Shownotes
In this episode of the Athletes Compass Podcast, the team is joined by Andrea Zignoli to explore a unique research study conducted within the Athletica platform—analyzing over 55,000 athlete comments to detect emotional patterns through sentiment analysis. The findings reveal that outdoor workouts tend to trigger more joyful responses, certain body part mentions signal pain, and beginner athletes generally report less joy compared to advanced users. The discussion dives deep into how this emotional data could shape personalized AI coaching and highlights Athletica’s cautious yet promising approach to integrating sentiment with training load decisions.
Key Takeaways
- Sentiment analysis on Athletica reveals emotional insights from athlete comments.
- Outdoor workouts, especially running, are associated with more positive sentiments than indoor activities.
- Mentions of specific body parts (e.g., shoulder, calf) are often linked to negative emotions—usually signaling pain or injury.
- Beginner athletes express less joy in comments, possibly due to physical discomfort or lack of adaptation to training.
- Advanced language models now enable accurate emotional classification of athlete-written feedback.
- Athletica aims to use sentiment data as an additional layer in training load monitoring—alongside HRV and RPE.
- Coaches like Dan Lorang value athlete comments as a key insight into readiness and performance.
- Future Athletica features will include conversational AI and more nuanced integration of comment sentiment into adaptive training prescriptions.
Transcripts
it was interesting.
to find that, for example, outdoor activities, such as running in the outdoors, were associated with a more joyful and more happy comments from the user than cycling on an ergometer or rowing into the
Paul Laursen (:you
Paul Warloski (:Hello and welcome to the Athletes Compass podcast where we navigate training, fitness and health for everyday athletes. Today on the Athletes Compass podcast, we are discussing features in Athletica, including a fascinating study done on the language from the comment section. Plus we're taking questions from our chat bot and the forum and answering them here.
We'd like to welcome Andrea Zignoli to the podcast to discuss a study on sentiment analysis from Athletica's comment section. Andrea is part of the amazing back end side of the program. Andrea, could you explain more about the study and what it suggests about our word use?
Andrea (:Yes, thank you for having me first. Yes, we came up with this idea because we realized we had tons of comments on the platform from the athletes. And we know that comments can hide in their language and tone something about the emotional status of a person.
⁓ semantic analysis was born for studies in the marketing or in social network studies to understand what kind of impact can a new idea have on the market or on users of a platform or to sense the sentiment behind a tweet, for example, or a comment on
social networks. we know from theory, from the HIIT Science book, especially from the chapter, yes, monitoring the training load. So we are talking about the surveillance system that
a comment from an athlete or the logs into a journal can hide potentially very valuable information to better understand how the athlete is digesting the training load or the emotional status of the athlete at large. So, yes. Yes, sure.
Paul Laursen (:Hey, Andrea, can I just interrupt there, just even for
our listeners? Because I had to learn this a little bit myself, and I'm sure there's some listeners out there that will like me, where it's like, okay, well, what are we talking about here? What is sentiment? What is semantic analysis really? And we're talking about the level of joy and sadness ultimately in the comments that we make, isn't it? It's like, am I...
Am I talking, the language that I'm using, is it expressing joy and I'm happy or am I using sad words and I'm whatever? You know what I mean? It's kind of like that, right?
Andrea (:Yes, exactly. Yes,
yes. You can extract the sentiment from a sentence or a piece of text. I mean, you have language models that can process the text and classify the dominant emotion or tone or sentiment behind that sentence.
Paul Laursen (:Yes.
Andrea (:Of course, with written text, cannot extrapolate anything from the level of the voice or the kind of non-verbal communication that you can have when you're talking with someone. Recording the voice would be a step further. And showing the video can be another step further. But in terms of text, you can process the text and extract the dominant sentiment.
based on the choices of the word. And of course, you can quickly understand that a word can mean many different things, but it depends the context you're using that word in and what you're using that word for. language models have been built exactly for that specific...
goal and task to extract the words in a sentence and try to understand the dominant emotion. So there are models that can provide you with like a binary or like if a comment or a text is positive or negative or neutral, but there are more sophisticated models that can extract like emotions or sentiments like happiness, anger or
other few that might be important for you to assess the emotional status of a person. And of course we can debate like a lot whether this might be accurate or not. It likely depends on how serious it is the attempt from the athletes to report the logs on a journal and how many emphasis they can make and behind.
behind the writing of those words.
Paul Laursen (:Yeah.
