6. Manhattans and Moneyball to Kabaddi: How analytics evolves with sports - podcast episode cover

6. Manhattans and Moneyball to Kabaddi: How analytics evolves with sports

Jul 26, 20211 hr 8 minSeason 1Ep. 6
--:--
--:--
Download Metacast podcast app
Listen to this episode in Metacast mobile app
Don't just listen to podcasts. Learn from them with transcripts, summaries, and chapters for every episode. Skim, search, and bookmark insights. Learn more

Episode description

For a lot of people, their first introduction to data and analytics happens through sport. Fans have tracked batting and bowling averages for many decades now. In the 1990s, with the coming of satellite TV in India, cricket fans had their first brush with bar graphs and line graphs, with “manhattans” and “worms” respectively.

In the last two decades, following the publication of Michael Lewis’s Moneyball, the field of sports analytics hsa exploded. A couple of months before this podcast was released, it was revealed that footballer Kevin De Bruyne had hired a sports analytics firm in order to better negotiate his contract with Manchester City. And along the way, analytics has entered smaller sports such as kabaddi and volleyball.

Today’s conversation is a double header, featuring the husband-wife duo of Arvind Sivdas and Dhanya P, who are also founders of KabaddiAdda, a Kabaddi platform. They have worked in analytics in cricket, badminton, volleyball and kabaddi, among other sports. We talk about the evolution of sports analytics, how to quantify “continuous sports”, the role of fantasy sport and several other things.

Show Notes

00:02:50 - How they got into sports analytics

00:13:00 -  The popularity of “matchups” in sports nowadays

00:20:00 - How Roger Federer used analytics to transform his game

00:25:00 - Why performance analytics has limited impact in (association) football

00:30:30 - The importance of buy-in from the management, and evaluating success

00:32:45 - How Kabaddi has evolved in the last few years

00:37:15 - The parallels between Kabaddi and Basketball

00:46:00 - Analytics in Kabaddi

00:47:00 - Data collection for sports like Kabaddi

00:49:40 - Biomechanics studies in Kabaddi

00:51:30 - How to fund analytics in smaller sports?

00:55:10 - The role of betting and fantasy in developing analytics

00:59:20 - “Moneyball” - where is it being underused, where is it being overused etc.

01:01:00 - Convincing CSK that cricketers peak in their 30s


Pradeep Narwal's 8 point raid

On how CSK won IPL 2018 with "dad's army"

Data Chatter is a podcast on all things data. It is a series of conversations with experts and industry leaders in data, and each week we aim to unpack a different compartment of the "data suitcase".

The podcast is hosted by Karthik Shashidhar. He is a blogger, newspaper columnist, book author and a former data and strategy consultant. Karthik currently heads Analytics and Business Intelligence for Delhivery, one of India’s largest logistics companies. 

You can follow him on twitter at @karthiks, and read his blog at noenthuda.com/blog

Transcript

Intro / Opening

How do you break down these vices in the sport of badminton and that was quite a revelation. It was just a one hour meeting. I think that taught me more about Badminton than seven or eight years of playing the game itself. Right? Don't even I'm Leticia is not an accident. It's not going to deliver returns in the first three weeks is somebody because I didn't, you know, they should work with who can offer that.

I really like, is a long term investment, right parallels, between comedy and basketball, is fascinating. And that way a lot of our analytics and analytical thinking also, Hello and welcome to data chatter the podcast, on all things data. This podcast is a series of conversations with experts and Industry leaders in data. And each week. We aim to unpack a different compartment of the data suitcase. I am your host. Got the chassis that I'm a blogger. You. Economist book author, and a

former data Etc consultant. I currently head analytics and business intelligence for delivery. One of India's largest logistics. Companies. You can follow me on Twitter at Karthik s that is Kar thi KS and read my blog at no Intruder.com. That is n 0e. N th you be a.com or opinions expressed in this podcast, belong to be and my podcast Pious and it do not reflect the views of any organizations. We might be Associated. Nothing discussing.

This podcast, will be taken as Financial or legal advice. One of the first encounters that a lot of normal people have with data, and analytics, was Michael Lewis book, Moneyball, which was later made into a movie. Starring Brad Pitt. Moneyball spoke about how the Oakland Athletics used, analytics to punch above their weight. And in the last 20 years analytics has taken out pretty much every sport in football. We have new metrics such as expected goals or XG in IPL teams.

Talk about the Moneyball strategy as they built their squads every season and it's not just the big games like football or if we get that, I've been taken over by analytics, even the relatively smaller games, such as badminton volleyball, and covered the, have a fair amount of analytics going into It in today's episode, we have van Erp and I've been served us founders of kabaddi at De. We will be talking about sports analytics in general and sports analytics in India in

particular. How did it all?

How they got into sports analytics

Start, how did you guys get started in data and the in sports analytics? So I think we'll start with that and then we can go on to other things. So, see my previous startup was an e-commerce startup and I know as we were we were trying to build a B2B e-commerce. And as you are building it, you know, we started meet. We're trying to build a distributed where also we are like seven miles across the country and we want to run it

with 60 70 people. So I know I restarted and Inez been doing datebook and data for ages, right? Even before that after PhD. She's been that way. So I reached out and asked her to help us solve the problem. And you know, what time we kind of brought down our working capital to just three days that was pretty exciting because it wasn't easy to solve and we had a solution and then we were able

to implement any kept going. And as I stepped out from the e-commerce company, we both kind of thought, you know, let's do something together. So analytics was a Pint. At that time, I think we were doing more of analytics for the startups right here, fmc's Pace, or Telecom space, and it just so happens that I play badminton and be a giver closely following sinus career. Yeah, and she lost silent even scarier. And she lost three matches in a

row and in spite of being world. Number one, at that time, right? So we were just wondering, I mean, candy helper some way. To resolve. I mean, what is it that causing her to lose these matches? Right? And she lost all three games to this player called tyrosine was a young upcoming player from. Not from China, from Chinese typing. Okay, everyone basically, so, yeah, your end. He ties up to used to beat Saina every single time when China was right at the peak of a game.

And at that point there was this This would have been in 2015. Okay. Okay. Yep. Interestingly sign. I had moved to Bangalore. She was training with in the prakash Padukone and Academy. So can we do to we welcome our search who's India's first Olympian and badminton? Yep. Is India's national clothes and you know, she was also coaching sign up back then. Right? So we went reached out to him and said sir. We do all this work in analytics. Is there a way we can help you

work with you to help Saina? And then it will maybe make it more meaningful. Young day. Those interactions with the muster and Professor were were Priceless. They were. I think, the reason I got into sports analytics is because the introduction was exciting. Yeah, I mean, the clarity, especially when it comes to prakash Padukone, right? We talked about all this analytics, you collect a lot of data. And then we met prakash after we had some insights about sign has game.

