Robert Mastrodomenico is a fellow of the Royal Statistical Society as well as owner and founder of his statistical consulting company Global Sports Statistics.. He is also the Chair of RSS’ Statisticians for Society initiative since its inception in 2017. He is also an RSS Statistical Ambassador, which involves regular work with the media in assisting with their reporting of statistical issues.
Episode Description
What are the odds of your favorite team's victory, how much should they spend on a big name player, we discuss this other topics this on today's episode of Stats+Stories with guest Robert Mastrodomenico
+Full Transcript
Bailer: How much should a club spend for a player. We discussed this on today's episode of stats and short stories where we explore the statistics behind the stories and the stories behind the statistics, i'm john Baylor stats and stories is a production of Miami University's Department of Statistics and media journalism and film, as well as the American Statistical Association. Joining me is regular panelist Richard Campbell, former chair of media journalism and film, Rosemary Pennington is a way. Our guest today is Rob Mastrodomenico, Mastrodomenico was a sports statistician and a Fellow of the Royal statistical society, Rob, thank you so much for being here.
Mastrodomenico: Thanks for having me.
Bailer: It's great to great to chat with you. Can you talk about how did you get involved in sports statistics and you know maybe what are some of the things you've worked on at least give us an overview.
Mastrodomenico: Yeah, I mean, like most things in my life. It was entirely by luck. I teach a degree in math stats. First Class degree four I've got all the answers I'll go get a great job, didn't know what I was going to do so I was like, do a Master's applied, and they're like, Oh, we got funding for a PhD, did you want to do one, I never thought doing it so I did a PhD in statistics applied to statistical genetics, and then out of that I was like Oh pretty work in finance seems like a good bet, and then one randomly one day an email pops up from a random stats mailing list said, Do you like football. Yes. Are you good at statistics, well I thought I was, um, so I applied for a job and ended up working for a consultancy company in London who were doing predictive modeling of sporting events for clients you use them to speculate on, and it was a world, especially back then, completely unknown to me, I If I'd known this job existed, I would have, you know, at least prepped a bit more than some more research, it was very unknown. And I suppose I got into it because I thought I had a decent knowledge of football and my communication skills were pretty good and I could communicate the concepts and kind of link back the concepts of, you know, sports and data and bring it together. And that's how I got into this crazy world and kind of never really left, you know, been doing it ever since and it's very interesting you know, like you say, lots of applications is it predicting who's going to win the title is it coming up with bookmaker odds for bookmakers is it trying to find out who's of what player, what League, should you should be looking at because, statistically, a goodbye is very varied and you know it's it's one of those you know obviously with statistics statistics, you can apply your all the techniques but it's having an application that you're passionate about that makes it much easier. It doesn't really seem like a job. When you're, you build a model to try and predict, you know, how many volts or in a game or. Who's going to win the Superbowl things like that, you know, those are that's fun is that you know I get paid to do that.And that's really cool.
Richard Campbell: Hey, Rob I'm interested in the analytics analytics part of this if you follow American baseball we just had a world series in which one of the teams Tampa Bay was criticized because they took a pitcher out who was doing very well based on the analytics that the next batter and the pitchers they would bring in would be able to shut down the Los Angeles Dodgers. Well that didn't happen. So the analytics came in for a big hit. Can you talk about maybe compare the way the US is using analytics in sports, and it's our professional sports team versus UK, the UK.
Mastrodomenico: Well, I mean, I think you guys are very far ahead in terms of how you use it, you know, when people say is they'll talk about Moneyball, and I'm like that's all very good so far away. I think the kind of franchises in America just a bit more professional than a lot of kind of clubs which historically in, if you think in terms of football which is our kind of soccer, which is our major sport here you know it historically, it was kind of owners bought clubs, as a kind of a, you know, some of the show, some, you know, some might be a local businessman, how does a club and it was you know the the it was just something he did when he was successful and that's with the money in the Premier League. That's more than now we've kind of this is big business and I don't think internally clubs have caught up, and that's not a. That's not a generalization a lot of clubs, you know, invest in analytics and invest in that you know what they need to do. But the kind of analytics community is there, and people are coming into close but I just don't think it's as sophisticated. And I think, you know, people with the iPhone advances and things like up to date are expected goals models. You know you can very easily get into it's a great thing to get into, but that kind of higher level research, you know, really good statistics, don't think that's infiltrated it yeah and I think there are a few examples of clubs here who were doing it, and a really successful, but that mindset of how you really need to use the data, and that kind of the level you need within the franchisee, we can use to work. I don't think it's caught on yet But my hope is that it will, we're seeing the term people, people are much more data savvy now than they were a couple years ago.
Bailer: So I just as I as I'm a huge soccer fan huge football fan so one aspect of this that I'm intrigued about is, in your experience in sports statistics. What's, what's uh can you give, I'm gonna ask you for two examples. One of the first example is what you would view as a real success story that you've that you've had in terms of some predictive model or some analysis you've done in sports statistics, and the second example that I'm going to ask you, is what's, where's the case where it's really not not worked. You know what's a really hard outcome to try to measure in sports statistics.
Mastrodomenico: This is good question. I do talk where I kind of. I've done it loads and it's like, I explained how you build a very simple kind of predictive model using goals so using the kind of Dixon calls person approach, and in the example I use some very old bodies of Liverpool when before they won the title when they had players like Luis Suarez, and it was a big example of where a model predicted something that the betting markets didn't and it makes it look like this model is brilliant it but I mean, in this case, it was right, because it got it right. But actually, there's loads of examples where it gets wrong in terms of if you're going to predict against the market you know your edges these days are so small because betting markets are really sophisticated So, you know, you have to be used to the fact that you know you're what you're going to get out of it is no small margins, but it's so small margins that make the big differences and generally you can get pretty close to what's going to happen unless there's nothing that really strikes out as things that have got you get massively right or wrong. It's generally within those kind of realms of where you need to be you might have an example like the Liverpool ball where everyone thought Liverpool would win the tie or a man seated, where the model looks really clever but there'll be other examples like when less than one the lead, and you know, there's not gonna be many models at the start of the season we're saying less that we're going to win the league back then and, you know, what's going to happen this season, you know, anyone's guess.
Bailer: Well, Rob. Thank you so much. That's all the time we have for this episode of stats and short stories Rob we really appreciate you being here. Thank you. stats and stories is a partnership between Miami University's departments of media journalism and film and statistics and the American Statistical Association. You can follow us on twitter Apple podcasts or other places where you find podcasts. If you'd like to share your thoughts on our program, send an email to stats and stories at Miami. edu, or check us out at Stetson stories dotnet, and be sure to listen for future editions of stats and stories where we discuss the statistics behind the stories, and the stories behind the statistics