Resources
Nov 18, 2025 |
53 min
|
543 views
|

Integrating video & data in sports analytics | Arielle Dror | Data Science Hangout

ADD THE DATA SCIENCE HANGOUT TO YOUR CALENDAR HERE: https://pos.it/dsh - All are welcome! We'd love to see you!

We were recently joined by Arielle Dror, Director of Data and Analytics at Bay FC, a team in the US National Women's Soccer League (NWSL), to chat about working out loud, integrating video with data analysis, technology they use in sports analytics, and quantifying intangible player traits. Oh, and ICE CREAM!

In this Hangout, Arielle discusses how she works to integrate data into Bay FC’s decision-making processes, including recruitment, tactical analysis, and game preparation. She uses tools like Quarto and Posit Connect to automate weekly match reports. To enhance understanding among non-technical staff, Ariel’s team also uses proprietary sports software (Sportscode) to build dashboards on top of timestamped game footage with specific events tagged in it. This allows end-users to click on specific data points, such as those related to chance creation, and immediately view the corresponding video play that demonstrates the data's meaning. This visual context is essential for translating data results to coaches.

Resources mentioned in the video and zoom chat:
🔗 R and AI Conference → https://rconsortium.github.io/RplusAI_website/
🔗 Bay FC Data Video on LinkedIn → https://www.linkedin.com/posts/wearebayfc_bayfc-activity-7390100562731712512-xKIY?utm_source=share&utm_medium=member_desktop&rcm=ACoAACSkp_4BRz9mhkQZvnAk0Wdehn749sDDYJY

If you didn’t join live, one great discussion you missed from the zoom chat was about the "passion penalty". Attendees discussed whether working in a field you love, such as sports, typically comes with lower pay than other industries, especially given the high supply of passionate people who want to work in the space. Do you think the passion penalty exists? 🤔

► Subscribe to Our Channel Here: https://bit.ly/2TzgcOu

Follow Us Here:
Website: https://www.posit.co
Hangout: https://pos.it/dsh
LinkedIn: https://www.linkedin.com/company/posit-software
Bluesky: https://bsky.app/profile/posit.co

Thanks for hanging out with us! 💛

Timestamps
00:00 Introduction
03:19 "What traits in players do you wish you were able to quantify?"
06:48 "How is the data science work implemented on the field?"
10:50 "How do you translate results to coaches?"
12:44 "What tools do you wish you had access to?"
18:27 "Tell us about how you began working out loud and what that led to."
23:26 "Why is it so hard to recruit for sports?"
24:31 "Do you feel like there's a passion penalty for working in the sports world?"
28:31 "Are you working with PyTorch, scikit-learn, or OpenCV computer vision?"
30:28 "Do you ever correct anyone and tell them it's football, not soccer? Is the same set of data available for everyone?"
34:08 "Were there any sporting myths you've been able to analyze, like home ground advantage?"
37:28 "What's the best way to break into sports data roles?"
40:11 "Are athletes willing to share personal health data from wearables?"
42:06 "Is there commercially viable technology for 3D modeling movements?"
43:43 "How do you handle scenarios where play on the field contradicts data predictions?"
45:33 "How did you pick the analytics stack you use?"
47:41 "Do you have any advice for new people learning data science skills?"
48:44 "Is there a danger players will optimize individual statistics over team performance?"
50:22 "Is there a 'Believe' sign in the locker room?"
50:59 "What's the most important thing you learned about communicating analytics to stakeholders?"

image: thumbnail.jpg

Transcript#

This transcript was generated automatically and may contain errors.

Hey there, welcome to the Paws at Data Science Hangout. I'm Libby Herron, and this is a recording of our weekly community call that happens every Thursday at 12pm US Eastern Time. If you are not joining us live, you miss out on the amazing chat that's going on. So find the link in the description where you can add our call to your calendar and come hang out with the most supportive, friendly, and funny data community you'll ever experience.

Can't wait to see you there. I'm super excited to welcome our featured leader today, Arielle Dror, Director of Data and Analytics at Bay FC, which is a soccer team in the US National Women's Soccer League. Arielle, welcome. Thank you so much for being here.

Hi, Libby. Thanks for having me. I'm really looking forward to today.

Me too. Okay, I would love it if you could give a brief background about yourself. Tell us a little bit about what you do, and also something that you like to do for fun. I know that's a two-part question. I'll remind you about the second part.

Love it. So I am currently the Director of Data and Analytics at Bay FC. We're one of the newer expansion teams in the NWSL, National Women's Soccer League. I work on the football ops side, where I work to integrate data into all of our decision-making processes, whether that be recruitment, tactical analysis, game prep, match review, salary cap analysis. If there's something that the sporting side is working on, I'm trying to integrate data to try and make our decision-making processes more informed.

And what I like to do in free time, well, I have a puppy. His name is Bingo. He takes up all my time these days. He's not here right now, unfortunately, otherwise I would have him make an appearance, but maybe I'll post him on Blue Sky after. I'm also an avid knitter, currently working on my second sweater of the season, and I also really enjoy swimming.

Nice! Wait, sweater, where are you based? Because I'm in Texas, it's hot, I never get to wear anything that I knit or crochet.

I am based in the Bay Area, in San Jose specifically, which means I also don't usually get to wear the stuff I wear until I go back home to New Jersey, but it was chilly the other night so I finally got to wear one. Yes, it is pretty warm, but I struggle to show, I suffer just to show off all my wares.