And just to pause you there, if we had done this study three years ago, it probably wouldn't have worked, right? But now, after many years in existence, we have a whole lot of data on Athletica. Isn't that right?
Andrea (:Yeah,
a couple of perspectives. Both the language models were not as sophisticated as they are today, or well-developed, or usable and widespread. And of course, we have a lot of comments now, so we can build and test and see what is working and what is not.
So both perspectives are valid to justify why this is timely and why this could happen right now, this year and not previously, at least in the context of Athletica, I think.
Paul Warloski (:MJ, what do you? Yeah.
Marjaana (:And so what did you find?
Andrea (:⁓ that was interesting. First, sentiment analysis is a broad terminology to express other different analysis that you can carry out. For example, we wanted to check if...
⁓ the sentiment coming out from the sentences was associated with the behavior of the athletes. I wanted to find out if for some specific reasons going out for a bike ride on Tuesday was like, was not as entertaining as like going out for a run on Monday. We didn't find any specific pattern in terms of days of the week, but it was interesting.
to find that, for example, outdoor activities, such as running in the outdoors, were associated with a more joyful and more happy comments from the user than cycling on an ergometer or rowing into the gym. So we found out that
Paul Laursen (:you
Andrea (:I'm not saying that runners are happier persons than rowers are. I'm just saying that we found that there was a significant higher happiness score when an athlete is training in the outdoors.
And that was fun to see in this quarter. And for example, we found out that also running sessions were associated with happier comment than cycling sessions. So someone start joking about saying running is better than cycling for that specific reason. But that is a ⁓ minor finding. ⁓
Paul Warloski (:You
Paul Laursen (:Yeah.
Andrea (:Then we enjoyed so much trying to see if there was an association between the specific words and the sentiment behind the comment. So for example, not surprisingly, comments mentioning ⁓ people.
athletes we're training with, for example, comments mentioning wife or daughter or friends or team were associated with like higher happiness scores. And other words, for example, calf or shoulder or ⁓ back were associated with negative comments. And we think that
when you name like a part of your body in a comment, it's because that was painful during your activity. So Some things are only present in the comments when you notice them.
Paul Warloski (:Hmm.
Andrea (:Otherwise, you would never report that your shoulder was amazing during your swim workout. You only use the term shoulder to tell your coach that the shoulder was painful on that specific day. So that's why some terms, like some words, only use it in the comments when they have uh like a bad the same.
Paul Laursen (:context.
Andrea (:for sleeping. Sleeping is not 100 % clear, but most of the comments that were using sleep were using that word for expressing a negative sentiment. So people notice sleep when it was a bad sleep more than when it was a very good.
Paul Warloski (:sure.
Paul Laursen (:of
Yeah.
Andrea (:night
sleep. these are the things that we investigated with our analysis. And then this has been done at the population level. Let's say we put all the data together to find
these kind of patterns, but then we scaled down to the individual level and we tried to see if there was someone consistently reporting the comments on the platform and we tried to reverse engineer and to see with time how their happiness or sadness changed throughout the training season. And we found a couple of athletes which were like
especially good at reporting in the comments there, like their training journals, and we asked them to give us feedback because we detected in their oscillations in happiness and sadness, we asked them to report back and give us some feedback.
about, for example, the dips and the valleys in happiness. And we found out that a couple of them answered that in that specific period of their life, they had an injury or they struggled to maintain a training regime. there was something going on. This needs for their research. But we do think that semantic analysis can be a tool that we can add together to the ⁓
For example, the other tools that we currently use for monitoring the response to the training load, for example, HRV analysis or training load analysis and et cetera. So that was the purpose of the study was to try to see if we can add an additional tool in the toolbox to assess the training load. And in this case, it's an interesting one.
Paul Warloski (:Hmm.
Andrea (:A,
because it's a new thing that is coming up with the new technological solution, and B, because this is something that has been in the back of the mind of the coaches for very long. If you read those chapters in the HIIT Science book, like eight and nine chapters, there are a list of tools that you can use to evaluate how...
the athlete is coping with the training load or how the athlete is doing in their lives. And usually those are visual analogous scales, for example, that you need to fill out when you wake up or when you end your training session. But at least in my opinion, an athlete would rather talk or write down their thoughts rather than like filling out a scale. So in this case, this could be like a very good way in which we assess
the ⁓ how our athletes are coping with their training regime.
