From Football Manager to FMDB Pro: Behind Sports Interactive's new scouting tool

the football co business podcast the most creative minds and innovative thinkers in football hello and welcome to the football co business podcast the show where we get to spend time with us innovating across football media and marketing on this show we're talking about fm db pro a new scouting tool from sports interactive the developers behind the football manager video game series which takes its lead from the famous database and scouting system fans are used to in the fm games we'll be talking about how fm has evolved from a video game to a tool for professional football and how scouting works across the game and fm db pro i'm joined by richard trafford director of business development and tom elliott senior research coordinator for sports interactive guys welcome to the show thank you paul thank you that was a bit of a brief intro rich what's what is kind of the elevator pitch when people ask you what is fm db pro what do you tell them it's changed over time but right now we would we'd like to define ourselves as the the starting points of recruitment so that the long listing phase there's there's a lot of tools and data sources in the market we would like to position ourselves at the front because we've got the widest net in terms of player numbers so you would use the fm db at the very first point of recruitment or the definitive point we're pretending to call it whereas when you begin to search for whether it's a right back or right winger our database is the place you go first so we're going to start with the fm db pro and we're going to start with the first and then the the other tools follow so there's always been there's obviously we will draw comparisons out between the game and the tool i appreciate there is a difference a big difference between them which we'll get into but there was always talk back in the day back in the early days of football manager that real clubs real managers had used fm to scout players it was apparently responsible for bobby famino internet but hoffenheim camoni using it to lose rooney at dc united now as a fan i'm not going to say that i'm not going to say that i'm not going to say that i'm not going to say that i'm not going to say that i'm not going to say that i was kind of viewed those as like really how much of that is kind of like media spin or just like a good story but at what sport at what point sorry did sports interactive realize that you had something that could be an actual tool which clubs would would pay for subscribe for well i didn't know that bobby famino won but i'm a liverpool fan and he's my favorite ever liverpool player so i'm very happy about that one before my before my fm time certainly the work at dc united that's whilst i was here i was a fan of the club i was a fan of the club i was a fan of the club we were working with them on recruitment but a lot of the stories that you've seen in the press are real we've worked with a number of clubs and different organizations during my tenancy here at si but also i inherited a number of clients before we started here so there was always the opportunity for us to to create a it is a different tool but to create a professional tool based on the market needs but also the experience that we had within the team we have people like my self and matt have worked at huddle and white scout previously to to introduce a new data set to the market the the market is quite saturated right now and what we wanted to introduce is something completely brand new and to have 1600 evaluators is probably you know nothing's impossible i don't think but very likely to be impossible for a football club or a federation to do due to cost and we already had that team and and that base layer of information of which the game is one expression the fmdb uses that data but also includes that third-party data sources and our own insights to make it to make it perfect for for modern day football recruitment so how did that actual decision come about was there somebody one of the most interactive kind of turn around and go well if we did xyz this could actually be a revenue stream for us or yeah full transparency when i joined it was the reason that i wanted to join fm i played the game and when i was when i was younger i still play it time to time i don't have so much time to play it as much as i would like but we used to talk about it at white scout and we used to talk about it at huddle and how interesting the data set was and a number of the clubs that were working with there were using using fm data i took the job and then it was probably about six months in where the question was actually sort of posed in earnest you know what should we do about this and i was very receptive to it because it was kind of a question that i wanted to be posed and something a project i wanted to run so about three years ago we we started with a very small project like how do we do this you know there was a bit of an rfp looking for engineers and developers and team members and ultimately i ended up working with a good friend of mine who who built the white scout platform and so it was somebody i knew i trusted and and then we started the the initial scoping so it was a bit of both it was there was already a need in there i kind of hoped work would want me to do this at some point and and then i was asked maybe through