I love it. Well, there are lots of knitters and crocheters and sewists in the data community as well, so you are in fantastic company. If anybody is a knitter, Isabella, knit-are, yeah, I love it, both. If anybody else is a knitter or a crocheter in the group, let us know. I know that, I feel like Marcy is often knitting whenever she joins, which I don't think I see her today, but I mean, love seeing that. Tim says he's a crocheter. I know Hadley's a crocheter as well.

Quantifying player traits

Okay, well, let's start chatting. First of all, SlidoLink, everybody, if you don't remember, we'll put the SlidoLink in the chat. It's where you can ask questions to Arielle today, and we do have an anonymous one that says, what traits in players do you wish you were able to quantify, and do you think we'll ever be able to quantify traits like drive in athletes? Oh, wow, that's a great question. We're starting off with the hard questions. This is great. I know, it's anonymous. If this is you, feel free to put it in the chat. Congratulations on a fantastically tough first question.

Yeah, this is great. So I will say, the data that we have in soccer is slowly evolving, and it might be helpful to just give an overview of what we have available to us right now. So most of the data we have is event level data. So if you think of a soccer game, there's one player that is usually on the ball, and then there are every other player on the pitch is not on the ball. And so this event data only collects on-ball information, which means we are missing like 95% of what happens in a game. So we have things like passes, shots, defensive actions, things like that. We don't necessarily have off-ball runs, where people are in space, et cetera. And so most of what you can do in soccer is actually quite limited.

We're now getting to the point where we have that off-ball information, where you can actually quantify off-ball impact. So we're slowly getting to the point where we can quantify a lot of things. I do think that often in recruitment decisions, we start going into these personality level questions, and things like drive, or dog, or how fancy they are. I would love to be able to quantify those things to be able to actually give an informed discussion to actually drive what we're doing. So I might actually just take the example that you gave here, which is drive, or just tenacity. That's something that often is really talked about in our discussions that I really can't contribute to, and I have to try.

It's tough. I think that anybody who works with people data, data about people, qualitatively or quantitatively, will say that anything that's a personality trait is extremely hard, and the results are dubious. Humans are complex systems, and anything that humans participate in are complex systems. I don't think we can ask, based on how recruitment works, we can't give them psychological evaluations either. Right. And the question is also, should we? Was that a thing that you even shifted to? Yeah, which is true. I think realistically, most players would be like, I'm not going to come here if you're going to give me a psychological evaluation.

Right. It's kind of like when you start a data science job or data role, and you get a really weird take-home project that you're like, I'm not doing free work. I don't want to work here. Exactly. Yeah, very weird. Okay. Fantastic question. These are all curious. And we suddenly got a bananas amount of questions. So let's get going and see how many of these we can answer.

Integrating data into tactical decisions

Can you share from Suraj? Let's see. Oh, and there's a follow-up. Suraj, if you are available to ask that question live, I would love that. And I'm going to highlight it in Slido. So if you're in Slido, you will be able to see Suraj's question at the top. Oh, fantastic.

Yeah. Thanks Libby. Can you guys hear me? Yeah. Yeah. Hi Ariel. I was just curious about how the data science and analysis work that you do is implemented on the field. Specifically like when it comes to you presenting some insights and whether they are sort of like implemented on the field by the team. Sometimes there can be an opposition. Players may be, some of them may be kind of like, depending on their personality, relying more on intuition and their skills. They may not be receptive to like data aspect of it. So I was just curious, like what is your experience like and what are the instances when you see that, that kind of work, data work being kind of acted on in the field and results generated? Just curious.

Yeah, that's a great question. So I'm really lucky at Bay. We have two video analysts that work under me who do a lot of the work in terms of integrating data into our tactical decisions and sort of in our weekly scouts. And one of them is particularly data driven. He worked with the U.S. Women's National Team for several years, won a World Cup, two World Cups with them, two, one, won several World Cup, won a World Cup with them. And so they had a lot of data folks on their team. So he's actually pretty data literate. And so he, whenever I build our weekly pre and post match reports, he will take a lot of that data and integrate it into our video scouts that go to the players and to the coaches. And he uses it to inform some of the tactical recommendations that he makes each week.

We will occasionally put like data in front of the players to sort of emphasize a point. We went through a period this year where we were dropping points where we shouldn't, like we're probably, the probability of us dropping the points we did was like less than 1%. So we put it in front of them to sort of emphasize the point that they were doing well. So we do it a lot. We do a lot of work like using data to increase the confidence of our players. But we do use it, like I said, for tactical decision-making too. So a good example is we have a pre-match report that I have automated that goes out every week. And one of the plots that I show is like who is the top chance creators on the team based off of a lot of different things that we do. There were two players who nearly all of their chances were through these two players who were not forwards. And so we designed an entire tactical plan around keeping them off the ball. And they did not get on the ball. And then we were able to win the game. And we were able to also use data afterwards to check our process. So we were able to see that they actually did not have any chance involvements. So that was like probably the cleanest example we have of where data is used to inform the decision, to find the information, inform the decision, and then actually evaluate after whether it worked.

So that was like probably the cleanest example we have of where data is used to inform the decision, to find the information, inform the decision, and then actually evaluate after whether it worked.

But that's, we're doing, yeah, we're doing lots of different stuff.

So many sports balls terms right now. I need a translator. Okay, hold on. What are chances? Chances. Okay, that's a good point. Yeah. So chances to score. So like a shot.

Okay, so how can someone who is not, who doesn't have the ball provide a chance for a team score?

