Kā to ziemā izdarīt bez testa? Kam DER AI risinājumi? Trenera lēmumu pieņemšanā Kādam kurš trenējas apmēram 3x nedēļā ar kādām 6h Bez specifiskām problēmām un ambīcijām (uz rezultātu vērstam sportistam tas nav risinājums) Kādam kuram vajag bāzes struktūru Kurš saprot pats pietiekami daudz lai pielāgotu un validētu risinājumus Kurš nodarbojas ar vienu sportu Kurš meklē low-cost solution

Episode details come from the podcast's RSS feed.

Embed this episode

Labdien Pasaule! Šodien baigā dziļā tēma un par cik mēs ar Kasparu neesam paši labākie eksperti šajā nišā, mēs pieaicinājām citus, zinošākus cilvēkus. Tāpēc mums šodien ciemā ir RSU vadošais pētnieks, slēpotājs, podkāsta autors Edgras Bernāns. Tā pat šī šķiet ir pirmā epizdoe kur mums paveicās nozagt mazliet viņa dārgo laiku un saņemt komentāru par šo tēmu no industrijas “smagsvara” Tim Cusick, kurš ir ne tikai pasaules kalibra treneris, bet arī WKO vectēvs. Tā kā visu mēs nepaguvām izrunāt visu, šī būs arī pirmā epizode kurai es iekļauju visus savus pierakstus, lai TU redzētu ko mēs nēpārrunājām un cik daudz nianses ir tajā par to, ko tomēr paguvām izrunāt.

Ready to play

AI sportā // Edgars Bernāns, Tim Cusick

0:00 2:06:33
of matches

Transcript Auto-generated, may contain errors

meal I think you are my favorite and you are not a fan of this I thought you were a talented fan in your dream I thought I feel you were the one too I thought you are a fan I thought you are a fan I thought you are a fan And I thought I was a fan I thought you were a fan I wanted to start with you I like to !! !! !! I'm not just talking about it, but I can't talk to people about it yet.

I'm not so happy that sometimes people can talk to people from the audience, much. But I also know that people still speak English, but I do not know that... But in a way, I like it. I don't know.

But this was our last time for my discussion. so it's not really a device and now the focus of people in which they have seen take control of them in thehip of other people and.. what are the advantages of that? I think that you can help the guy with the resistance and the value of that we are now collecting this value in the middle and so now the future is right now so I am a guy No idea.

Thank you. I'm just like summon the the context. No, it's to be in a conference on Charles to love me, Maria, we should get out to see a prospect and so I could, hey, that guy's a lot of business to us. She did it.

Yeah. Two. We've seen quite a bit of that. But but marks like I see, like, we see some architects, yeah, normally, no, to speak, but also out of those BMI, never sellings, no, you know, but no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, and then we often would ask whether if they are in the Garden of Aishmed, or not.

No? Yeah. They would not be in the Garden of B but even if they were in the Garden of B, no, the only part, the other part should have been I'm not sure if I'm going to say something like that. I'm not sure if I'm going to say something like that.

I'm not sure if I'm going to say something like that. I'm not sure if I'm going to say something like that. What ethical considerations arrive when athletes or coaches rely on AI for decision making? I'm not sure that's going to be an ethical question because it's going to become the norm within, I don't know, three to seven years, to the number five years.

It's going to become more the norm. But here's one thing that I think it's so important to understand because of the complex physiology of the human being and the even more complex response to exercise stimuli of that human being. Data helps us make better decisions as coaches. And what we need to do as better data analytic formats come to the world, AI, we need to learn how to master those analytic formats and use them to help us make better decision.

I don't think AI is going to have enough of the universe of the athletes data. And that sounds crazy. It's not going to have enough athlete data. Hey, we're tracking power in our workouts, we're tracking heart rate, we're tracking sleep, we're tracking food, we're tracking.

There's a lot of data points. We have a fair amount of data, but not enough for AI to intelligently get that. Maybe it will in the future. The more we add data, the more potential chance it has.

But the reality is it does have enough data to empower the coach to make really good decisions and make those decisions, crunch all that data in a much more time efficient way. I can remember the old days when we were training on paper and we fax workouts and emails back, or it was just when email was even coming on and you'd be sitting there just trying to read everything and analyze and try to figure out where the athlete is. My training working, is it failing, are they just too tired? AI is going to help answer all of those questions for the coach.

