It’s back: the Betting 360 podcast returns for 2017 with Mark Haywood as your new host.
This week he chats to an experienced sports modeller who heads up a ‘diversified sports analytics organisation’ focusing on NBA and AFL.
- How their ‘feature engineering’ helps create an edge
- Why they operate differently to other betting syndicates
- How focusing ‘off Broadway’ can offer great value
Get More Betting 360
Make sure you don’t miss our punting tips to come! Subscribe on:
What Do You Think of the Show?
Mark: My guest on today’s Betting 360 Podcast is Brett. He’s a professional sports modeller and trader, and he’s the man behind the new Champion Bets NBA package.
Before we discuss why you put together the Champion Bets package, we might just delve a little bit into your own background and your own history. When did you first get into the punt?
Brett: I’d say that I probably seriously got into sports betting in around 2009. Prior to that, myself and my current business partner had a pretty deep background in poker. Straight out of high school, two young blokes with some large imaginations and some half decent programming and just general skills and logic. We thought we would get together and do something with poker. So we started building poker bots, back in 2003. We had a five to six year history with poker botting, and playing poker. That generally transformed into a sports betting capability by about 2009. We grew out of our poker roots. We started to learn a lot about machine learning and all of these processes coming out of data science. That was new and exciting, so we basically moved away from poker and got into the much more scalable sports betting market.
Mark: Was it always a more serious thing for you? A full-time pursuit, if you like, rather than a hobby?
Brett: Yeah. I think I’ve rarely recreationally had a bet in my life. Around 2009, I surveyed the investment landscape and made an assessment that sports betting offered the most bang-for-buck for return of capital. Because if you can find an advantage in the sports betting market, your so-called return bearing period is the duration of one single game, whereas in conventional financial markets that can be weeks, months, even years for long term investments. So I really liked the idea of engaging with sports betting markets, given the frequent cycling of capital with that asset class.
I also liked the idea of the benefit of sports betting markets having recreational money. I think it’s the only market in the world where recreational money fuels some significant portion of the total amount of capital circulating through the market, whereas compared to any other conventional financial market, that’s just not happening.
Take the example of currency FX pairs. No one’s punting the FX AUD/USD pair for pure entertainment. It’s being done to make money, whereas some bloke might be sitting down on a Friday night watching the footy, and his money is contributing towards the market inefficiencies that we see in sports betting markets. I saw that straight away back in 2009. That’s what attracted me to sports betting markets, and the rest is history I suppose.
Mark: Sure, and as we said with Champion Bets you’ll be doing our NBA package. Has it always been predominantly basketball for you, or did you start with other sports, and have you moved around other sports a little bit?
Brett: I’d say that my first ever model was the AFL. We still have a pretty deep interest in AFL football. We expanded to other Australian sports, and then when we met our U.S. based partner back in 2010, we were introduced to North American sports.
We basically moved much of our focus away from Australian sports. We came to the realisation that there was some pretty deep, pretty liquid markets in North America, so we made the conscious effort to sink in much more time into the body of work that our business partner had worked on for a number of years. He was also a sports modeller. We basically built on his body of knowledge, his knowledge of algorithmic processes. We basically entwined what we were doing with what he was doing, and we sort of melded together and grew on that. What came out of it was a focus on Australian and North American sports at this point. So NBA is just one of the stable of sports that we currently do.
Mark: As you mentioned there, you’ve got business partners and you’ve also got a team. How would you best describe your current approach, and the group that you look after at the moment, and how you guys go about it? Obviously without giving too much away.
Brett: I’d say that we have a very, very data heavy approach. Putting my feelers out there in the sports betting world, I’ve actually never met anyone that goes to the extent that we have gone to. We use tens of thousands of features in our sports modelling, which is pretty much unheard of. Even in data mining circles, in machine learning circles. I know of some pretty smooth operators in sports betting, and none of them have gone to that extent with their data orientated approaches, whereas we have.