Yeah. So, you know, you don't just win the ball and immediately shoot, right? Sometimes there's many passes in the build up before a player takes a player takes a shot. So there's an assist. And then if you think of hockey, there's like the hockey assist, the second assist. And you can go all the way back. Maybe there's 10 passes before the shot. Like we still want to know that player is involved in that build up. Because that means that they are getting on the ball, probably in dangerous locations. Okay. Interesting. Thank you. Sorry to interrupt. Had to like get a few. No, that's great. I appreciate like, if there's anything sports wise, that is confusing, please let me know so that I can clarify.

That's all right. And there was a follow up to Suraj's question in Slido. In Slido, by the way, you can hit a little reply button and like add context questions. That was how do you translate results to coaches? And I was wondering that too, when you said that you had automated like a pregame or a pre match report? Is that like, scheduling a Quarto document report in Connect? Or is that like a quick what? How does that work?

Yeah, so it is we use Posit Connect. We love it. And it's just a Quarto document that is scheduled to run every Sunday. And then I should automate sending it also. But I was going to say, do you automate the emailing? I would love to. But I want to be able to check in first. So I download it. And also, something that I've learned, and I'm sure this is true in other industries is that the less technical you are, the more less likely you are to actually log on to the portal. So it's usually better to just download and send it to them manually. But our analysts will go onto the platform and look at it. You're meeting people where they are, and I think that's really important.

So the follow up part was, how do you translate results to coaches? And I guess that's part of it is like, meet them where they are, if they're non technical people. Yeah, be really open to questions. I think I was really lucky with this coaching staff, who's unfortunately moving on at the end of the season, they had never worked with a data person before. So they had no preconceived notions about what was possible, or really like a lot about a lot of concepts. So I was an open book, they were really eager to learn from me. Last year, I actually sat in the office with them. So just being able to integrate my thoughts and share my insights, and allow for conversation was a big, a big help for us. But yeah, it's just been a lot a long term process of conversation, teaching, informal moments, and just trusting that we both have the best interests in mind.

Tools, technology, and integrating video with data

I love it. And there's something, there's an anonymous question that is so in line with what I want to talk about. So I'm going to combine them. The anonymous question is, what tools or methodologies do you wish you had access to? If, you know, the National Women's Soccer League had the same funds as something like the NFL, what I really want to talk about is the fact that you do have access to so many cool technologies. And I put a link in the chat with a video on LinkedIn that just came out like a few days ago, about how you guys, you use drones, you've got like cameras out, out the wazoo. Also, you told me that you like, I don't know if it's Shiny apps or Quarto reports or something, but you actually embed video into your reports, so that when you have a point to be made about, like, here's what I'm showing you with the data, somebody can click it and watch a play that demonstrates what that data is trying to tell them. Can you talk about that?

Yeah, I can talk about that, and then I can also answer the question because that's a fun one. So in terms of using data and video together, James, I see James is in the chat, and he'll be really familiar with this. There's a program that's used in sports called SportsCode. It's a proprietary software that's used specifically in sports, and it allows you to basically take a video feed and then tag it with specific events. I don't know a lot about SportsCode. This is like not my world, but something that's really cool about SportsCode is that you can actually import your own, like, specially formatted profile, like a file that is timestamped to the video, and then build things on top of it that really just look like dashboards. They're called SportsCode windows. They really just look like dashboards, and so our analysts and I have worked together where all of the data that we care about is timestamped, and then I format it in the format of this SportsCode format that is necessary, and then he's able to actually build dashboards on top of it. So, for example, we have one around chance creation, so he can click all the chance creations that are created, or all the chances that are created via through balls, and he can click on it, and then it'll pull all of this data that we have on all the through balls, and it'll give us, at the timestamp, he can watch the video along with all the data tidbits that I add at the bottom for information for the coaches. And at the beginning, I mean, I've talked a lot about collaborating with coaches, and I found out halfway through last year that when I talked with just pure data terms, they had no idea what I was talking about.

I wish they had told me sooner, but August works too, and so we were able to build these SportsCode dashboards, basically, where whenever they had the video, but they also had the data, and it worked incredibly. So, some of our reports don't have video, but the ones that our coaches use most often do, and then I was able to automate all of that, and so now, because of PositConnect, now we have a shiny app that downloads all these files, generates all these files, and our analysts can put them on our lovely SportsCode with us, which is great.

This is amazing. You are so far ahead of most data-driven organizations. I see a lot of nodding happening in my grid of faces right now. There are those of us who have worked at these corporations who have millions of dollars to spend on their analytics functions who can't get anybody to use any of their data products effectively. You're killing it. This is amazing.

I appreciate it. I know it's easier said than done. It also helps that we have a really small org, right? There's 20 of us in the sporting organization, so I can actually have touch points with people. Scaling things is tough. It is really, really tough. That was fantastic. Thank you so much for the question, but I do want to go back to the question.

What do you wish you had access to?

What do you wish you had if you had access to it? Magically, more funds. More funds and also a league office that is able to do technological innovation. I think that's the big one. NFL has next-gen stats and has a whole internal data science team, and my understanding is that those products are available to teams.

There's a lot of really advanced data. I've talked hand-wavy about this tracking data, which is essentially sensor or GPS data. It's a lot harder to work with than event data, and most teams don't have the capacity to work with it to the scale that they would want to, and so having a league office that is able to actually process that data and give out-of-the-box metrics and maybe even some dashboards would go a long way, not just for us, but also for any team in the league. We probably have one of the most built-out analytics departments in the league, and even we are struggling to think about how we would approach that, so just having people at the league office who are able to process data is really a huge help.