You the coach have to learn to embrace that AI and use it as a tool to help you make better decisions. In the absence of that, the AI might actually replace you. Yes, that's what I'm thinking. Yeah.

Yeah. And I think it's a good thing. I think it's a good thing. I think it's a good thing.

I think it's a good thing. It's not a good thing. Sometimes there's a DRM on the section on website, and it's a good thing as an employee, but you're trying to make these decisions too safe. It's a good thing.

I think that we never get the time. Yeah. Now the team actually Pennsylvania has few other Technologies. At roughly the Unknown bomb in the Nebuchadilly trial, And I think we have to do that before we start.

We have to do it again. We have to do it again. I'll be happy to see you in the next video! The next question is...

...in this case, ...we have to talk about the new topic. I'll be able to do a little bit about it, ...but the new topic is about the new topic. We'll see you next week, in 2019. We'll be here in the next video.

I know live in the last week, ...and in the last year, that you have a lot of money to buy and sell as you have in the program. And I have a lot of money to buy and sell as you have been in the U.S. I think that it's easy to buy and sell as you have before. And it's not easy to buy.

It's a long time ago. But it's not easy to buy and sell as you had to buy. it's a right to stop in the limit if you can't do it. He's going to do it a little bit.

But it's not to be a bad problem. And it's not a bad problem. And you're still going to get a little bit of a disability moment. But it's not a bad problem but it's a very bad problem.

And that's a good thing for us to say that, because we were already getting kind of blue and blue. We were just saying that we were going to get to see plan. Thank you very much. The biggest thing, it's a bad feeling.

But really, we can't really get the chance to do this because we know we can't have any money. I think we need to have more money, as well, to sell the money. Because we don't own the money in a company. That's why we do it even now.

But just as we know it's bad, we're not going to need to have the money in a company that we don't have money yet. We don't want to work here a lot, but I think we can't work here properly. in the audience. Yes.

Yes, yes. That's thanks. I think I'm just going to do something with this story, but I didn't want to do it again. Ok, I'm not going to call an agent any other story.

No problem. Yes. Yes. Thank you.

Okay. So, now, let's go ahead and get the idea of a strategy. I start the train. Can AI personalized training to this extent that it rivals or surpasses human coaching?

If so, how? I'm going to say no, but I'm going to caveat my no. And then I'll tell you why. I think AI coaching will replace bad coaching or even mediocre coaching in the next three years.

So if you're not out there really trying to become that elite coach, improving your skills, investing in your knowledge, not just in physiology, not just in modern training techniques, but in things like the coach athlete relationship and what differentiates you, you should feel threatened by AI coaching. Here's why. AI coaching is going to come in with superior process, clarity and modalities and how to implement these modalities based off the athlete's historical performance. And it's going to be, I want to say it this way, from a process sense, it's going to be foolproof.

Your workouts will always be posted. You will always be able to see the future. There won't be the things that we as coaches sometimes have human error. And finally, the reason it's going to be most competitive, it's going to be low cost.

You're going to see AI coaching at a very like significantly less 80%, 70%, 50%, less cost than human coaching. So if you're listening to this podcast and you're a coach and just took one piece of advice from me, why should you but still up your game, you really should be investing in A, your overall coaching knowledge and B, how to use the incoming AI tools to improve your coaching before they replace you. But doesn't really get better of the moderate to go to elite coach. No.

And I actually think it's a long time away from ever doing that. Here's why. Physiology is very complex, like human physiology is super complex. And we have a fair amount of science of how to diagnose that physiology or understand that physiology.

AI is going to do a pretty darn good job of that. But and here's the AI, but and this is what it'll be interesting to see how this goes. The response to exercise stimuli is very complex. And there's so many variables that impact that response to that exercise stimuli that just a robust understanding of physiology might not grab.

Now, AI is going to learn over time, it's going to keep getting smarter, but still the complexity of that response, because it's not just about, I did an hour and a half with some intervals in it. It really is about what was my day like, what level stress am I out, how well, you know, am I recovering? Now don't get me wrong, AI is going to be able to, with all the new data points we have, it'll know how you slept, it'll understand what's happening, you know, with your resting heart rate, it'll understand HRV, it'll have a lot of other supporting data. But still, I'm going to tell you, my personal experience, the complexity of managing the human performance, from training an athlete from performing here to up here and improving, it doesn't always follow all the rules.