Yep. You just took out a sheet of paper and Drew the badminton court. Oh, and he said, there are these seven positions on the court. Two in the front, one in the middle, two at the back and basically a 3 in the middle 2 3, 2 2 3 2, right and basically a player has to hit from each of these seven positions to all the other seven positions on the court in a is free manner, but it just so happens that every player has a bias, right?

If shuttle goes to the front. I have a bias may be to hit it to the backhand side of the opposite opponent, right? Yep, and he said, badminton or analytics kind of comes from this right now. Do you break down these biases in the sport of badminton and that was quite a revelation. It was just a one hour meeting. I think that taught me more about Badminton than seven or eight years of playing the game itself, right? And five months of politics

before that. Yep. So how does so, how do you how do you Will collect data for this because I imagine that, if you have to look at seven positions of the code means that, first of all, you need to actually watch the game or do you need to have somebody watching the game? Most of most Indian badminton? And I guess is not telecast. So, so how do you even Catholic? And I am a pretty confident that at the time you guys started.

There was no badminton and analytics going on, at least in India. I'm I'm getting. So, so how do you get the collection as Always been, you know, as we did and it is became relation data collections, the core problem to be solved. Yep, every sport including Cricket for that matter. So we we have always invested very early when we get into a sport to build the data collection tool of our own or look at what are the data collection tools out there and try and see if anything can be

repurposed. Right? So the first thing we did is built a data collection tool in today's language and quality, and he will thinking tool broadly does a tool that would I love so much and I love person who understands the sport to collect one piece of data. Yeah, and the tools that will give this is the most likely

return shot. So he connects lesser data on every data collection Excel, but it is a data collection, exercise, a fairly intense and involved data collection exercise, the power of the tool, just minimize this error that the agent has and data collection. I would. Okay, so they actually had people going and sitting I've certainly got almost an order in the lucky cases watching someone TV and sort of tagging every shot. It was it from police.

In a you proceed from position. Number one or two position number seven on that side, and then they return came to position. Number three. And is that how you you guys actually did it? So in as we are doing this initially for Saina Nehwal, it was easier to get it off of YouTube. Read all the key mindset of all the big games at Sinai spreading in size. We were online and all of our big He's against upcoming opponent were all online.

Yep. So, we were able to get a lot of these matches on YouTuber, or in different platforms. And then, you know, viewer, then collect the data. Yep, and delicate quality was good because I know was definitely India's best player. So, of course, we almost anything it was. Well, I was well, documented and prose their online, right? Yes. So it makes Turtles possible to do it.

So the so the investment was about watching it from YouTube and like you said, you know, the That election was a little more detailed and that's one is the positions. Yeah, the other thing that precursor and women served. It has a kind of broke down. The sport to saying, when you pick a shuttle there are, you know, you can obviously call it a Continuum of places where you on the y-axis on the vertical plane, when you can pick the shuttle, but he's a little, at

least, three big pieces. Is it above the head? In this case? The player has reached the shot Valentine. Yes, it between the shoulder and waist. In this case. The player has just reached a And if a begin below the waist it means of restored really late. Yep. So it's a 3D model. So here's the good thing was like the me a separate precursor and criminals are able to break it down into a way that

analytics. And they were able to understand and analytics, on continuous data sets is not reasonable. Yes, if you into analyzable units. They can they can advertise the value. They brought was breaking the code into the seven important places having the short in the three places because then you still need to watch the video. But then what it allows is, it allows them to look at a specific inside and then was all relevant videos for that Insight.

Yep, and that you get. This is really the error happening, right? Yep. Yep. This is our first kind of Entry itself into performance analytics. There was another data will be collected. Today, which came purely out of what women had noticed. He said that there is a thing going on between the Chinese players that sinai's impatient. Okay. So I always serving is the wait for a good, three, four five seconds before they actually get into position and they wait for

5 seconds before they serve. Okay, and that file s, it kind of puts Ina and Restless place, right? Because she's waiting for the shuttle to come. It's still not come. Just breaking this thing. So they used to do that. And then because he said that we started looking at them that date also and we realize sign, I was reading 4.8 seconds or of the order of one set in before she serves, on the other hand, other players were happily taking three to seven seconds while they sort of to her.

Yeah. There's no rules. The difference is start when they serve to cyanide is 327 when they play anyone else. It is between 1 2 3. Okay. It is not in nature the They always take time. Yeah, right. It was very conscious attempt on part of opponents to unsettle her game at all levels, right? And it took this observation for us to even look for their data. It is not data. You will collect as an analyst ever happened.

So I think one of the things, I think a lot of a lot of people in analytics, I think the mistake they make is like there is they'll just they'll think that any data can be unless in a sort of a domain, 3 manner. I mean, we all have this, we all cover come with the sort of arrogance.

I have all the data, you don't need to tell me anything about the business, but I think, unless I mean business in your cases sport, but unless there is some fundamental Insight or some sort of starting point that the hypothesis that you start off with from the business. Think it's pretty much impossible to tell if look for that. I mean, if you might be lucky, sometimes the data might throw up something neat, in terms of like need patterns and so on.

But in most cases, like, sort of, you need to look for the, you need to wait for the, you need some insight from the business, in which case you go. Got it here from from percussion rebuilding the, the initial sort of the insights, which help you break down the court into the into seven things. It's one other thing that it reminds me.

So what, what what's interesting is that, like the Chinese players as you mentioned, like they were taking longer to serve to sign up at like, taking a shorter amount of time to serve

The popularity of "matchups" in sports nowadays

to each other. Right? So what this reminds me of like this year in the IP and the little bit of it that happen. I watched it all. Exclusively on The Dugout channel in hot star and they're like, pretty much every 2 3 overs people were talking about matchups. They were like Burak:.

I mean, it's sort of well, known that we're actually is weak against leg spin like that and then like and so people would teams would bring in Alec spinner while playing CB or for example, like, they would bring in like you though, going back. A few years. People would play Left Hand Spinners against people like such In scientific or Kevin pietersen because they were known to be weak against that. So, so I guess this matchup sting is not a cricket specific

thing. I guess. It's Universal to All Sport and guessing. Yeah, matchups is a universal thing. But, you know, one of our observations in, in specifically in cricket, right? Is that there is an ebb and flow to a cricket game, right? And you know this well, we have seen all the work you've done in cricket. Very exciting stuff and the append flows every game. Is unique in itself. Yep, and and it is very rarely a standard answer that. Fix it.

All right, when you're playing against cska, you can definitely open the bowling with the left armor and Watson was batting. Yep. With the Phantom, you if you keep holding back a spinner for the barn and do need to come out to bat. You would have lost out so much of the event from the match and maybe Watson and fast. Have just gone at it. Yeah, right. And and that, that is what makes Cricket so exciting.