I would love to have something like next-gen stats. That is so sick. Also, they do a really cool hackathon every year that helps teams find talent, and it's really hard to recruit in sports, and so having a capacity to do something like a hackathon would be really, really cool. Amazing. I mean, I think hackathons can be as small as you want them to be. It could be really fun. All right, let's get in some more questions because we have so many. There's no way we're going to get to all of them, but let's see what we've got here.

Pardon me while I take a minute to pop in here. I think that there was one about breaking into sports analytics. Sorry, y'all. When the questions pop up in Slido, they all move, and then when y'all upvote them, they bounce around my screen.

Breaking into sports analytics by working out loud

Okay, so let's assume that I did see the question about breaking into the field because this gives us a really great opportunity to talk about how you got started, and it's my favorite part of REL lore. Tell us about how you began working out loud and what that led to. Yeah, so I got this entire job by just obnoxiously working out loud until this paid off for me. I went to Smith College in Northampton, Massachusetts, historically women's college, also one of the first schools to have a data science program, or at least a liberal arts college. So I didn't go in thinking I was going to be a data science major, ended up being a data science major.

So my senior year, I was taking a class, an advanced programming class in R with my professor, Ben Balmer. I'm going to shout him out because he's fantastic and quite involved in the R community. We had to build an R package, and I had started watching. It was right after the 2019 World Cup. I was feeling really inspired. I started watching the NWSL, where I now work, and I wanted to learn more about Soccer Through Data.

I basically scraped the website, all their stat pages, put it into an R package, and then over winter break, I just put it on Twitter. I didn't really think it was going to pop off or anything. I just put it up there to see what would happen. I wanted people to use it. I wanted feedback. It went mildly viral, so that was really helpful. From there, I was able to get some opportunities working for American Soccer Analysis, which is the main open-source soccer writing blog. I was writing for them and then doing some work for them behind the scenes, and then that turned into my first job. While I was doing that job, working for a multi-club organization, doing data analytics and some data engineering, I was also putting up post-match reports for NWSL every single weekend, just to keep my name out there.

I got this entire job by just obnoxiously working out loud until this paid off for me.

They weren't really that great in the beginning, but I was getting lots of feedback from other people on Twitter who were using them. After doing that for two-ish years, the Bay Area team was announced, and I really, really wanted this job. I made sure everyone knew I wanted this job, and it paid off.

Bekkah, what kind of timeline? How long between you posting on Twitter and being like, YOLO, I'm just going to put this out there, it's not perfect, between you securing a job in sports analytics? I want to say that I am very lucky, and it doesn't usually happen this fast, but I put the R package up maybe January 2020. I think COVID helped here. People were really on the internet a lot, because they had nothing better to do. 2020, and then I got my email from my first boss October 2020. Again, I don't think that it usually happens that fast, and I was prepared for it to take a lot longer, but it really worked out for me. Then from that job to the BFC job, that was like two and a half years. Would you say that you endorse working out loud?

I do endorse working out loud. Even if it doesn't turn into a job, you make lots of really great friends, which is maybe even better. I've made so many great friends just through working out loud, and people who eventually became my co-workers, and people who I've traveled with, people who I use as professional sounding boards even now, now that I work alone again.

Shout out to Tan. Tan was on Instagram. Is Tan on here? He's great. Tan is here. Tan also works out loud a lot, and has done a lot of great work. Tan and I were co-workers at Zealous, and he's the best. Shout out to him. Tan is the best. All right. Thank you so much. That was amazing.

By the way, with the Discord server that we put the link to, that's a fantastic place to go work out loud. If you are worried about posting your work on Twitter, or Blue Sky, or LinkedIn, and you're like, I don't feel like I'm in a safe place to do this, we have a channel called Only the Silliest Projects, which was spearheaded by Dami Pak. If anybody knows Dami, she is an amazing silly project person. That's a fantastic place to put stuff, but there's also all kinds of places to just share what you're doing, and the only people who will see it will be nice, supportive data people. Please use that to your advantage, and I would love to see what you're working on.

Recruiting challenges and the passion penalty

Okay. Let's head over to the Slido again, because I see we have a question from Aaron. Adam, thank you for helping us get that onto the Slido, because Slido is blocked for Aaron for some reason. It is, why is it so hard to recruit for sports? What are the challenges?

Honestly, I think I saw someone write this in the chat. The pay is just not on the level of what we would expect for people with technical skills, and that's something we're trying to change our pay. That's something I'm trying to work on behind the scenes, so that's definitely a challenge that we have. Also, we often want people who have some level of sports experience. It's really hard to get that experience, so working out loud is really helpful. Those are probably the two big ones. I find the most talented people that I would really want to come work here are just not willing to work in sports, which is super fair. It's a really hard, it's definitely not your typical tech job or typical data job.

Yeah, there was a question in the Slido about this. It is not a nine to five. That is right, Jane. I think the question was, oh, Isabella highlighted it for me. Isabella, you are killing it. Thank you. This is amazing. It's, in the past, sports analytics jobs I've seen have tended to pay a little bit less than other industries. Do you feel like there's a passion penalty for working in the sports world? I have heard this sort of theory a lot, that if you get fulfillment out of a job, you have to make less money for it, which I don't think is true universally. What do you think, though, about sports?