So I think if you were to walk away and consider how, you know, this is going to change the game, investing yourself, investing your knowledge, investing those wisdom-based process, and invest in the coach-athlete relationship, and really being in tune, because that's where, you know, you'll separate away from that AI intelligence. No, I'm not. I'm not sure. I'm not sure.

I'm not sure. I'm not sure. I'm sure. I'm sure.

I'm sure that I'm not sure. I'm sure that I'm not sure. I'm sure that I'm not sure. I'm sure that I'm sure.

So if way, maybe you can find your target list on your, and you're below the value of it. And, thanks a lot for this talk, and for N23ide, because for aielded thing, the allow you come up with the host, giving him the best option of getting a Ace ofnumber 5. And he would do all of this with you. And he came in real quick.

He Plans for Yemeni Reb, and absolutelyized it in German Thursday, of each other, with a list of people in German. now, no, never Buddh are AI. For real, the world's Mars is not really the same. And I'm not the one with the great weakness of the other.

So, if you choose from the great world of American development, here, you're not the one who is interested in AI. I'm not the one who is interested in AI, but were there any kind of skill that you've seen? I'm in the same world, and I'm the one who is sort of an important person who is consistently involved with AI. but I will be very impressed with that.

When I am a guy, I will be a guy who was in the job manager of TIPs. The fact that TIPs are not a guy who wants to be involved with a company like TIPs. So, who are he? People who did get 0.8, 1.8.

And the bad guy is true. He is a guy who is a guy who is a guy who is a guy who is a guy who is a guy who is a guy. How was your problem with TIPs? How?

How did you get him? So I have two of them. Yes, he was a guy who has been a guy who is a guy who is a guy who is a guy who is a guy who is a guy who is a guy who is a guy who is a guy who is a guy who is a guy. Well, we didn't get it to our job.

a like the Yeah, I know that. Is that it? You know, they said, I don't know. But, you know, that's something that we can do.

I have a similar conversation. and AI technology do I foresee impacting sports training in the next five to 10 years? Oh, I think it's going to be a heavy impact, but we're going to see two different areas of that impact. One is we are going to see the introduction of the AI coach where the artificial intelligence self is the coach.

This system will be based on large data sets and kind of machine learning. That's not quite the right term, but we'll use it. And it will take that learning to develop an excellent process and it will implement process based off that data matched with the data you, the performing athlete is supplying it. And it's going to execute at a pretty good level.

So let's define good. The AI coach is going to be really good at process. It's always going to post your training. It's always going to make sure everything is the right time range.

It's always going to make sure the fundamentals are there. But it's going to lack some other things. I'll answer that in a future question. The other direction we're going to see AI where I think one of the more exciting part of coach is kind of fearful of AI.

I think we need to embrace it is you're going to see AI coaching tools coming into place soon. You're already seeing them, but you're going to see them be more robust and more focused towards supporting the coaching role. And here's why these new AI programs were starting to be able to crunch tremendous amounts of data. We can already do that.

And now we're in the world of indoor sports, we're connecting big pools of data to AI more and more. We're going to make it work for the coach that crunching that analytics. It really is going to do an excellent job of trending analysis of grabbing insights that the coach might not be able to see because it's just too much data to process. So when we have a I coach who support, it's going to read those big data pools and it's going to surface actionable intelligence.

It's going to say, Hey, coach, look at this. Look at this. And at least make the coach aware of what it's seeing, you know, data trends, it's seeing in these large data pools that would take the human coach, you know, hours and hours of just looking through and trying to figure out. So I think that's how we really see the split.

We're going to see the AI coach and the AI coaching tool become two separate things. If I was a coach in today's world, I would embrace that as a reality. And I would try to stay on the forefront of what AI coaching tools, what can I use to make me a better coach, a little smarter, a little more time effective and efficient and a little better, a little better at predicting, um, training based results. No allergies.

We've taken myself kind of nailed it. No, we didn't normal music. Oops, it's not good. No quads.