Paul Laursen (:And I think it goes back to, so I had Dan Lorang on the training science podcast and Dan Lorang, if you don't know, he's a legendary coach in the area with many successful athletes. And he said right out very clearly, that is the most important thing for him is reading those comments, reading those comments. He gets so much out of reading the comments and it's the same sort of thing here. That's probably why he...
Marjaana (:Mm-hmm.
Andrea (:Yes.
Paul Laursen (:probably has developed the art of reading those sentiments himself as an experienced coach. And I think that's what Andrea and his study here have really highlighted that Athletica can also do this as well as one more tool in the training load response. In other words, how you're responding to training, how you're responding to that stress.
individually because we know that we all respond individually to the different stressors and training that we have. And this is one way we can get insight into that. So pretty exciting, both for athletes and also for coaches too. You can imagine if we instill this into the coach platform and you can now, know, coaches with multiple athletes, they just, you know, it's hard to read every single comment and pick out every single thing that's going on. And here we can almost have a little bit of a flagging system where, you know,
we've got, there's a sentiment flag that kind of goes up and you can go to that athlete and just investigate a little bit further just to check in on it. So I like that, that's just super promising. Andrea, there's one more fascinating finding I thought, and I'll just quote you in the paper, said that beginner athletes reported significantly less joy compared to both intermediate and advanced athletes.
Now why do you think that might be?
Andrea (:Yes, that's an interesting one. Yeah, that's interesting. The reasons might be like very, very different. Perhaps they're not already in that kind of training regime that intermediate or advanced athletes are enjoying already.
But also it might be that.
They're less prone to provide very, very happy comments and very sad comments as well. They are more stable in terms of the expression of their sentiment throughout their comments. It's not guaranteed that is like... It's just that they're not expressing as much joy, but at the same time, as much as sadness. This can happen with age as well.
And there are studies suggesting that it happen with gender as well. providing a more stable response is not like indicating that they're not experiencing maybe that kind of joy. They're still in the process of understanding what is going on maybe.
Paul Laursen (:Yeah. I want to ask
the same question to Marjaana an experienced coach with dealing with a lot of both beginner and more advanced athletes. Why do you think beginners in Athletica are detecting, or our study here is detecting less joy in the beginner's comments?
Andrea (:Yeah, that's interesting.
Marjaana (:It's super interesting. was going to ask you like, what was like, were there a lot of people commenting? Because what I find as a coach is sometimes it's hard to get people to comment on their sessions, right? ⁓ But after like top of my head, was looking back my own ⁓ athletic journey, coming back from like three childbirth, it was hard.
Even 30 minutes was hard and everything hurt and why is this so difficult and I feel so out of shape. So like your head is in the dark cloud and you just want that run to feel good and you want to feel fast and it just feels really hard. So that's really hard too. Like you said that when people are going through injury or illness, there's more sadness.
in the comments or when people are struggling with something there's more sadness than joy. So I'm just thinking maybe there's a connection there.
Paul Laursen (:Any thoughts,
Paul Warloski (:Yeah, I'm curious about, about the experience level and whether people, you know, there's that initial struggle of what it's like to be running every day and your body kind of hurts. But I wonder if older athletes just kind of start to accept that, you know, some days are going to be crummy and, you know, some days are going to be good and it's all, it's all good. It all kind of washed it you know, comes out in the wash. ⁓ I'm curious about that.
Paul Laursen (:But just to your question, those are all great, valid, realistic potential reasons behind that. think that's sort of spot on. Just to your comment, MJ, on the data size, 55,000 data points on this analysis. Pretty solid.
Paul Warloski (:Wow.
Marjaana (:Nice. Yeah.
Paul Laursen (:Again, that's getting more as we go and growing too, right? So this is just gonna become more and more powerful as we go. So we can certainly confirm this in the future, but I think it's pretty strong. I think it's, I'm kind of in a hybrid belief of the three of you. And it's just like, it's sometimes it's just a little bit tough to get going at first. But once you get that, once those adaptations are kind of in play,
Marjaana (:Mm-hmm.
Mm-hmm.
Paul Laursen (:the resilience develops a little bit and then you start feeling better. I think it's just something that's really important for beginner athletes to be aware of is that it takes a little bit of time. You should almost expect it to take a little bit of time to push through and get you to that place where you've adapted to the training a little bit. It doesn't happen right away. I think just if you know that going into your training,
Paul Warloski (:Mm-hmm.