I would say definitely. Actually, let me backtrack. I majored in data science. I also majored in political science. I was really interested in doing data science for social good before this. I feel like every job I've had has paid a passion penalty to some extent. I would say it's true. Especially here in the Bay, I look at pay ranges for jobs. I will say it is an anomaly, but I think this is true throughout the country. It's definitely true. I think it's getting a lot better compared to where it was when I started. Maybe Tian can chime in here, too, because he's been in this world. Or James. It's definitely gotten better, but there definitely is some level of passion penalty, especially because those jobs that you're comparing it to, the work-life balance is just not the same. You're usually working in a smaller team. The projects you're working on, the type of people you're interfacing with, it's definitely very different. Not necessarily bad, but there's definitely a passion penalty.

Where are some of the benefits of that? I think that's a great follow-up. You're working on a smaller team. Your hours are not normal. Oh, Chris Fabic is also here. Hey, Chris, who is an alumni of our data science hangout as well and is a ref for high-level soccer. I'm wondering what the benefits are, because I feel like when we're like, it's non-standard, it's lots of hours, it's weird, it's small teams. I think you also get to do a lot of cool stuff that people in much larger orgs don't get to do, don't get to put their hands on. Something that I really love is that I get to see my work actually put into play every single day. I know I'm making an impact with the work I'm doing. There's also the fun stuff of working in a team. You don't have to be experienced all the time, but I travel with the team. I've gone to see the world, quite literally the world, and meet really interesting people, work on really novel ideas. Those things I don't think I necessarily would have had the opportunity to do in a different industry. That stuff is worth it for me.

Something that I really love is that I get to see my work actually put into play every single day. I know I'm making an impact with the work I'm doing.

Ice cream and travel

I'm putting the link to that video again, that Bay FC video, because you talk about something I really like in there too, which is you developed a habit of going to get ice cream. I think it's before a game, right? Yeah, the night before a game. The night before a game. You went to go get ice cream one time, and then y'all won that game. You were like, well, now we got to go get ice cream before the game every single time to make sure we continue winning. It's true. But it means that you've gotten to explore your favorite ice cream places all over the freaking place. Absolutely, absolutely. It's been so much fun. We haven't done the data set. Oh, go ahead. Sorry. I was going to say, how have you not studied this? We really want to. The problem is that we don't have time.

So we have our non-official ranking, but we wanted to actually do a formal study. Unfortunately, we know that the number one is far and away the best one, and that is Blacksmith Ice Cream in Salt Lake City area. If you are there, really best ice cream I've ever had in my life. It is so good.

Cool. Maybe you can get on the Discord later and give us some tips on ice cream and also share some dog pics. Absolutely. Some puppy pics. We have some more questions in Slido here, so let's keep going because we've got about a half an hour left, and it's not enough time to answer all of these good questions. We had a question from Edward. Edward, if you would like to ask that live, that would be fantastic. A little bit more on the technical side.

Computer vision and modeling in sports analytics

Hello. Thanks for choosing my question. Yeah. Sorry. The question is a bit more technical because y'all are talking about you've got some really data-oriented people. Are y'all working with PyTorch, Scikit-Learn, or OpenCV, computer vision framework stuff for feature extraction to analyze the game video for more subjective traits like tenacity in players? Because I need a player position, but that tenacity concept was really stuck with me. Yeah. I love that question. I would love to be doing really fun machine learning and cool novel work like that. Unfortunately, we have a lot more low-hanging fruit that we need to meet first before we can do really cool stuff. I actually don't do a ton of modeling in my day-to-day. It's a lot of data engineering and a lot of much more simple data analysis, data viz. One day, I would love to do something like that. I do also think we probably need to get a larger quantity of data or a video before we could feel comfortable doing something like that, but it sounds really cool. If someone does that publicly, tag me because I would love to see it.

For sure. Thank you for asking the technical question, Edward, because I also have this question a lot for a lot of people, and I don't want to ask it because sometimes they're like, oh no, I wish I did, which was your answer as well. I wish I did all that technical stuff, but it's really nice to hear, though, that not everybody is doing these really super complicated computer vision models because it makes me feel more normal where I'm like, oh, the stuff you see on LinkedIn where everybody is doing cutting-edge, bleeding-edge stuff is not really reality. I think it makes us feel a little bit bad about

ourselves sometimes, so fantastic. A plus answer. Let's hop over and grab some more questions here.

Oops, I marked one as answered. I hope that I marked Edward's question as answered. When I mark them as answered, they disappear from Slido, so it makes me nervous. Okay, this is anonymous, and it's a two-part question. The first part says, do you ever correct anyone and tell them it's football, not soccer? And number two is, is the same set of data available for everyone beyond, say, FB ref and Opta stats? I don't know what those are. So those are like FB refs. Oh, Rajesh, this is you.

Rajesh, do you want to give some background to this question?

Rajesh, it was anonymous, but then in his reply, it said his name. Hey, Rajesh. All right. Can you hear me now? Yeah, perfect. So yeah, a brief background is...

A brief background is my questions background is basically I used to work for an Asuka racing team and the data that you see on the tv is not even close to what the data that we used to get behind the scenes.

Question – How would you go about getting that same kind of data when the players are actually playing, even for the opposition players? Because when you jump on a player and especially given that football now has become football, it's actually becoming more structured and less about players and their talents themselves. So how would you go about getting that data? I'd be very curious to hear that.