Because there is a function of not just matchups, but which wind dryness humidity grip, right? And all of those factors are really, you know, if you want to be honest about entities, you have to quantify it of that, a military context of all of you have almost no data to make a matchup meaningful. Exactly. Yeah. The number of combinations. Is this too much. It just was great. And that's what Cricket is such a week.

Beginning and end. You also notice that the role of the captain and T20 is priceless because these are decisions are making after you see 10 balls of the game. Yep, and those, I think those decisions are it's a very different Dynamic decision-making entity itself right here. Versus for example, baseball, which is often a common comparator.

The biggest issue. The baseball doesn't have a pitch component, which takes away one, Big Driver of differentiation that it's only the end wins that you need to correct for. Yep. Yep. Yep. The picture actually doesn't pitch the ball. So correct. Exactly. And I did again, like the other thing I guess is that like, in baseball and MLB has 162 games per team per season. Using it with post. So it's like the amount of data

is simply not comparable. So a, you could have some noise, which can kind of make end up making, you think you have some deep Insight on some player? And so on 100% 100%. I think it is something that we need to often. Correct for, right? So when we look at so we have a metric is very similar to a lot of that. You have to predict. What is a win like to you after every ball, right? And what happens in that particular ball is in assigned to the pattern of the Border

accordingly. Yeah, makes sense to use that whole most cash for outrageous jumps. So, for example, sometimes you take two wickets in an oval, right? That's a my changing exercise in itself. You know, what is the border is done that? Just once in his career, you know, you have to give it that way. They said you can assume that he's always a match winner.

Yeah, of course. Yeah. Yeah. And and but what he would have done put 10 because any do aggregated into look very Exciting because he's had 14 ticket Hall. Yeah, and the season. Something looks like it looks very different. But if you look at WP is actually done it. In one game is added. You fantastically given a win in a game, but then, it has just been a setup, you know, is not lost two games.

Yep. The drawing that difference in data is very tricky because like you said is 14 games and and The Fringe players get to play 34 games, if they are lucky. So Cricket has a challenge because depth of data is very very limited compared to MLB. Oh, yeah, so I - Cricket itself as a challenge with respect to MLB. What's it like in like this is sports like badminton volleyball covered. Eh, it's a the same thing, you know? Yesterday.

We were reading this article on the Olympic site where they were where sine squared, which Carolina's Martinez, given an interview. She has been collecting data for 15 years of all her matches. And now, Fifteen years. How old is he? She's about 20 cheaply the lead with his 2007. He won the gold in 2014. So I guess she must convince not that young as well. So p60 they didn't do anything serious, 2016. Yeah. Yeah, that's when we started

working with sign erect. It was in the run-up to that obesity and and even doing this for 15 years, right? And they have 15 years worth of data bank and update that. She said she said my biggest Different issues that I have these 15 years worth of data on my opponents and on myself. Okay, and and honestly, that is the investment. You need to make in a sport. If you want to really stand up a right Djokovic, did this in tennis a few years ago when we were following ten is very closely.

Okay, he bought out the company that does 10 is analytic and make them work with him exclusively. Okay. Okay. And to biginsights, right? One was he? The job Djokovic was world number three for the longest time, if you remember, right? Yes, he is. And he's 49 percent of the shots. He played. So it's not like he was, he was doing exceptionally. Well, yep, and then, he moved ordinary person up to 51. Yeah, and I'm from World. Number 32 such an Undisputed world.

Number one. It is fascinating, right? And then he will start, he basically needed to break down the games of his two major opponents. Nadal and Federer, you have to Tote The Note, right? And yeah, of course. And he he always believed a model of the breaking down Federer's game, right? Where they resist, it is Stark. So he used to hit more than 50% of the boys to Federer's backhand. Okay, and he kept doing it even if he was in a bad position.

Even if Federer was expecting, but he still kept going and his model of hitting at the back and through the game because he said over a match. If I do that long enough. I'm going to get enough errors, so He started playing the percentage game there, although he moved is more aggressive. I guess. But on the contrary there are

How Roger Federer used analytics to transform his game

very few players who are able to apply the insights from analytics and transform their game, you know, if we look at even the but by that time that would keep that is according to us. The main reason that Nadal Federer and Djokovic have dominated intense generation and a dominating the Next Generation. Also, right, we're evolving their game, which is just, I mean, it is not reasonable to win the number of B is a three of them everyone here it is. Just not reasonable.

Yeah, especially when you're playing in the same generation rate. Someone should be with you. We also saw this. A lot is federal, right? He had a terrible Year in 2012 that he was tried in the peak of his game was winning titles and then he went through a year and a half its own retail. Single finals, obviously capable in a dollar. Federal is not like he kept losing the everyone understand. Why branca won three titles in

that period also, right? And then so Frederic came back and Transformers game just to be in a game of shorter rallies. He started sacrificing points. To go for shorter rallies and he basically started cutting into the net. So he realized that people are going people as an Italian joke, which are going at his backhand through an entire Minds. Yeah. So he's only way to counter that is to either walk into the net or play even more aggressive.

In the first three, four shots in the denominator. We have read that, he tuned screaming also to being that he stopped doing any long rally trainings. He only did for some trainings is really even care. Obviously. He had the game to completely change training schedule for this happen. If the dings I'm leading with the pinch of the player being able to execute, right? He always had the belief that he can pay the rest of the Murray. Just played for Scott rallies in

all his practice. So, it changes practice partners. He changed his approach to the game. And now we're at 3940, if it wasn't for the locker, I can email know the last two years would have looked the semifinals and finals. Now, we lose a few even win a few, but it's just a percentage is game way that he gets that obviously Djokovic is at this point. Two years ago, pointed probably the peak of his career and he's held up in for two. Long, however, however, you slice and dice the data.

They are all standout players. The thing is what you thought of you were there. With this example of Critter is that like analytics, in some sense is a bit of herb competitive game, right? Like you figure out some weakness in your opponent. And if you are playing each other regularly, the opponent then figures out how to counter what you found, assuming that they can, they can follow through on the idea that they got and so on.

And then it's a sort of a back and forth from there in terms of like how to sort of how Oh my God, what do you say? How to outsmart each other in and so on terok, nor spot on? Because I see I would think about daytime's portrayed in a few Dimensions where we are right now, only talking about performance analytics till now the conversation, right? Yes. Yes, performance analytics.