Paul Laursen (:It's
just, it helps, like, of expect it a little bit, you know? Maybe it's going to be a little bit more soreness and whatnot and fatigue in the beginning phase. But man, once I get through that, I know what's on the other side. And I'm keen to become an intermediate or advanced level kind of athlete, because I know I'm really going to enjoy this training. I know that it's going to just keep me coming back for more and more.
⁓ I reflect on my daughter. ⁓ I talk about her all the time, but it's fascinating just to see how she's developed my own, call it addiction for training. And it's just, we were just, we were having a chat about it and she's just like, she goes, dad, now I understand you. I see how you always have to go and train now. So I think it's a really nice place to kind of get to, but it wasn't always that way. So that is my thoughts for you guys.
Paul Warloski (:How does the AI coach use the comments and the RPE and the feels like sections? How does that all play together?
Andrea (:Yeah, the main reason why we are developing this feature is to drive the better the AI coach to take decisions about the upcoming training sessions. We are currently experimenting, but not releasing the feature.
because it takes a lot to understand how much you have to touch the load or the duration of the intensity of the upcoming sessions to make an impact and eventually alleviate that deviation from the baseline that you just witnessed and reported. It is quite difficult for us even to understand
how much is too much in terms of RPE or feeling reported from the users, let alone those, ⁓ understanding how much you need to deviate from your baseline, from your sentiment analysis is even more difficult. So there are not ...
strict references in the literature of how much we should change the training load for the upcoming sessions when we detect a deviation from the baseline in a response from the athletes. So usually we play on the safe side and we try to instruct our coach to never
make a mistake in prescribing a session which is harder than it should be. But for us, we've been discussing internally a lot about that. The stirability of the system when you provide an input to the platform is something that we deeply care about. And we think that...
you should not be able to steer too much the system with the slightest comment or with the feedback you are reporting. Because we are already putting in place some safety nets, which is, for example, deleting the session or the workout wizard. So we think that if we make the system too steerable,
with inputs from the athletes, we are going to end up with an unstable system that is going to adapt too fast to comments and maybe, I mean, you wake up the next day and you say, this is gone. So, I wanted to bring up my training load again. But yesterday, I made a comment and now I regret that. We think we are going to mess things up more than ...
we can make them right. And as I always remind myself and the team is that we need to make these work for as much athletes as possible. And most of the times you end up doing some design choices that are good for the average and you lose the sense of personalization. But those choices are made in the attempt to make the
Paul Warloski (:Hmm.
Andrea (:the greatest number of athletes as happy as possible, if it makes sense. So if we can make happy the average athlete, we can make happy the most of the athletes. So I see where you're coming from, and I totally get it. And we are working on it to better understand, and we are shaping the system so it...
It's steerable to some extent with the comments and the feelings you are reporting, not as much as many users or athletes would like to, but we are playing it safe. This is a learning process for us because in the literature, you cannot find that kind of...
scientific paper that tells you how to do that. It's experimentation, but it's experimentation with athletes' time and training and bodies, so we cannot allow ourselves to mess up with that. So we're playing it safe, but for good reasons, at least this is our position.
Marjaana (:I mean, ⁓ it's a million dollar question, right? Even as a coach who has daily contact with an athlete, this is the hardest thing. How do you know as a coach when to ask the athlete, just take a day off or just go for 30 minute walk? It is so difficult and advanced experienced athletes, kind of know.
Paul Laursen (:And if you
Marjaana (:the beginner and intermediate athlete. I almost want to say that they almost always lean towards, I'm going to do something. I need to do something because that's in the plan. I don't want to go to Workout Wizard and make the workout easier. Or I don't want to delete the session because I have this goal. I need to get to there.
You know, like I think it's so hard to know when to say, just take a day off.
Paul Laursen (:You know, what's going to be interesting, MJ, is another feature that Andrea and his team are building. And that's really alongside the database. Andrea and I were talking about this a little bit before, but with the respect to the conversational element that's coming out soon, we're going to have that conversational AI is going to be able to inquire about history in the database. And this may assist with decisions.
Andrea (:Yes
Paul Laursen (:that we always feel should be alongside the athlete, right? I mean, and again, back to the Dan Lorang podcast, he was really talking about on that, and maybe this is for more the experienced athlete, but he was just saying how important it is for athletes to make these decisions alongside the coach. they need, and if they can't do that just yet,
They should be working towards doing that. They develop their own agency. And I think that's really what we're trying to do in Athletica. We're trying to create a really useful tool that you can use to help optimize your own pathway and your own training. And again, back to the, you know, the Luke Evans podcast, Luke Evans won Ironman Canada with Athletica. He used it as a tool. He describes in the podcast how he uses it as that tool.