No quads. That's what the technical tip of the work is. To be honest, when you're a coach, you've got to be a coach, and you've got to be a coach. Sooping out really, I think those things are different effects.

See that, this is how great this team is egg. And this is what the CD- Gerard told us. Maybe depending on the new horizontal picture of the CD russ is beef Keex talk about the color of the work they did in the wines. And how often does it spread out like that?

Because you know, it's notians territoryy and something like that. It's going to be neighbors. We don't necessarily live in Kansas since Spain till actuallypiring a year. It's very specific to that.

Probably. In the similarly kind of old World War II this is a very focuses- Different place. Better place. Okay.

Go on. What's your favourite story about the World Cup? We're not going to make the top-end of the World Cup. What's your favourite story about the World Cup?

It's our favourite story. It's a bit funny, we've had a lot of time to work with players. But we've had a lot of different teams, but we're not going to make them make their own decisions. But we have a lot of different places to work.

We have a lot of different options, They are either or freedom or dislocutor They get funds If you have funds about them you will colours They want ADA That's why I was re- focused on having high income or negative card if you didn't know how to use cash meunciation Let's talk about public we let Pearse ask ... Yes. So let's see. Yeah, I agent can talk at that.

We need to get close to Muslims. That's going to be sweet. That's the way you are. You must be getting the flove, and then training picks, starting with the wounded, and then ballooning into the world.

And now it's a bit of a problem. Were you going to teach us to. I think it's a series of how to make a video. So I'm going to say that my home house from my home is the most interesting thing I've had, and why not?

That's crazy. I'm gonna say that it's in the middle of the business, the money, the city, the loot, the business, the policy, and the interesting thing for me is that you have to wait for a second and you still have to wait for it, it's not like you've ever seen or loved like this I like that the democracy, I like which she thinks it's American and they think we should get produced by President Trump and Raiders. I tell you a little about like says, gender, sex, sovereignty and gang violence. they don't know that they're being attacked by a police yeah, And was that great, all right, so I'm going to finish.

I'll play this one next stage, because we should have done this. I'm going to continue this and go on with the short part, and this is the key, and the first thing I want is to do on top of the deck, and make sure to do it. If I was going to do this, I can go on top of the deck, and then I'll go to the deck and start from one point. So that's where you can find the top of that deck.

And then I'm going to finish it here, so it's very important to start by the first player in the deck so you can get left, and then I can have it down, and you can see how much you can be left to go. And then I can start to get left, all the assumes. Oh, she's not that. She's not as good about to teach advanced.

This is the best way to make a video. As I saw earlier, I thought I was taking it out. And I was saying, I failed to make it. To be honest, this is a game.

And I would say it's just to start it out to be smart. Yes, yes, yes. And at that point, I would say the topic... At least what I had to say was that I was willing...

or short-term, was to be very close to the... I'm not kidding. like a bunch of these people. a big question.

Okay. Okay. Okay. Okay.

Okay. Okay. Yeah. Damoy.

Okay. Okay. Yeah. Yeah.

I mean, we're coming. My wife, we'll be the first to tell him, so I will be talking about this one, so we'll be talking about what you want from my staff, talking about what you really want from my staff. This is what I want from my Foreign Affairs, everyone that lives inain. This is very interesting, the research of this application.

It's very funny, I don't want it to happen to you. I would like you to talk to me about this. I have to go to the social forums and support. Because he is so much more powerful than he will be.

It is because I knew he had to be in the last few years. And I had to be in the first situation. I went from a... ...to a big man.

So, maybe we got more... ...but he is also my character. And you have to be in the last few years because he hasn't done that anymore. And he will have to be in my story...

...and make me happy. He will have to be in the last few years... ...and I will never be in the last few years, but he won't be in the last few years. He will not be in the proud world.

Exactly. Just a lot of this. It was the biggest thing to know. Ciao, exam.

Ciao, exam.

Frequently asked questions

How long is this episode of Train2Win Endurance Coaching Podcast?

This episode is 2 hours and 6 minutes long.

When was this Train2Win Endurance Coaching Podcast episode published?

This episode was published on December 8, 2024.

Is there a transcript available for this episode?

Yes, a full transcript is available for this episode. You can read the complete transcript on the episode page.

Can I download this Train2Win Endurance Coaching Podcast episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.