With this is probably the only thing where you are looking for an inside yourself and often inside will take a couple of years to combine, and then another couple of years to really Implement. Yep. And then why? By the time you implement in a year of other people has to go to do. They are all starting to come to it. Yep. So, you already have started looking for her next inside you needed or started training for

an extension. So, that is where analytics kind of helps push the game forward when, you know, as you can imagine has you as a low-hanging fruit fruits, get discovered. The next level of insight is much longer to discover a much much longer to implement date. Yep. Play people think of Spain in the World Cup right? Where they played the style of short passes. Yeah, once it worked, the way that it got adopt adapted very quickly or countered, very quickly, was also interesting to

watch, right? The style did work in the next World Cup itself. Very next World Cup with drop-down spectacularly in the defense of solder is too short passing game. I mean, while Spain won those three titles 2008-10 912, the two euros, and the World Cup. And then like they have not been a force on the worldsheet of it. But somebody like same Pep Guardiola with Manchester city has now won three out of four titles. So in the last four years by slightly adapting is here.

But again, I guess Club football you play at a much higher frequency than than its Russian Football. So it's easier to sort of adapt in the, in club football. I guess. Compared to International. Seaman human trait of having lot of data on football or say we have a slightly different point of view analytics, right?

Okay, football is a very continuous game and and I almost believe that, you know, the largest value analytics can bring into football, is selecting the right squad for the opposition in front of you and selected the right Squadron and postseason. You know, what? Because in season In a match, if you to break those who take analytics right down to the match, right? Given that this is it you have a only value analytics release to

Why performance analytics has limited impact in (association) football

pre-plan for, who is our opponent. What are they typically doing then? What can you do to counter that? After that? What happens on the ground? That is how good is your execution versus a plan? It also requires at least, 45 people to come together for any part of the plan to work. Yep, right. So often deviation from plan also is not a function one. Person and the analytics is honestly the ability of analytics to take five moving

bodies and take a plan. Measure how these five movie bodies are doing against the plan is impossible, especially when you don't even know what the plan is. It has an analyst unless you're a journalist sitting right in setting, this was the plan and these are the big deviations from the plan. So in football, I feel like all of the thinking unexpected goal. It's a lot of Statistics, but Well, I believe that the impact of all of this on actual performance is almost negligible, right?

Analytics is really a planning tool in football, but I do not believe anything is a fairly controversial. Belief, in today's world, where I think the the value of analytics and football is really to help help a coach with planning or the manager planning for an upcoming opposition and the bigger package was really, how do you build a meaningful score, which you, which can deliver two?

Ian's having input voltage, especially like I see all these like sort of. I hate those these spider charts that keep getting proper put up. I mean, recently be athletic as recruited one. Analytics writer is their football team and like this full of fur spider chassis, like okay wind. I got injured would you replace them with? Look at all the spider charge of all the other Defenders all over Europe and say like, okay, maybe we could go with this guy that again.

So it's a bit of a sort of and XD itself is a bit of a In my opinion itself reasonably use this metric because it just dip. It's just a function of where you shoot from. It's a function of where you should try. Mine. It doesn't. You don't it doesn't matter whether you have an open goal, whether there's a full wall in front of you, whatever it doesn't matter. It's just a function of x and y coordinate in front of the goal. What's the likelihood?

You are going to shoot from. There is the herb, in my opinion. It's a meaningless metric. Yeah, you know, we we all agree, we completely concur on on that. We think also is listen to be for us to work in football. There. Only choices are Indian clubs. Yeah, it's let's be there is not at the Pinnacle of the sport, right? Yep, and I guess it, what we've observed in India. Also, is that working with with franchisees and teams are not requires, a lot of buy-in at the

other end of the bar. One of the content of what you're sharing needs to at least be Even water sharing, they need to invest their time to grow the asset even angle. It is not an acid that it's not going to deliver returns in the first three weeks. If somebody believes that then, you know, they should work with who can offer that. I believe. I need is a long-term investment, right? You have to go to the process. Has to kind of be done and the outcome will slowly start

falling in place. Especially in something like sport because you need to, also your whole system needs to align to work with the data you're getting, or with the principles of using data in decision. Making right. The esight probably has to come from the other side, right? From the management, from the coach. You just give the tools and when the internalized it, that's when there is some connection to implementation, finally, on the

field. Yeah. I need the think loaded Sports, Sportsmen and sportswomen. Experts in their field in their own right and to expect them to kind of apply analytics which is not their core competency. Right. Is some unfair to do. Yeah, I guess I'd also like a I guess in some sense some sense, a lot of people the stuff that they do is like this is do things because it's sexy because it's like sort of it's a cool

girl cool thing. Okay Liverpool got a PhD in physics to head their Analytics. See so Manchester United with also higher, if PhD in physics to hit their analytical, something like that. I would think what analytics right is for the gentlemen take for big Dimensions, right? There. Is that what we have open till now is really performance analytics. There's a lot of Indonesian Squad selection when there's a lot of the count, there's a lot of analytics, which is relevant

for fan engagement, right? Yep. There is this whole layer which is what you and we have done for a living, which is an analytics for the, for running the business. Yep, and I've been running the sports business. Also, you need to look at a time. It is not just a single fit. All right, and and often any talk of sports analytics, right?

Everyone focuses very, very quickly on performance analytics and why I believe the account of what we vote was saying that outcome and performance analytics is dependent on so many factors. Beyond just the analysts ability. Yep. It is dependent on the ability of the athlete to execute

command. It is the ability of the athlete coach and the analyst to come up with an Eatable plan that means analysts needs to not just do. The analytic were also converted into into language that the coach and athlete can buy into. And to be honest you and we should know this, right? Converting an observer and insight into and into something that the athlete or the course can execute is not an easy

The importance of buy-in from the management, and evaluating success

switch. Absolutely. Absolutely, especially in sports. I guess the coaches usually not somebody who is like, even in general very well versed with Data and things like that. Exactly. So they have to, they have to understand the process, right? Yep. Take for us working with Chennai, Super Kings has really been exciting because, you know, the management and the coaching staff have made the effort to understand what we are doing.

Yep. They don't just the understand what we are able to and what we are not able to. So sometimes there's no expectations of. Obviously, we are doing a lot of predictor work with them. And, you know, when you have a predictive, when you do predictive of you, say there's a Like business 35 percent likelihood that this will happen and then they come back and Signal, will it happen or not happen? And I googled, you know, there's a 35% chance.

It will happen. Yeah. Yeah, and then how we measure success of predictions, right? Thank you. The league at least has 14 games. So we can say, we predicted when likelihood for a team to be 57 percent over 14 games and, you know, over 14 games. Are we, are we on track that attics times? We have been working with Chennai, Super Kings 44, Yours now, right?

And and obviously it is a it is an investment on both sides to to understand the other side and honesty, try and deliver value in that period to each other, right? Yeah. Do you think it's a very continuous continuous process that needs to keep happening? Yep. Okay, let's sub slightly change tack again. I mean, I have a brother. So we, I think I was very excited by how you broke down the badminton court into your 7 into 3, grid and so on.

So, how do you buy? I'm assuming you guys have done similar work in, like cupboard be volleyball Etc as well. Because, like there again, I'm guessing there is a sort of a, I mean, if you think about something like football, which I watch a lot of it's like, usually you talk about 4, 3 3 4 2, 3 1, and so on which is the main we will.