Like, you know, he knows when to go to the workout wizard. He knows when, you know, he knows how to use it as kind of a guidance and he knows it, it can't know every single thing all the time, but it's a heck of a great tool to keep you on track.
Paul Warloski (:Thank you.
So it sounds like it's important for Athletica users to use the comment section. Would we agree on that?
Andrea (:it would be more and more important going into the future, for sure. We in the process and the history of the comments would be leveraged, so it can be retrospectively looking at the past comments. So again, to establish a baseline. And we need to understand what kind of fluctuations are normal for
an athlete and what is something that is too much, is going outside the normal range.
And we are taking, in this case, are taking an approach which is very, very similar to that of the HRV fluctuations. So we establish a baseline and a normal range. If we can see that the weekly average of the happiness score in your comments is deviating too much from the baseline, so if it is going outside the normal range, we start raising a flag and we make modifications to the trend.
Paul Warloski (:Hmm.
Andrea (:plan. But again, we are slowly but steadily developing these tools, and we'll give them more and more power. But to...
to give the athletes something that we can take responsibility for, we needed to build something which is like deterministic as much as possible. So we need to be able to explain to the athletes why we are making those choices. And when you push too much with these kinds of tools, you end up building a system you don't have like full control over. So
we are happy to take decisions that might lead to a platform which is less flexible, but is more predictable so we can stand, we can back our choices and we can reproduce the mistakes and the errors and we can track them down and build on top of those.
If we're pushing too much on systems which are not predictable or deterministic or subjective, we end up with something that we cannot use on a platform like this.
Marjaana (:Is there a structure of comment that you would recommend or is it better to just have a free flowing comment after the workout?
Andrea (:Well, what I would suggest is to stick to what is important for you and not talking about something which is not related to the training sessions in particular. Something you think might be relevant so we can better understand. what...
athletes think is relevant in the first place. And again, we can establish a baseline for each one of them and push on personalization for all of them. But what I think is most important is to write down
in their own words, how they perceived the session, if the session was appropriate on that specific day. That is what I would recommend them. I would not recommend them to report the RPE or the field or something which is already integrated in the platform with different...
⁓ input formats, but they can freewheel and just put their comments down, how they felt during the session, or if the session felt too hard, or it was like they did that in the morning. If something that can be related to the prescription of the session that can help the AI coach to adjust the prescription down the road.
So this is what we're working currently.
Paul Laursen (:I think you should just, yeah, almost treat it the same way as you talk to a human coach. That would be my, I don't think it should be any different than that. Be honest with yourself and talk about the session, whatever you need to say about it. Good, bad, ugly, that's all going to be helpful.
Paul Warloski (:All right, well, thanks for talking about the study and getting some great suggestions for the kind of comments that we use. But we also did some research into the Athletica chatbot. And this is going to be a series of questions that will be mostly rapid fire for Paul, Marjaana and Andrea, ⁓ except for the first one, because this is the most common one. Is the Athletica right for me?
I there we go. Let's leave it there. We'll come back to that at the end. What sports do you use? Yeah.
Paul Laursen (:Of course it is.
Marjaana (:Yeah
Paul Laursen (:Well, I will say one caveat.
The one thing where it's probably not going to be right for you is if you don't have a wearable. If you don't have a wearable, ⁓ it's probably not right for you. It's going to respond with good wearable data. We always say garbage in equals garbage out. Gigo is the old adage. ⁓
Paul Warloski (:Yeah, that makes a ton.
Paul Laursen (:So you got to have, you got to be feeding it with the right info. So you need to develop a wearable and a wearable that is specific to the exercise mode sport that you are, that you're using. If you've got that, Athletica is definitely the right platform for you. I would, believe you, you connect your wearable, plug in your race date and Bob's your uncle. You're on your way. So on a daily basis, it doesn't matter what you do. Uh, you can do more, do less.
Do the session, you're gonna get feedback right away from your AI coach to keep you on track and keep you moving. Yeah, and that's so, if you got that wearable, Athletica is right for you.
Paul Warloski (:MJ jump in whenever you feel like it Andrea as well. ⁓ What sports do you currently support?
Marjaana (:What we have running, cycling, triathlon, duathlon, rowing and Hyrox .