a little bit of probing and then we started in earnest you mentioned dc united and some other clubs that you've worked with before which clubs are now using the fmdb pro as a tool and also you should mention when did this lodge this is quite recent is it we're talking about about it was last month or this month two weeks ago today or yesterday so super super quite super recently we we have a partnership with where all of their clubs have access to it and we've had a it's not every single one that would be wrong but it's it's it's above 80 you know and obviously when we go through location and promotion certain other teams become eligible for it so that was one of the first things we did which is a good partnership to start with the second thing we have is we have a partnership with fifa so within fifa there's 211 member associations and they are all entitled to a limited part of the data set so for example if i am the greece national team i could have access to all the greek players within the fmdb and then they can they can extend their access should they so wish then outside of that we go from club america in in mexico i'm heading north here because this is the way my mind works up to up to angel city and the nwsl up to the vancouver whitecaps and then as we we head towards head towards europe two of the premier league clubs i'm going to say which ones but they're very known for their data usage of always being partners of ours or not always probably you know five six years then and as we head south monaco are a partner of ours we also have dynamo zagreb and obviously you know they've got a great reputation for finding finding talent and then we have a number of businesses as well so from different industries other data companies who use a baseline of our data to to supplement their data sets and then companies in other industries that are specifically asked to remain private so in total we are probably just over 40 40 clients at the minute so 40 customers you know 40 partners there's quite a skate it's quite sizable ones that wouldn't include the fifa partnership and the premier league partnership that's already well i i'm not including the 211 because it balloons the number a little bit so we're probably closer to 250 but the problem with uh well the status of that situation is we're just launching that so why don't we just launch that and we're just launching that and we're just launching that because they are whilst they're entitled to access we haven't approached all of them yet so they're coming to us but we have a number of them already on board but i would say 40 is of 40 is a good number at the minute and since the launch yeah it's it's been pretty manic over here i think matt who works in in our in our team probably something like 25 meetings this week with various football clubs and his diary's packed all the way through till august it's launched so yeah we're very positive about the future well let's look at that then i know obviously you asked revenue and and prices or anything on those lines on this is there like an anticipation that in terms of the bottom line this could have as much of an impact as fm the game or is fm the game always going to be the key revenue driver i mean certainly now fm's the key revenue driver you know it's a significant and historic game that plays a large part within society and football culture really we're a brand new business and we're a sports data sas business within a video game company which i think we're probably the only one of them in the world right it's quite a it's quite a unique situation we haven't set any forecasts on the business we haven't set any goals for the business the the goal was to create an elite grade product that could support football clubs and federations now you know if over the next two or three four or five years it becomes a significant revenue stream for the business that's great because we can put that that revenue into more projects but i think it's really important for us to be able to create an elite grade product i think it's really important to state that the team that's built this product apart from me because i mean i i work in the commercial department so i work on the game as well they were hired specifically for this project so you know matt who i've referenced or dave or the pat or the people that work on the project tom's a little bit different because he supports us from a knowledge perspective they were hired to do this and their skill set is this and they are not people who probably could learn to work on the game but that's not what they've done in their careers people hired to do this and their skill set is this and they are not people who probably could to do this was specifically for the fmdb pro but we've got high hopes so there's going to be people who are listening to this who've probably seen the podcast kind of pop up on their feed because they're interested in football manager the video game who may not kind of be aware of the fmd pro products how these are very i know these are different products and kind of how they're treated internally but is could a fm player have a look at the fmdb pro product and work their way and kind of benefit? Is there that kind of synergy or is it basically on a data set at some side? I think if you gave the FMDB Pro to an FM player, it would obviously benefit them because you wouldn't have to discover the players. They're