Formations in zone. So I'm pretty sure there's something similar that is there, which I have not watched too much of it, but in sports, like, company, volleyball Exeter. So, how do you kind of how do you go about analyzing a new game? New sport that you sort of get into and Etc? See, couple D.

How Kabaddi has evolved in the last few years

Obviously, we are is very close to us. Right? We have, we have not lie. We started doing comedy analytics. We have now built a couple D business itself. You've commandeered L is aiming to kind of bring all the information of cricket of all the information of covered being together, you know, we're trying to connect the comedy fan to the play. Are we doing a lot of that? So obviously analytics for the fan is very is integral to the cupboard, get the story right? Yep.

Help me the cookie fan, just look at the sport, like the cricket fan looks at sport. Right? And that is that, that something dear to us. But, you know, coming to Cabo Verde, we haven't recovered. T is really nice. Isn't all said, and done, right. All credit to the league with brought the sport to the Forefront. That's when analytics even came close to the sport. We have actually this season on season here on here as a sport, is the world.

The way the sport is being played itself is getting very different. So at this point analytics that I did two years ago is almost completely redundant today. In the way. We started the way we break the players away. We analyze players we Rank and build a squads all very different now, because when we started when cavity probably started the something we started observing. Professionally, what the prevalence of all around us was quite significant ready.

For example, can you break it down for our listeners, many of whom I have played comedy, when I was a kid, but I think many of our listeners may not have played it. So, can you just break this down into the? You just make some noise on the sport? Yeah, maybe that will help. Yeah. It's a maybe 30 seconds. One team will be ready. The other team will defend. Yep. There's always one Raider who's reading against a defense which starts with People every time but he loses a point, a player

leaves the coat. Yes, so if you think about it, so you're not really getting a point. You're also losing the opponent, also losing a player. So you would think that the reading the team that is gaining points. Gains in power very, very quickly. Yes, but what really happens in Cabo Verde is when your, when your defenses down to four and five people, it's almost impossible for the data to get points. It is really tough tough

balance. So in some sense, when you lose those first two point, you're actually make your Defense much better. Okay, so that man's then, pulls it back in favor of the team. That is down a couple of points. And then we go down to three players If you defend you get two points instead of the typical one point. Very, very Priceless. Is that a PK lonely or is that in all be about, 220 root even defenses down with three people, you get two points. If you get a defending point,

you have to make it a one-point. Yeah, and I guess when you get a point, you can also bring back one player onto the field, right? But you bring backup clear. Not equal to point players. Equal to number of players going on that. Many players come in. Yep. Yep. Yep. So in defending the three people you get two points, but one pair comes back in. So you both to for third offense again dangerous. He walked 25 your defense again. Dangerous.

Yeah, right. And so through this transition from disorder since they were players who could you know, as a league started never players who could defend and Ray? Yeah Ray Dozer, just barely ever even had multi-skilled. Eels, and when the sport is being played on mud, you know, the defense is much better than reading grade has really struggled in the game environment because it's slow. Yeah. So the trader speed is slower. Whereas in the game is move to Madison become much faster.

So Defenders struggle, much more open areas are a lot of readers are able to get points, especially if you have in mind, and then you play on Mac to become much faster. And the other way doesn't work. Right? If you've always played in math and going to Mud becomes not easy for the for a reader. But Defenders, of course it

makes sense. And now what we are using is a sport is about to becoming one of our Specialists also because there are more matches happening, you know, it's very tough for the body to record if you're doing too many things you

need the time. So there is kind of two ways of playing, you know, when there are small tournaments, the 34d tournament you would see both Specialists and all around us being relevant and when you, when you see the law, Longer format tournaments, whether a lot of matches happening over to 3 months, then people start saying, you know, start noticing special to someone important. I play one role in the in that

team. Yeah, and you would even look, which would we have noticed the parallels between poverty and

The parallels between Kabaddi and Basketball

basketball is fascinating. Okay, and that would be the Lord of our analytics and analytical thinking also and storytelling. And so please, oh, please. Oh, please tell us about that. Because I think this is a very Fascinating analogy. I mean apart from you have like in one you have five players in one you have between three and seven players. So apart from that click were the how are they similar? So, I mean you can think of basketball.

Also, as a Quantum sport, right? Just like a birdie cover T. Every 30 seconds, some event happens when a radio goes to the position. Same thing happens in basketball team because actually, we have 30 seconds to actually shoot their basket and learning and another Parable, when it especially comes from, what is it? This strong link weak link. Sort of analysis, right? Yep. So basic In basketball, your best shooter can take practically every shot that your

team takes into the basket. Right? So we're obviously improving your probability of winning. Where is similarly? Similarly, poverty, your reader can go for alterations. Your team has to go for. So, thank you. When you getting a team, you get one Superstar readers, a smart thing to do because he can actually go for 40 of out of the 45 raids. If he has the Well, make sense. On the other hand. It's more like football, right? It's a clean sport. Do you want one the weakness in your defense?

Because that guy is going to get caught more often than not very similar to basketball. Great defense needs to work together. When you can keep hitting the ball to your 1, Min shooter. If you have the ability, obviously in the NBA at the highest level is now become different. That is one big battle is with the evolution of spirit. Quality in be involved. It was considered a black sport as if its early days of a solution, right? It's still dominated by a lot of

athletes of color. For, what is really exciting is that, you know, it keep it in its Evolution. It really came. It was a sport played by, not the rich by the not rich State segment of society. Yes, but the best players of the sport version of the home country of this port, right? Yep, all you can probably create 10 basketball teams, 10 USB. Basketball team, which will win the Olympics. Yes, you mix up your similar to a birdie, right? You can create 10 teams.

Although India lost the Asian Games, gold this time. You can have 10 teams from India, which will be in the top ten or twelve teams. Yeah, with an Iran, in The Courier coming in baby. And which again credit to the last two years of development that has happened, even in these countries that is parallel in how the sport Ebbs and flows, right? Is something you can see, Manifest itself in analytics, all back. She said the strongly how the Raiders, what the reader, that

contributes to your team. There is often very, very it's very tough to quantify because there are there are two impacts supposed a reader who gets a gets one point every rate, right? Yes, you're not taking out the player and that player goes, make your and then you bring the opposition down to five players. At that point. You're not really able to get ready and then the opposition sends a Raider, he gets out. So they get back to 16 to take another pointing. You haven't read.

Better take single poems is very different in value delivery to a player who can take a multi-country. Here's a multi-point, a movie of would opposition of 624 very quickly. And I know that in price, he does that in two games, you are likely to win those games if we can pull it off two or three times in a game.