Paul Laursen (:Yes we do.
Paul Warloski (:Do
you have an app?
Marjaana (:We do have an app.
Paul Warloski (:What apps does Athletica connect with?
Paul Laursen (:Well, I don't think it really connects with an app per se, but it certainly connects with lots of ⁓ devices. So it connects with Garmin, Strava, maybe, yeah, Velocity, guess, is an app, right? yeah. Concept2, Wahoo, intervals.icu. I guess these are all apps, all the various different partners, I like to call them, but yeah, platforms.
Paul Warloski (:Mm-hmm.
Paul Laursen (:So yeah, did I miss any?
Marjaana (:What's Watchtletic Watchtletic . Yeah.
Paul Laursen (:Watchtletic that's right. Watchtletic,
which integrates into Apple Watch. We just did that one. Thanks for reminding.
Marjaana (:And through intervals, basically, you have a Suunto, Polar, whatever watch you have that is not directly supported by us, then you can still hook them up to intervals and a Athletica
Paul Laursen (:That's right. Yeah. So say for example, you've got like an aura or ⁓ a whoop strap and you want to get your heart rate variability in. While we don't connect directly with those just yet, you can easily just connect in with intervals.icu and now you've got your HRV data on the recovery profile, which is super handy and a lot of people use that. So yeah, good point. Intervals is super convenient, super versatile as a result.
Marjaana (:And if you're using HRV for training, can connect it through interval via Dropbox. So there are ways.
Paul Laursen (:Yes, that's right.
Paul Warloski (:Can I import my training data from Garmin or TrainingPeaks into Athletica?
Marjaana (:Yes, of course you can.
Paul Laursen (:It is a zip file you can use. under history, I believe, and you can, yeah, back load your historical data as much as you want. You can put in there if you'd like that.
Paul Warloski (:Can I upload .fit files into Athletica?
Paul Laursen (:Sure can. Yep, just finish your session, click on the session itself if you haven't connected to the automatic sync and you'll actually be able to upload that fit file directly to the session that you did.
Paul Warloski (:My primary training platform is Training Peaks. Does Athletica integrate with Training Peaks?
Paul Laursen (:no, we don't integrate with TrainingPeaks just yet. So no, that's the one we don't.
Marjaana (:But you can
bring all your training peaks data to Athletica by the import. Just get your zip file from training peaks and come over.
Paul Laursen (:That's what I said.
Paul Warloski (:Hmm.
Paul Laursen (:Correct.
Paul Warloski (:Does Athletica send workouts to form goggles? I didn't even know what form goggles was.
Paul Laursen (:No, we don't do that, unfortunately, yet. We've asked for a conversation with FORM and they haven't responded yet. So if anyone from FORM is listening, it would love to chat to you.
Paul Warloski (:little zig in the kind of questions.
What do Athletica's strength workouts include?
Paul Laursen (:⁓ I tell you, MJ is our strength training queen. But yeah, I mean, like I'll start in general with the predominant types of sports like ⁓ cycling, running, . Triathlon These are mostly running based ones. And there's a standard plyometric session that's in there that anyone can do anyplace. You can have it, you can do it in a gym or you can do it on your
outdoors or in your basement or wherever you want. So ⁓ that's kind of stock standard across those programs. But then we have an entire Hyrox suite of sessions of that MJ video demos in all of them. It does an amazing job. And that's just everything you could possibly imagine because the strength training sessions within Hyrox are extremely diverse.
Paul Warloski (:Does Athletica include yoga?
No. Not yet.
Paul Laursen (:No.
Paul Warloski (:Although you can certainly ask me for yoga plans, but you can also ⁓ get yoga online as well. Does Athletica include swim drills and open water swimming workouts?
Marjaana (:There's a few open water swim workouts in the global library if you're not a triathlete. ⁓ There aren't many swim drills, but we know where you can find them. That's swimsmooth.com.
Paul Laursen (:Yeah, there are a few drills on the technique. Search for the technique. Remember with Athletica, you've got something called the global library. So you can actually just actually do a search in there. So do a search for technique swim in your global library and you will find a few drills to get you started on the technique workouts.
Paul Warloski (:I want to cycle, run and swim on specific days of the week. Can Athletica be set up to only schedule these on specific days or would I need to keep rearranging workouts each week?
Paul Laursen (:Andrea.