all in there straight away, right? You can see all their data. You wouldn't have to go scout it because it's all there. A key difference between the two is that the game obviously comes out usually around November and then there are X amount of data updates throughout the game cycle. The DB Pro is based on Tom's team and Wood, who is in his team. They're 1,600 researchers. As soon as somebody updates the data within our mainframe database, which is called the ORDB, following Monday, it will be updated in the FMDB Pro, whereas the game is an entertainment product. So if everybody on this. This call started a save, you know, playing as Liverpool, for example, within 10 minutes, we're all in completely different worlds. So, you know, it's your journey and that's always been the beauty of FM. Whereas the Pro was built, you know, for visualizing our database in a platform, which we spent a lot of time and consulted a lot. We wanted to build an API. So, I mean, a lot of smaller clubs use APIs now, but the bigger clubs have been reliant on them. You know, 10 years now. So we wanted to build an API so we could fit their recruitment needs. And then we built recruitment specific tools as well, insights to support long listing before clubs move into video analysis or modeling or like scouting. So whilst the data set creates the game and it creates part of the FMDB, it's not the entire FMDB. You know, the things that we've layered on top, the tools that we've layered on top and the additional data sources that we've integrated and the consultation with clubs about what's needed in the market right now. You know, there's no point as replicating one of the existing providers, because as I said, it's very congested. It's very saturated. We wanted to create something that's essentially unique and to supplement their existing workflows. So you mentioned additional data there. Like the FMDB data. Yeah. The one that fans would use within the game. It's known for being extremely deep and breaking down players in more ways than people might think. What else are you doing with the FMDB pro database that clubs have access to? So we have a partnership with Huddle where I used to work. A fantastic company that, you know, at the minute dominates elite sport from a performance analysis and recruitment perspective. They've made a number of acquisitions over the last 20 years across multiple areas. We have a partnership with them for their Y-Scout product, which weirdly as a product I worked on before I went to Huddle. So their event and physical data is going to be ingested in certain capacities into our product. It's important to note that we're not going to, you know, abandon the attributes or the scorings that our evaluators do, but we use them to benchmark and we use them to evaluate different countries and league levels. And then there's certain other assets. You know, like a player's top speed, you know, it's, it's not necessarily a subjective thing, you know, it's, it's a pace, it's a number of kilometers, you know, someone's top, top speeds, probably around 40 kilometers an hour, rather than a scoring of 20 out of 20, you know, so, or 15 out of 20. So their data is coming in, looking forward. We, we recognize that there's, there's other areas in, in the global market that there aren't. Um, let's say necessarily a preferred provider for, we think we can really help football recruitment from an injury perspective. And we've got tools that we're working on right now, research and methods that are going to massively expand our, our, our injuries that we, that we will, will visualize within the platform or from obviously compliant means and looking at world football and all available sources to try and accurately forecast player salaries and transfer fees. And that's a big areas where, you know, I think the market could use a lot of our help. So it really is a hybrid of the original FM database. We'll be using our existing numbers, whether they're on injuries or whether they're on financial data and blending them with the other data sources that we have access to, to, to try and try and create a unique set that is beneficial. So I've seen plenty of places I have pre covered in interviews before the whole process. Around the FM database and the scouting, and you've mentioned 1600 scouts already in this conversation. Has that been extended for FMDB pro? Are you still using the exact same scouting networks apart from obviously kind of what you're bringing in through the partnership? Why scout? Yeah. I mean, it's the same exact same scouting network. We choose to call them evaluators. I think that while we've got scouting personas, people who've worked in, in football analytics and scouting roles for a long time, I think that, you know, we acknowledge that we're not a replacement for, you know, experienced football scouts. So we like to use the word evaluators or researchers when we talk about them, but yeah, I mean, it's the set that same network of about 120 that we'll look after an individual nation and then about 1400 1500 below them that are sort of based around one club or a few clubs at a time. Obviously the success of FMDB pro we hope in the future will allow us to sort of scale up and add more coverage. We're always looking to sort of get further down the levels, increase the amount of data that we collect and the depth of the data. The data at the top levels too, but