So yeah do the the multi-point shooter becoming very valuable, ya know like the image I have in my head about cup of tea, maybe some movies and stuff is that The reader goes pretty much six out of the seven guys would have touched him and the rate of which great difficulty comes back and manages to slap The Lion and the the other team gets all oh does it? That's the image of Comedy that you have from Bollywood. Let's say and then like from mastered their most recent

movie. I ever watched any movies in the last 10 years. So I'm talking about like some I wrote on some old Salman Khan movie. I'm thinking of I don't even know which movie but but But he is so in pickle that doesn't happen. I guess, I guess you only get like one or two outs in a raid and so on, once a to say, it's happened, once, Okay, who wants took six points in a rate of Fox Point, 6 Plus, and then the Slick guy and he got six points with the. There were only six players on, Matt.

You got all six on the mat and and because the team were already got two points on a date, so it was neat. Point rate effective bomb. Obviously, the game was over, right? The team won the game. It didn't matter to bigger lead. So this is a short after that, the game just. It was a good game to let point and it just dissipated. Oh, how is company in science? Some say some music, for example, in IPL what we saw this year is that like, there are some games which are which are

like very well in contests. And so on, there are some games are let's say there's some gives a weak King, Solomon, made him 110 120 and you you that there's no point watching the game from the Not that it is pretty much a foregone conclusion. It quickly becomes one side and I see this more in ODI Cricket than a T20 where a game can very quickly become one-sided, while about 10% of the game's called, possibly like, remain exciting, till the end zone. So how's it going, buddy, and

other sports. That you cover that is, you know, you brought the other thing that is very similar between Capaldi and basketball trade, but still has a lot of close games. It's a district Ali, right? And nobody is very similar because the ability to come back is actually Tight, I have seen a few games team has gone up. 10 0. Good happens very quickly. You get to three first user ID, a first 10 points, but once other team regroups way, they can also get back to the same

evening points. And, you know, it it's a game where a couple of Errors can suddenly turn the game on. Yeah. So I have seen games with the teams are one being eight points down with three minutes to go. It should not happen. You have as to error different and you will certainly have a lot of pressure and because it's it's So fast, like worst of all the time for the code, there's no time for courts to step in and, you know, kind of calm the situation if you don't have a timeout left, yes.

Yes. Yes, you're acting under pressure. And you know how you'll react when the president only made more errors, right? Yep, that would be so exciting. There is, you know, there is a very small percentage of games that were actually, you know, where you can predict the result very very quickly into the game, right? When to average teams play, the still high quality match. When two really strong team

spirit, still very hyper. And over the users, like what you seen with, you know, even in the cup of tea Nationals Talent is spreading over. So there is not two or three, who teaches already eight or nine good stating, otherwise, Ariana Railways and services of the three strong teams, even in the tournament. We conducted in the K7, you know, there were 16 teams. We think there were 12 outstanding team went because we were limited.

We couldn't conduct a longer event because of code can the quality of gameplay between teams so high. They were there were three years at that ended with a 1-point difference, right? Okay, click yes to time. And these are all the nude, all these big debates on, who should rank higher because the fact is, you know, using a small sample set of decided ranking team is not fair also, right? Yes. Yes. Yes. Yes. Yes, interesting property. That is what we are seeing.

We are seeing a lot of Close Encounters because quality is generally stepped up, right? That's what we are. Open the K7 and even the senior Nationals on the other hand. If you look at International matches, very, very well deserved. It is putting them in the local matches us quite interesting. Whereas, when you look at it at an international skills. It's still not scaled across the country yet again like basketball. Yeah, exactly. Was the word we use an analogy.

But every day we see it. You know, they just, it is staring at you of is actually, you know, what, the next time I watch basketball. I don't think based on the energy that you give me. I don't think I can unsee it when I see, but I keep wondering why in NBA. First one team attacks, then immediately, another team attacks. And so. And so now if I just think of it as a game of kabaddi, it's really it's just going to be the it's going to make more sense to

Analytics in Kabaddi

me. Yeah, it is. Okay, so being in. So I guess it's covered in terms of analytics and so on, you do things like cup. How does it rate of perform given? How many players are in the picture near the side and so on? And and those are some of the things that you quantify. And I guess you also have like your regions of the cortex things like that. I'm guessing select and happening difference is very exciting to do and latex for

right? Okay, because the defense just not a function of how, you know, we, when we started doing analytics and Cover TV were really visually how each player was doing in the Did you ever hear of? So, and then we started talking to coaches and present in coaches kept saying. No, but this is not how it works. See this way didn't do enough. We just started the tackle. Those four guys who came after are much more important, others is really useless tackle.

Yeah, right. Okay, so and that is almost impossible to quantify. You think? Tell you how you would quantify that, right? Yeah. You guys know if I didn't quantify, how do you collect the data? Like the okay. I see.

Data collection for sports like Kabaddi

Now tell me about how you collect the data for somebody having liquid. Because it's a Continuous within each. I mean each rate is a discrete event. But within that there is a lot of micro events that happen unless it's in badminton with each shot is a discrete event and women is very easy compared to cover D because every shot has a has a life of its own, right? And it has a starting point and an ending point yet. Clear within the rate. There is no starting point and ending point.

There is something that's going on. Their number of humans. I would say on the court is their biggest difference and it also there are eight. People on the code, red. I mean number of Defenders plus one Accord during any rate continues. Yes. Yes. Yes. So because he has very nice scent in covered a great. I would say, I mean, if I have to qualify quantify where your, I think there's a long way for analytics, to grow in comedy still, but we are doing things.

Like, who are The Defenders on code, how many of them, how are they positioned? Well, when people stand, during any radiant defense also, very big impact, right? Some people are very good in some positions. On the mat that some people are very good in the same position when there are lesser people on the map. Yeah, right. Yep. There are these. If I had to name the defensive position just for swear. I don't, I'll explain them.

Yeah, that the do corner positions And Then There are the to cover positions, the two covers play, inside the two corners. What do the four main defensive positions, but the to Raider, it's often the Raiders who stand with the two Corner Defenders is one of the Traders is an all-rounder. We are very valuable Defender because the guy standing next to the corner Defender, right, whether the opposing writers, always looking at the corners.

Yes. So, the person standing inside, the corner can get a lot of good points. If your corner, Israel is standing in the right position. The Raider is trying to avoid the corner and this the corner in as you call. It has the ability to get a lot of good tactic points, you know? Good as in three or four talking points in the game, extremely Priceless. If one different is going into three teams in the game.

Yep. Here's the ability to do that very consistently if he has a skill so specific skill, right? And but, but the minute you become four Defenders, that guy was standing next to the corner is a pointless Defender. You can do very little. Yeah. Yeah. Yeah, sodium skill level player. So that's and then you have to start quantifying. What are the skills, you know, we've tried to break down the skills of each defensive position. So there are two kinds predominant skill for each.