Very cool. First of all, that sounds like a sick job and someone also asked in here wanting to learn more about your past life. I also would like to learn more about your past life. That sounds so cool. In terms of live data, we get a little bit from the league from Opta, mostly just in the form of a dashboard where there's nothing we can really ingest, though that might be changing next year, fingers crossed.

All the data we get is post-game. So we get stuff from our data providers ingested via API. Does that answer your question? We get more data than what is available publicly, but it's not from the league. So some teams in the league do not work with those data providers at all.

Great question, Brijesh. If you do want to hear more about Brijesh's past life, Brijesh is going to be on The Hangout next year. Yes, we are scheduling it next year. It's so weird when we get to the end of this year. So Brijesh is going to be with us February 12th, right Brijesh? Yes, I'm flipping through my schedule right now, February 12th. So we will get to talk to Brijesh and now I know to ask him all about his NASCAR past. Yes, we'll look forward to that. Thank you for that fantastic question. And there was one part of it, which was, do you call it football? Do you call it football or do you call it soccer? I use it interchangeably. Sorry for the boring answer. Depends on the day. Depends on who I'm talking to. Depends on the vibe. Okay. All right, cool. We have another question here

from Ntobeko. Do you want to ask that question live? I'm going to highlight it in Slido so we don't lose it as things are bouncing around. Let me see if I can find you to unmute.

Yeah, hey. Hi, how are you guys? Okay. Can you hear me? Is my sound good? Yeah, you sound great. Okay. You've kind of answered this question already from Edward's question, but what I'm asking is probably based on the more social sort of community audience base, where I was wondering if there were any sporting myths that you had prior to landing a role that you were eliminated by after joining the team. I'm thinking about things like, is there any data around the home ground advantage or audience participation having any sort of influence in the game's performance? Because I think about things maybe like if they're having war cries and stuff, if you can measure things like the decibels throughout the stadium, does that, can that have a correlation that you can sort of look at later to sort of say, hey, we always say this, but here's the data that actually sort of supports that. Yeah, that's a good question.

That's a great question. So we don't do a ton around audience home field advantage, etc. Most of the stuff we use there is just stuff that's publicly available because I try and use as much as publicly available as possible because it's just me. There's been some stuff around different tactical decisions that have happened or in-game moments. I want to share them, but I feel like they're kind of competitive advantages, so I don't want to share them too much that have been interesting to prove me wrong. But I will say a lot of those types of questions, we use publicly available information. Just again, there's not a lot of time to do some of those analyses.

I know that this sounds like a fun thing for like fans to tackle. There are some questions that like if you are curious and if you think this is a fun thing and you can find data that is publicly available, it would be really fun. I'm sure there are some home team advantage analyses out there for some sports, right? I think that would be super fun.

Yeah, that is my favorite question of the day. That was such a good question. I am always curious about these things and in analytics, it's like every idea is interesting, but you know you can't do them all and you have to prioritize and choose between questions. So, the fun ones often like fall off the end of the priority list. Unfortunately, there's so many fun questions I would love to do, but the reality is if I spent all my time doing fun questions, I wouldn't have a job anymore. So, yeah, one day I'll get to go do the fun ones again.

And also, I wanted to call out Rory mentioned in the chat that soccer is actually British as well. I learned this just recently that like the term soccer is not an American term. It came from England. We somehow ended up just using soccer and the UK ended up using football and like very interesting. So, go Google that. Let's hop back over to the Slido and get some more questions and in our last 15 minutes, y'all, let's see. Oh, we have an anonymous one about breaking into sports. We kind of talked about this for the working out loud part, but is there anything specific that might help someone break in to sports analytics like projects in a GitHub repo or how to, you know, network? Are there conferences that you could attend for sports analytics? Let's talk about that to cover this anonymous question in Slido, which is for someone coming from business analytics, what's the best way to break into sports data roles?

Breaking into sports analytics

That's a great question. So, first of all, I'll say that a lot of clubs now also do have BI roles. So, that is totally an avenue that you can take. Like we have two or three business intelligence people on our business side who are lovely and I've talked to many people who do that, which is great. I don't know enough about BI to say if the work translates very well. So, I don't want to say anything there. I do think like in terms of networking, working out loud, et cetera, there's a couple conferences that I've gone to that I've found particularly friendly.

America's Soccer Insights Conference Summit. I can't remember the full name. In Houston, I went last year. It's only the second year. It was the most friendly conference, except for maybe Posit Conf that I've ever been to. And they especially, they like particularly plan things so that industry professionals are talking to people that are interested in breaking in through like lots of breakout sessions. So, I really thought what they did was great.

I've gone to a lot of like women's sports data type conferences. I've also found those to be really friendly. So, women in sports data, it didn't happen this year, but in the past, it's been super friendly and lovely. So, I highly recommend following people on Twitter. I mean, Twitter is dead now. So, Blue Sky, I guess, is the best option. Like following people on Blue Sky is often how I did most of my networking and just talking to people. I'm trying to think what else. In terms of like projects that people do, I find like the stuff that is quite interactive and not just like a single report or article. It's the stuff that draws me to profiles. So, packages, dashboards, that sort of work is, to me, just much more compelling than like a single analysis. I would agree. Finding some way to, quote unquote, or air quotes, deploy something, right? Just put it on the internet for other people to click around in. 100%. And maybe interact with, do an interactive thing. Fantastic.

Finding some way to, quote unquote, or air quotes, deploy something, right? Just put it on the internet for other people to click around in. 100%. And maybe interact with, do an interactive thing.