yes, it power is powered by the same network. So how'd you get a part of that network? Like it, who, who makes up these, this, I'm not going to call them scouts. We said not scouts, this evaluator network can, could I be one by going down to watch my local team, whether they're premier league or third division or Eastman Lee, whatever else it might be and make notes or do you train them up? How, how do you kind of create an evaluator? In some part? Yeah. I mean, the network is, is, is a wide range. Of, of people, as I say, there's a lot of our, our group that have have experience in football and will use us as a kind of, I guess, a, some kind of a stepping stone, a CV addition, or a, an ability really to showcase that they're able to look at strengths and weaknesses of players and go on to work in football. But yeah, I mean, generally speaking, everyone goes through, through training calibration. We've got a set of definitions and guidelines that you'll see in FMDB pro. We try and give that information to clients so they understand, well, what is a nine for finishing and what, what are we looking at there? But there's always difficulty with subjective scoring. We know that, and we're always looking to improve, but yeah, it's to, to put it briefly, it is a wide range of, of people and we've got to make sure that, you know, the, the data in China is the same data in France is the same data in England. And that's a challenge for sure. And I think, sorry, I'll just add to it. I think we have an in-house data team as well. And what the responsibility, one of the responsibilities of that team is to look at the, the real world events and physical data and to, to model it and benchmark it against how we score. So, you know, that's a relatively new addition over the last few years where we've started to do that with the, with the prevalence of those data sources becoming available. And, you know, every day we look at how to make it more accurate and the strengths and weaknesses of various leagues and the part of the FMDB's evolution over the next three to four to five years will be to become more and more and more precise. But we don't want to lose that subjectiveness as well, you know, because otherwise we are essentially replicating the event database in, in one to 20 or whatever you want to call it. So we are that, that hybrid, we are essentially that power of people, you know, going back to what I said at the very start of the chat, it's impossible, virtually impossible to, to have an army, let's call them, you know, for want of a better word of people all over the world in men's and women's football, collecting data for you. And we don't just collect data on. On players, we collect it on bathroom staff, we collect it on stadiums, we collect it on weather, tax regulations, all these different things that power the game, not all of them are relevant for recruitment. So we strip the things away that weren't, but they evaluate the whole club, you know, men's team, the women's team, the academy, the unders, the finances, et cetera. And that all forms part of this hybrid model that we've formed. There would have been a time not too long ago, actually, where women's football wasn't, wasn't part of FM, wasn't, wasn't even part of a lot of fans kind of consciousness. But it is now. And obviously it's part of the database you've mentioned as well. Does, is there more challenges to evaluate in the women's game than, than the men's game at different levels? I think if I take this one, I think that definitely to start from scratch was probably the biggest challenge. I think that, you know, there are clearly when we, when we look at finding data from publicly available sources, it can be really difficult to find stuff on women's football. There clearly isn't the coverage that exists, whether that's in video in terms of providers that we can watch players on, whether it's finding information, even on transfer. And squad lists is difficult, but certainly starting from scratch, as we did five or six years ago, when Tina Keats joined the company, our head of women's football research, I think that, that, that initial phase of adding lots and lots of players, thousands of players and clubs, I think we're up to about 40,000 active players in there is a huge effort. And also actually developing an evaluation network from scratch, you know, to have the number of researchers that we have on the men's side replicated on the women's side is a challenge. I think that, you know, we look at bringing in people with lots of women's football expertise, but also that ability to translate it into the football manager scale, I guess. So I think we've called upon a lot of experience from the men's side as well to help with that in terms of the unique sort of challenges compared to men's football. I think we were very adamant that we were going to run off the same one to 20 scale, I think. it's fair that we have two separate sort of databases there. And we say that a 200 CA player, the best in the world on the men's side is separate to a 200 CA player on the women's side. And each of those individual one to 20 attributes the same. And I think there's obviously differences in the style of the game. Certainly game side, we tried to replicate that in the way that you see matches played out. But from a recruitment tool, I think you're always looking for very similar attributes in players regardless