Defensive position is slightly different right for the corner. The predominant skills would be to put in an ankle hold to time the start. Yeah.

Biomechanics studies in Kabaddi

In this context, which he just said now, right? We did a small biomechanics sort of study. Okay, so there is the call for deep novel and we're deep level of 8 points cream. Yes. He was. He was a huge Nemesis for all the teams. So we thought. Okay, let us break down his successful rage and unsuccessful rates. Really? What is the difference between the two from the aspects of his positioning? And when the defense tackles, right? Yeah. And the thing which we came With.

It seems obvious in hindsight, is that every time his center of gravity, slightly outside his feet, right? Yeah, then it's the dash comes from the Defenders. It's a, it's a success for the Defenders. Okay? Okay, but it's the timing of the difference is one second before that when he is, you know, comfortably stable. Then he takes away all of them. Within when you do go to the code.

I mean, it is a very interesting piece of analytics, but I don't know if but Deep can use it or the Defenders can use how you would connect the data for this. Because I think it's almost like you are. How are you? I mean, it's such a continuous support. If I had only one microsecond one millisecond later. I get it's almost like you have got back to your center of balance and so like you can do that.

Same sort of analysis of very center of gravity is where his body is, how is like angle is. And we looked at a lot of these things. But the thing, which matter was the co g. Wow. Wow, this completely fascinating. It was a one, one and a half month exercises just to give you context. I will these important for

How to fund analytics in smaller sports?

people to recognize, it's not to go with my dad, all your fears. So, I think now, this brings me to a question of a, I mean, obviously like sports, like a NBA there. Of the MLB. Even to some extent. IPL football etcetera, they have money. So they could they can afford to pay for this kind of a analytic and so on if it takes 11 and half, once you get this kind of an inside, I'll be like and if you are paid fairly for it, I can imagine it being like, fairly expensive but I can't.

So, how do you get, how do you find the analytics? I mean like analytics has a long-term project. So how do you really find it and what to what our teams and different sports doing on that so far? I do think we started covered data to make money. Your analytics, if we will recap, we really enjoy doing comedy analytics, right? And there was no way a franchise team, an athlete, anybody in the ecosystem, code, funded,

exercise, until we know. And we think the sport has a way to evolve and we wanted to be part of the journey. So it made sense for us to kind of, at least in comedy to be part of the entire growth story. Right? To help the growth along. So we'll cover T. I think it is very much an investment, a tyrant, and it is really our interest and incentive to do it. It. But in other sports it has to be the funded exercise and you know, you are seeing as you can

imagine. You know, what we did with the city Only rather than you just said is automatable. It can be you know, automatable has it can be moved into video analytics. If you get the right that you you have the video you just need to start cutting it. And you need to First build a tool. You need to train the tool, both our Global exercises painful time consuming but very doable. And then it will start collecting the data for specific readers.

And then obviously include a human, because you have to do error correction. Every time be that gets collected. Correct? Yep. So, we want to get there and we believe poverty will get the probably in a few years right there. In the meantime, once you've discovered the inside, you can you can expect that cameras will start collect recording videos. It said that will let you collect this data so you can set up cameras for data collection, right? They create exactly it.

Exactly. Yeah, we can has cameras for data collection. Okay. Yeah, you need live in poverty. For example, you will need a top view and the job you will it attract all players on the mat. But the fact is how do you track players? When when a tackles happen to those are suddenly eight players who look like one lump in. All of these efforts will will take its own related Zone interest. It will we believe that all of this will come if the fan.

Is tied into the sport, that is a fan, enjoys a sport. The analytics will follow that. And which is probably what both all the is drawn and MLB and we NFL of also new and of course, I will show you the same thing away. Yes. I just got fan Cricket. Had one thing going for a rated always at data over the last 20 years it is they have made data storytelling element. It's not happen today or yesterday, right?

Has happened over 20. Years, very actively, every commentator, every platform using data to tell the story. Yeah, and I think that that is a mindset that we are trying to build in kabaddi and it's, it will be a similarly, long journey probably shorter because, you know, people now and crickets already educated the audience to look at data. Yeah. So a lot of the analytical initially Camuto, Drive value to the fan, but if your ass and do

The role of betting and fantasy in developing analytics

you have your own narrative for the sport? Which brings me to the other thing? You have got you've led me on to my next question. What role does they say? Things like betting wherever it's legal and fantasy wherever betting is not legal. What what role does that play in sort of in sort of developing the game overall and in sort of the developing analytics and and things like that because the p7 tournament, right? We were getting around, say fine. 100 users watching the

match live at any point. So then nine matches nine out of 24 matches. We had dream level putting out a game on their platform. Yeah, right. On the fantasy platform. Immediately the data on our website, for a player called neeraj Kumar spiked up, because people are now looking at stacks of neeraj Kumar and the place even, I don't know who he is, right? Okay. So what is the need for consumption of Statistics data number? For the staff and suddenly we're fantasy.

Come, the equation. I think it creates that need and that's the that's probably where the meat in some sense. We noticed that it brings a fan closer to the sport, right? You are invested in the sport. We're just watching a person. Yeah, and that that that is what the u.s. Always had reduced always at Fantasy 4 years running rate. Yep, please my favorite Fantasy game was a part of sitcom culture sitcoms used to talk about people.

Running to create this Fantasy game in the middle of sitting in the loom. It was an Indian popular sitcom side. Yeah. And so you're talking about fantasy as a part of basic culture, paying your regular Frankie game. We never spoke about betting as part of the regular ones. With the always spoke about creating a fantasy inviting

friends as a couple of episodes. Two and a Half Men has a girlfriend that principle to Modern I'm doing all the long-running sitcoms have fantasy brought up enough times. Yep. So it's a part of basic culture and and like Daniel just said, right, the traction bestowing our own data for games where where there was an ability to put a fantasy team, Andre Miller versus game where you couldn't create a team on finances and driven.

It was still very interesting. Yeah, and a site went down, that is not happen to us in the last, okay.