Wearables and athlete performance data

I saw a question from Nick Drew that was about wearables. And this is a great question. I've seen it pop up in the Zoom chat as well, people talking about wearables. So, I'd love you to ask it live. Hi, Libby. Ariel, thanks for presenting today. Yeah, I was thinking about athlete performance. And I don't know if that's something that you do or work with that kind of data, but it seems to me there's a lot of factors that would influence an athlete to perform well or not so well and made me think about secondary data sources. Like, is that something you do? And if so, are athletes interested in participating in that by, I guess, willingly give up some of their personal health data? I know my smartwatch collects a lot of data about me. So, I've got to imagine that that could be a goldmine of information if people are really interested in modeling athlete and team performance.

Thank you. Yeah, that's a great question. So, to answer your first part, we have a sports scientist on staff who handles all of our sports performance data. We're slowly getting to the point where we can collaborate together more, but she handles most of that day-to-day. She's great. So, all of our players wear GPS units in-game and in training for us to track things around sprints, accelerations, decelerations, heart rate, all that kind of stuff. And in some leagues, that's required. In other leagues, it's voluntary. It's all up to the CBA. I think in our league, it's required. I've never seen any player not wear one. So, that's how they collect most of their data. The players also do fill out daily wellness surveys around sleep and whatnot. I don't think we're taking any data from their personal wearables. I think that might be a privacy violation. I don't think it would work. I don't think it would go well in our league personally. Let me see what else I want to cover there. Yeah, those are kind of the big things.

Yeah. Does that answer your question? Perfect. Privacy is an illusion.

Motion capture and tracking technology

We're getting distracted by the chat again. Okay. That was great. And there's also a question that is kind of a follow-up on that that I wanted to smoosh in here as well, which is from Nathan. And Nathan had said, like, is there commercially viable technology that you can place on players and on the ball to 3D model all the movements throughout the game? I'm pretty sure that technology is possible, but is it worth investing in at the moment? Because there's definitely motion capture, and we've talked about this with other sports analytics leaders, that's just from the video capture. Can you talk a little bit about that? Like, what is motion capture through video versus having them actually wear things?

Yeah, so I will say I'm not familiar with any 3D motion capture through sensors at all, so I don't want to talk about something I don't know much about. It might exist. There's tons of 2D motion capture stuff. So there's obviously the GPS units we've talked about, and then there's also optical tracking systems where the cameras are set throughout the stadium, and they're collecting all the information from all the players and also the ball. And then what we use more often is broadcast tracking, which does the same thing but uses broadcast video. So it's basically using computer vision to track the players in the ball that are on camera. The challenge there is that in some leagues, the broadcast quality is not as good, and they have to use models to impute the off-ball players, which is a challenge for many reasons. Yeah, that's tough. Imputing is hard at all. Imputing is really hard. Coming up with what you think the data probably was when you don't have it is really tough and like fraught with problems.

Handling false positives and data versus video analysis

There was another question from Suraj that has a little asterisk, so I'm going to read it for him, which was, are there any scenarios of false positives, like when the play on the field contradicts the predictions from the data, and how do you handle that if that does happen? Yeah, I mean there's plenty of moments where that can happen. So we always try and, like every half six months, we look at our reporting and see where we were right, where we were wrong, what we might need to add to capture things. Sometimes things happen in games that are not quantifiable in data, or at least not easily. And that's why, you know, going into this, I only wanted to use data to make all of our decisions, because that's sort of how I operated in my last job. And I still live in that camp a lot of time, but I've also learned that using video analysts and using their expertise in collaboration with the data is really like where our best work is done. So it's a lot of interrogation of where those things might diverge, fixing things, seeing where things might be breaking, and also accepting that people, like everyone's best skills work together to give us that answer that we're looking for.

using video analysts and using their expertise in collaboration with the data is really like where our best work is done.

Yeah, I like that. It's a nice little like reminder that subject matter expertise is like really, really important, and also that when you're wrong, it can be a feedback mechanism into your data to either say,

okay, clearly we're not tracking all the variables that are confounding what we're predicting, or we're just not going to be able to know that. Like we're not going to be able to track that, and we have to be okay with it. Fantastic questions, Suraj. I saw you saying in the chat, I don't want to take up too much time. I don't want to ask more questions. No, ask them.

I love questions. Yeah, we love the questions. Give me one second while I go through the Slido here and peek at more questions. Do you mind if I take one of the questions, Libby? Absolutely. There's one that I really like here. Someone asked how I picked the analytics stack, and I really like that question because I was in a pretty interesting situation when I joined Bay, where I joined in October, and our expansion draft where we picked players from existing teams was in December, and when I joined, we had absolutely nothing. We had no contracts signed except for one single data contract, and we were already making decisions on player transactions, which is really stressful. So I was tasked with deciding really early what that stack was going to look like, and so I did my best to try and find things that I knew were really flexible because I didn't really know how we were going to grow. So that's why we ended up with Snowflake, for example. I felt of all the warehouse solutions, that was the most flexible for what we wanted at the time, and then I also had to work really quickly to build pipelines to get this data into a workable format. So at first, I just built R pipelines, which is still great. I've worked in full R stacks for a long time, but I built everything working off of my computer. There was no orchestration. Nothing was deployed, really, and so in the second year, I was finally able to build a more coherent, error-proof tech stack using Astronomer and Python and R and a lot of different things, but the point I wanted to make here is that I worked really hard to build a flexible tech stack because I didn't know how things were going to work and also was okay with having something really crappy at first with the knowledge that I was eventually going to go back and make something more robust, and if anyone is in the situation, I highly recommend taking that approach. It's okay that things aren't perfect right away. Sometimes things just need to work. Sometimes things need to work. Sometimes you're already making decisions without any data and you need things to give you context.