Andrea (:Yes, you can. You can move the constraints of the platform. We have a system based on
layouts that have been written by specialized coaches. So, you can deviate from those plans, and this has been also a discussion going on for years now. So, if you deviate a lot from those template plans, the AI will find it quite hard to find a solution.
If you deviate less, it will be easier. regardless, can technically you can define your own ⁓ days and time for it's one of the activity you want to train for. It's just like harder for the AI coach to find a solution to your problem. But technically, it's possible to do that.
Paul Warloski (:How do you determine FTP and FTP improvements? am currently using TrainerRoad and they have a monthly FTP update. Is yours similar or the same?
Andrea (:we are not calculating FTP exactly. We are estimating the critical power, which is a related concept, but which is computed in a different way. So FTP is usually...
retrieved with constant power exercises for 20 minutes or one hour and you take a fraction of that with the critical power using the critical power model to fit your maximal efforts and to find that kind of asymptote which is the boundary between the heavy and very heavy intensity domain.
So we are using that kind of technique and that is something that is updated every day. You upload a new session. So basically you can have like a new estimation each time you carry out a new workout. So if you update your profile with maximum efforts, ⁓ week in, week out, your threshold keeps updating and you can follow that.
oscillations throughout the season. So, yes, you can check the updates to your threshold, so to speak, and you can do that on a daily basis if you're training every day. It's a sort of moving window. So, we're taking into account six months of data for computing that threshold.
So if you carried out a maximum effort more than six months ago, that does not enter the computation. So it is not considered for the threshold estimation. So that's why we encourage athletes to carry out maximum efforts, at least on a monthly basis.
Paul Warloski (:Can Athletica help me train for my half marathon, then a marathon, and then an Ironman?
Paul Laursen (:Yes, it can. So yeah, again, it likes to focus in on each one of those events. You know, it relates a lot to the A or B race setting that you put that under. The A race will be the primary race that it's really targeting, but it certainly can, you you can manipulate the in your settings in terms of what race that you're putting down there and a...
a training plan will be put forth to work towards optimizing that. This is also very useful in the daily availability that Andrea was talking about as well, where you deviate from the template plans by actually setting in, like, say, for example, you've got that run and that bike focus. Well, you might want to have specific days where you're assigned to that. You might also want to work on the duathlon plan.
But irrespective, there's lots of different options in there where you can train for both and continue to tick over and then with an emphasis on one versus the other depending on what phase of the training that you're in towards that event.
Paul Warloski (:All right, let's take a few more questions here. I'm skeptical. I did try another AI platform and the estimates were completely off for the first week. How is Athletica different?
Marjaana (:I think that's a good question for Andrea. Why do we start with the test week? Say somebody brings in no history and start the trial athletica. What happens?
Andrea (:If you start trialing with Athletica and you don't bring the history, you don't have the reference values for defining the different training zones. So that becomes problematic because we need to start somewhere and you can input by hand those values, but you don't have a reference for that kind of target intensity.
So power profiling or speed profiling is a big portion of what Atletica can do. And we are using a seven zone training model. So to calculate those thresholds that can set the boundary between the zone and the other, we need some data to work on.
And in this case, of course, the more realistic and honest the maximum efforts are in your profile, the better the estimation becomes in terms of pace or power or heart rate, whatever variable you decide to follow for the prescription.
Paul Warloski (:Good. All right, last two questions quickly. What does Athletica cost in dollars and in euro?
Marjaana (:19.95 dollars what's that in euros
Andrea (:almost the same.
Paul Warloski (:the same. Is the Velocity app included in the cost of Athletica?
Marjaana (:No, it's five dollars per month. That is it. Yeah. 17 euros is 19,95. So plus the five dollars velocity, which is awesome. We have weekly live sessions. Yeah. So for four euros per month, you can join us for live sessions.
Paul Warloski (:All right.
Yeah. Great classes that we get to do our hard workouts together and our recovery as well. You're doing that now on Fridays. That's awesome. All right. Well, thank you everybody. Thanks for listening today to the Athletes Compass podcast. Take a moment now, subscribe, share, and let's keep navigating this endurance adventure together. Improve your training with the science-based platform, Athletica, and join the conversation at the Athletica Forum.
Marjaana (:Mm-hmm. Yeah, Friday is this for recovery.
Paul Warloski (:Dr.
Zignoli, Marjaana Rakai, and Dr.
Larson. I'm Paul Warloski and this has been the Athletes Compass Podcast. Thank you for listening.