So what I say is they can normally you have the exam server core audience which are which will watch the game anyways, and so you would have the data from that and they'll be watching and I think once you bring in fantasy and a well-designed fantasy and so on, then that can that will also bring in a sort of a marginal audience, people who are in it just because like I know friends who live in the UK, who bet on dog, Gracie, I mean, I don't think they've ever gone to see a

dominating show. Dog. Crazy, dog race, but they bet on it because it's there on on the betting sites in. Yeah, I guess they just listing one of your events on deemed. Evan means that like, they'll be some interest in that. Will that can in induce organic interesting, cupboard, if you think about it, because people would have watched it just for the heck of betting and then like they like the sporting so on. Yeah, and and, you know, I think see fantasies are level of way

from betting regular scientist. He's is making, you think about a player level and that makes it very different, because you're not right, you're hunting. A macro decision, you're saying, will this one player be interesting today or Windows has been? And because has a fan, you're taking that move about these one or two players being interesting, you're making a call, right? This guy is going to be testing today. So you're suddenly watch. This one is already said that

playing who helped you are not. Yeah, yes. I did the watching the sport. You're not looking the result on the spot. You can we bring that the fangs. He's tied to the story. Come back was an extremely busy like watching that game because X obviously the steam will definitely come back. System doesn't mean he may not come back, but we feel like that connect to the sport grows. I mean, like, I think I had the so when we talk about like sort

"Moneyball" - where is it being underused, where is it being overused etc.

of sports and it against the analytics and stuff. Most people will talk about the word uses Moneyball after the, the Michael Lewis book and movie later on. So and I'm sure you would have gotten questions from everyone. Like what is the scope of my money voluntary K towards the core scope of Moneyball in comedy and so on. So that probably the Android that like, why don't you talk

about? Like, I mean, I guess this is the sort of you spoke about performance analytics versus like, Business analytics Moneyball I guess is more of a macro business analytics kind of a thing where you try to where you use data to sort of build build squads and so on. Right? So, how was that worked out in in different sports? Like, where is it under used? Where is it? / used in your opinion and like, and what? And yet talked about everything

and finally come to the ability. So I think see, we've done a lot of this work in cricket and comedy specifically, right? So in cricket, we think, See, one of the big differences between, you know, the most natural comparison is cricket and baseball, right? The biggest difference between baseball and Cricket is that baseball me? It is it's kind of unconstrained spending spending. So some teams can spend 4X of what other team spent.

So one of the reasons something like money will happen in baseball is because the teams had less money still wanted to be competitive and for them to be competitive. They had to step away from the traditional. Definition or understanding of this port. So if I take the equivalent in cricket, the tradition understanding is, you need young players to build a T20 scorer, right? Yeah. Thankfully in the first season and cska. That is only thing with this pain.

We said that doesn't make any sense, you know, players really

Convincing CSK that cricketers peak in their 30s

peak in their 30s. Yeah, and that that was our, that was our big inside for the season and that is probably why, you know, it would add is one of the places where we got a lot of buying because then we were able to buy players like so. Half Shane Wynn that once auction, we got fast. She wasn't in her words and sing, right? Yes. Yes, but, but to be fair, all three have been matched winners over this period of time, yeah. And and, and similar insights, right?

So, so that is, for example, one big. So money, collinsport really requires you to death first demonstrate, an insight, and then user inside to build your team before others catch up on that, right? Yep. And so, for example, in cricket, that was a Those one specific Insight which allowed us to build a team. Now, the other, for example, very interesting thing that happens between nobody in baseball and covered cricket. Is it Cricket? Everyone's running a limited purse, right?

All the twelve eight teams have the same defined purse. So it is now a function of how much do you spend on whom, you know, cska always retaining a large part of their core, right? Yes. They very rarely try. To bring too much change into their thinking if something works. Why break it? And in the process works, why I try to create a new one. Yeah. It's clearly a mantra of success. All said and done it is it is 10 years, 10 years of data to say no stealing something, right?

Of course they have. Yeah, and, and one of the other things we are beginning to notice itís. If you look at the, if you look at the cricket field, and if you look at, if you look at where the catches, I think, you know where the most teal events happen to have four or five. Dominant positions in feeling, right? So this whole theory of saying, you want fantastic Fielder's across a cricket field is is I think fairly Humber, you know,

it is sad. It's a little like baseball where they, where you want pictures, who have a strong arm and we'll throw fast. You don't you just need a mix of you beneath Regal fielders in baseball. Yep. You don't need several filters. Emily. You need two or three. Good filters on your field, probably, for if you want to be a great feeling, right? This one who thing, right? When we are forming teams for instance, in cricket.

I think while in forming a team, you need to have an idea about what the role of the person is. Yeah, right, like for instance. Do I look at dhoni as just a batsman or just as a batsman and a captain? So that was one question. We kind of came up with Ray. Yeah, and you come to the question. What is the role of a captain and what we realized was that every time? When he is on the field, the bowlers have a win probability 5% more than they have if he's not on the field.

So suddenly, let me stop you here. How did you even get this data? Because cfk the VA almost never played Without zoning. So how did you even get your cheetah? Three matches in 2004 Battle, of course, a few years back problem back spasm. Yeah, exactly. Exactly. See if this is noise or data. Also is not something I can even argue with heavily for or against yet. Yeah, we feel we have collected enough noise out. We have looked at a player by player.

We this something we did over a long period of time that we looked at all players who have played with dhoni. However, they played outside, how have they defend their best? So what do we do? We said, how do they play in cser? Did they play outside? Then in case k? How did they play? When he The renewal is not there in the four games. It's for bullers, then for players who have played with cs. Can Garner what? What have they done on set?

So we have tried to correct for noise when we still think there is a 5% Delta that he brings over to the bowling unit. Wow, right? Okay, and I think in some sense right? It's very tough to buy something like that. And if it is true, as you can understand, I mean, obviously the system management. Peter seen it yet when there is a doesn't exist right now. And if I have to quantify and you add 5% performance Improvement for a player, when you put them in the squad, you

can use what happens. Why it has neither. Why? The thing looks pretty I can imagine playing Tony, you know standing like Gail is doing it. Kind of brings you into one more thing right? Which we did in the course of this Moneyball, analytic sort of thing. Yes. We looked at just 25 metrics in cricket. Which might lead to a, you know, win in the end for instance to your top. Best Batman scoring or be the Matrix. We were looking at were trying

to address. What is the likelihood a team can qualify for a season exactly, right? Yeah. We looked at these 2030 Matrix, right? How much is your top order? Batsman, scoring? What is your death? Batsman, scoring? When is the first week it taken and the most? Don't metric which came out of all. This is again, a weak Link in your boning. So what's teams their match because of the worst 40 was of the match.

So I don't irate. Yeah, if I don't prioritize getting a footballer and I do not know how to imagine, what will happen in the Dead Two Worlds, right? If I don't have a problem that just connects back to Tony adding value to the bowlers, right? So we have is able to use them correctly at the right times. Sister and it was a standard metric. Right? Your worst four overs. Just the worst Horrors in every match. Can if you just tell me what the worse for was for both here.

I can tell you which team won the game about 90% of the time. That's how I couldn't admit. I mean, that in hindsight, it sounds, it sounds quite obvious, but obviously, to reach it, more forward-looking was quite tricky. Thank you for listening to data shatter. If you like this show, please leave a comment, share and subscribe to the podcast. You can find this podcast on Apple podcasts Spotify. Wherever else? You go to get your podcasts. Once again, this is Karthik

signing off. Thank you.

Transcript source: Provided by creator in RSS feed: download file
For the best experience, listen in Metacast app for iOS or Android