It's okay that things aren't perfect right away. Sometimes things just need to work. Sometimes you're already making decisions without any data and you need things to give you context.

Advice for learning data science skills

I love it. There was another question that was about the tech stack as well, and there was an add-on to it, which was, do you have any advice for new people learning these data science skills and getting experience? This is a tough question, especially when it comes to stuff like Snowflake or Databricks or whatever, where if you're not working in an organization that uses it, it's a little bit harder to get that experience, right? But there are ways that you can learn. Do you have any tips? Well, first of all, I will say that when I started working with Snowflake, I had never used it before, so that was fun, and that's okay. Sometimes you learn things on the job. Most of those things have free plans that you can use to try and build toy projects. I highly recommend trying to do that. There's tons of really good open source blogs that people write on Medium or elsewhere. I find really, even LinkedIn sometimes, I find to be really useful. Those are probably my two biggest suggestions I would use, but yeah, take advantage of that free tier.

Players optimizing for individual stats

Okay, fantastic. I had another one on here that I think we can slip in in our five minutes left, and it's from Neil. It says, as players get more involved in analytics, is there a danger that they will start to optimize the way they play to improve their individual statistics rather than what is best for the team? This is a question that spans not just sports, but everything. If you've ever worked in a call center or managed a call center, you know that if you give people metrics that are time on call, they'll start to optimize for that at the detriment of their call quality. Does that happen in sports analytics too? Are the players self-aware of their stats? That is such a fun question. First of all, I do want to say we have some players that are super interested in data. One of the cool but also sad things about women's soccer is that most of our players will have to have jobs after they retire, and so we have players that have majored in computer science or math or are just interested in data, so that's been really cool to give them fun high-level intros on what we're doing.

I'm not worried about them trying to optimize for stats because we use lots of things in our decision-making processes, to be honest. That is not my big concern. I'm more concerned about them I mean, it's good that data is publicly accessible. I would love them to just use us as the source of truth for data because the reality is that we can contextualize it more, so that's my bigger concern is that they're just trying to get data information without context from other people. That is a great question, and honestly, it's not something I've thought about, so now I'm going to think about it. Now Arielle is going to be thinking about this. Follow Arielle on Blue Sky to find out what comes of this. There was a great anonymous question that says, is there a believe sign anywhere in the locker room?

Not in the locker room, but we did have one last year in our comms office, and I did hit it all the time except the doorway was quite tall. Yeah, it was super cute. We have been thinking about putting one in our house. We are huge Ted Lasso fans over here. Ready for a re-watch. I think we're going to re-watch again over the holidays. Okay, can we fit one more question in in the three minutes that we have left? I think that maybe we can. Let's see if I can find a short question.

Communicating analytics to stakeholders

I wanted to ask a career advice one, so I think this one works. It's anonymous, and it says, what's the most important thing you learned about communicating analytics to stakeholders? Oh, that's a good question. Players, coaches. Yeah, well I think, to me, the big thing is meeting people where they're at, right? Like, I spent the whole first six months of my job, or even more, building dashboards and reports with, and you know, I would talk to them and be like, do you understand this? And they would be like, yeah, I understand, but it turns out, like, I wasn't meeting them where they were at. I was looking at things with the total data lens, and I don't know if they were just too polite to tell me or what, but once I found out that I had to use the video to contextualize what I was doing, they opened up this, like, communication that we didn't have before. So yeah, meeting people where they're at and just working with them slowly. Eventually, hopefully, they feel confident enough to tell you when things don't make sense to them, and I make it really clear to them, like, that I am a thought partner. Like, I'm not here to be, like, the, like, nerd in the corner that's telling them what to do. Like, I want everything we do to be really collaborative and that we build together, because they also have expertise, you know? I learn a lot from them just as they learn from me.

I am a thought partner. Like, I want everything we do to be really collaborative and that we build together, because they also have expertise, you know? I learn a lot from them just as they learn from me.

I love that. That's fantastic advice, and I second and third it. I think that making friends with the people that you are serving your insights to is really important. Yeah, just so that they understand that you're a trusted person, because when you say, does that make sense? Did you understand that? That can feel like a challenge to their intelligence, right, where they're like, oh, now I can't say no, because I'm going to feel dumb. But if you are their friend first, and you let them know that you're a trusted person, yeah, then they can be like, no, Arielle, that doesn't make sense. Quit using soccer terms. I don't understand what you're talking about. You know, I think that's fantastic. Thank you so much for being here, Arielle. This was really fun. I think everybody in the chat had an amazing time, considering how absolutely wild it was going on in the chat. I hope you had a good time. This was so much fun. I really loved it. Yay. Okay. Come hang out with us in the Discord server. We have a Data Science Hangout channel. Everybody come say thank you to Arielle. You can tag her in there and give her all the praise. And also next week, come hang out with us because we are being joined by Ryan Tempe. And Ryan Tempe, if you do not know Ryan, he works at Lego. He gave hilarious conference talks about his silly projects that he did to learn, like the Golden Girls drinking game and all kinds of other hilarious stuff. He is a hoot. Come hang out with us next week. Thank you, everybody. You can save the chat with the three dots in the top right if you really, really want to. And I will see you next week and on the Discord. Bye.