of the game side. Obviously the idea of FMDB Pro is that you look for the things that you're interested in, whether you want players with high pace or you're looking for particularly strong players or players with high anticipation or composure, you can do that yourself. So I think there's some of the challenges, but I think by and large, we try and make sure that the system is, I guess, equally the same on both sides. So I read in the PDF, I kind of outlined all the details for this, that there are 250 data points for the players. As an FM player, I'm used to a lot of them. I'm not used to 250 though. What things have kind of come into FMDB Pro that might be surprising? I think we've probably crossed over maybe a few of them already in this conversation. I think we have a different GBE scoring, right, is one thing that's one of the key development pieces over the last probably six to eight months we've been working on it. We noticed in the market that there was a lot of different providers, not naming names, but it looks like some better than others. And we consulted with a lot of parties and due to our breadth and depth of coverage, you know, across, you know, appearances now through our relationship with White Scout and the injuries that we already had, et cetera, and the documentation from the FA, we think we've created a really good GBE calculator. So I would say that that's one. Also, though, we can do things like injury insights, injury availability, right, which is obviously different in the game because the game, as I said, goes off in its own direction, you know, within five minutes, 10 minutes of playing. So you wouldn't be able, you wouldn't be able to use the game. You wouldn't actually be able to use the FMDB for the game because the game's on its own path. So I would say injuries is a very, very specific one. We have embraced AI as well. So there's two areas in the platform where you can create individual reports using all of our data and the White Scout data to get an AI summary of a player in a very concise format at scale, which is unique and doesn't sit within the game in that kind of format. And we can also use the AI technology we've developed to go and actually look for players on a sort of, you know, pre-tech search. You know, I am looking for player X between these dates of birth, who has these attributes or these features, who's going to cost this amount of money, whose wages are this amount of money, who fits our playing style, fits within our current level that we're playing. And it can bring that 500,000 players down to a meaningful number, you know, 15, 20, very, very, very quickly. So we've embraced that perspective of it. It's not the most major part of the platform, but it's something that we've built also very recently that's a key feature. And yeah, I think that's, they're the key things for now. Obviously, sorry, the contractual stuff, but with the new tools coming in the next couple of months, our accuracy on, transfer fees and salaries, whichever way you want to call them, we'll be up there with the best. And hopefully we'll save people a lot of time and work in the background. Another thing which would never be game related is we did a lot of ID mapping. So this probably won't be known to many people listening to this podcast, but every individual player within YScout or Opta, or we don't use it for transfer marks, we'll have an individual ID. And one of the problems at a football club is combining those IDs. And compiling the data on an individual player so you can make decisions. So that's been a big project that we've taken on. We've done a number of sources already. And so we can help clubs with that as well because it's a pretty laborious process. So there's a lot of things in the real world game that people wouldn't know about. Is there a chance, I know nothing's been announced for the new FM for later this year, but is there an opportunity, do you think, or possibility that there could be elements from FMDB Pro that will maybe, end up in the FM that we're all used to? Well, we did on the game side, we did a partnership with a great company called Transfer Room two games ago where we built part of their UI and it's the agent to agent, the pitching of players marketplace in the real world. And it was really successful. It's one of the features that our players really, really liked. And then we did a relationship with Hawkeye where they provide us data for the match engine that you might see. So we're integrating more professional tools, we're integrating more technical tools into the game. It's part of our agreement with Huddle. We want to do something with them. We haven't defined exactly what it is yet. It might be maybe like the iPads on the bench, their product called Huddle Replay. We might do something around that rather than the, you know, the fast forward replay or the rewind. We haven't completely decided. But as the FMDB becomes more prevalent in the real world, of course, right? And of course, as we're now generating more data than the researchers alone, and the product becomes more well-known in the world, and more clubs are using it, it will then become a natural part for us to include that data in the game. I don't think it will be for this iteration because we've only just launched, but it's something that's talked about a lot in the background. You know, obviously with these things, there's a lot of checks we need to do and development work and the design team, et cetera. So nothing happens super quick, but there's a good chance by, you know, maybe FM28 that parts of the data in the DB or parts of the information in the DB will be included in the game. Because I've seen a lot of the commentary since we released the product going, the WISCAP partnership will help the game. We hope so. And we're allowed to do that with them. It's just the development time of when we get the information and then putting it in actually to FM. But the door's now open. - We're recording this on Friday the 17th. We know the final is for the World Cup. This will go out after the World Cup. So we'll know who's finally lifted the trophy. - There's always conversation when the World Cup comes around around players who could have played for other teams and this kind of thing. And one of the things I picked up on going through the big PDF was the way that the database in pick up second and third nationality options for international teams. Is that something which you expect to be used? Could there be a time where maybe in about four years time or something like that, we don't lose out on Erling Haaland to Norway as an England team or more, or we end up with other clubs who may be kind of smaller, or other nations be able to kind of benefit by finding out who maybe had a curacao, curacao, sorry, grandfather or something like this. - I think I could probably tell a couple of stories and then Tom might want to tell the Diaz story, you know, which is the historical one. I mean, he'll do a better job of it than me. Even a couple of years ago, when the product was in the build phase, national teams were approaching us and saying, what we really struggle with is finding out who's got a second and third nationality. So with a number of federations, we started to trawl our database and we said, how many players do you think are eligible for you? And in one of those situations, which was a relatively small nation, compared to the teams obviously in the World Cup final, we found them three and a half thousand extra players who could potentially be eligible. And that got my mind sort of ticking over. So as the discussions with FIFA progressed and the discussions we've already had with national teams, it's one of the key features that they're interested in. And that's why we've kind of highlighted it in the platform as a key part, because it might also lead into languages, and it might be spoken by the players, but also really from a national team perspective, so you don't miss out on that talent. And if you think about the smaller nations, they're probably likely, with no disrespect to them, to know less because they've just got a smaller team. If there's a big, big nation, you know, Spain, let's say arguably, within the World Cup final, they probably know, in my opinion, most of them. But just due to bandwidth, I think there's a huge sort of untapped market for finding players that you could recruit early and are playing a national team. But Tom might like to tell the other story, because it's better out there. - Yeah. Yeah, there's certainly some, a couple of stories that, you know, we've heard may have been sort of a football manager and there was certainly an FMDB background, the Ben Brereton Diaz being called up by Chile. There's kind of a story there that, you know, potentially was one of those that was flagged by one of our evaluators from a Blackburn Rovers match day programme that his mum was from Chile. And then, you know, conversations happened. I think also the Young Husband brothers who ended up playing for the Philippines for quite a while. I think they started at Chelsea was another one. But yeah, there's absolutely value in that second and third nationality data. It's, I guess, one of those things that, you know, you can pick up in those publicly available sources, you know, whether it's in an interview, I think spoken languages as well is a massive one, even for clubs in terms of, you know, can they speak English? Are they going to be able to integrate quickly? But yeah, certainly I think that we've got a really good coverage of, you know, the Curacao example, and Suriname as well, those players with many of which born in the Netherlands and then go on to play for different national teams, I think could be a massive, massive plus for those FIFA nations for sure. - And that ease, you know, he goes straight onto the platform and you say, which players have second or third nationality that's Norwegian, for example. Within two seconds, you've drawn all those players out. And, you know, it's one of the values of the database, whether it's from a language perspective or a nationality perspective, what would have taken weeks and months is a matter of seconds. now for us it could be an interesting 2030 world cup seeing who turns out for who it certainly could be so i find it interesting the world cup you have brothers you know who playing for different teams and i think it's going to be a more common thing and yeah let's see let's see and maybe we're part of it right let's hope so guys it's been really interesting talking to you let's let's see how things work out the next few years see what see what the effect on the real world is thanks for coming thank you very much having us on cheers the football co business podcast the most creative minds and innovative thinkers in football