I’d say at the heart of our sports prediction is that sports prediction technology that we’ve been building, and that’s surrounding the feature engineering phase of our sports modelling. That feature engineering phase is where I think the heart of value resides in sports prediction. We have what we coin as the computational engine, which generates all these features that feed into downstream lending processes, and that is what I consider to be the heart of our value advantage.
So yeah, in summary I’d say it’s a very, very data heavy approach, unlike anyone I’ve ever come across. And I’ve come across a fair few syndicates now, in the last few years especially. We do things very differently to everyone else. That isn’t saying we have a pure machine driven approach. Everything that we generate is eyeballed and signed off by some human trader. It could be myself, with main markets. Or with this player prop staff, and other minor markets, with one of the blokes that work for us basically, look at our internal trading console and eye off any bets that we either make ourselves or that we release.
That’s our betting approach I suppose, and what we are. What we are as an entity, I would say that we are a diversified sports analytics organisation. I don’t really consider myself as a sports betting syndicate any more, whereas that may have been the case 18 months ago, two years ago. I think there’s a multitude of profit centres that can arise out of sports analytics. Punting is just one of them, and my goal at this time is to exploit advantages in sports analytics far beyond punting. There are some entities overseas that I sort of look up to, in terms of what that business model looks like, and I will endeavour to take our organisation there sooner rather than later.
That’s basically what we’re doing at this moment in time with our punting, and what we’re doing outside of punting.
Mark: Just a quick one, a little bit off topic. Obviously, because a lot of our listeners are very focused on the racing game. Is that something your team’s ever looked at, or has it been specifically sports betting?
Brett: Good question. Currently only focusing on sports betting. I’ve had a little bit of exposure to Hong Kong racing in my time. I’m really attracted by the volume in those markets and the kind of margins that can be had from those markets. So that’s something that I may look at or we may look at going forward in time, but I’d say we’re just taking things one step at a time. We’re going to consolidate on the sports that we’ve been working on for a number of years, and then once that has been perfected then we can move along and explore different opportunities.
Mark: As I said earlier, just moving on to the NBA. You’re actually coming on board with Champion Bets to produce a new player betting package. But you actually, I think it was around three months ago now, came on board for a small group of our higher value clients. It’s definitely unlike anything we’ve ever seen in terms of both how it works, and the returns. I think at the moment the initial results are around 400 units profit in pre-match, which is obviously absolutely unheard of. Did you want to give us a bit more background on that package that you’ve been running?
Brett: We started that package around mid-November. It was my intent that we’d keep it as private as possible, given that we were seeing these kind of margins prior to launch. We were like, “This is out of control. We can’t give this out to the betting public because we’d probably pull the whole thing down with hundreds of clients taking on the books.”
So we basically had a very small, private clientele of Champion’s that we started to release to mid-November. We’re almost 400 points of profit in three and a half months on 6,700 bets at 8.3% ROI, which is basically unheard of. I’ve never seen anything like that. I’ve never seen anywhere near those results in three and a half months in any sport, in any package ever. That’s what we’ve been privately working on with Champion Bets since mid-November. I think it’s time to go out into the open somewhat, and release a more liquid portion of our offering to a more wider circulation of clients.
Mark: As we do that, we’re launching the new NBA package in the coming days of course. Do you just want to take listeners through exactly what sort of bets they’ll be looking at with this package?
Brett: This All-Stars package will entail two subsets, or two different kinds of bets every day. There’ll be around half a dozen NBA player prop bets per day and that will be on the six most liquid players per day. So your Steph Currys, your Kevin Durants, your superstars of the NBA will be the players that we will be focusing on, because they have a wide circulation among all the Australian bookmakers, and Australian bookmakers can take a half-decent dollar on these well-known players.
That’s the first part of our offering. Then the second part of our offering is NBA first half plays. Going in to first half markets is part of a strategy to exploit our advantages in markets off-Broadway, so to speak. We made the conscious choice about a year ago to pivot somewhat, because we found that even after six or seven years of working on technology and such, our results in full game markets within NFL, NBA are just too inconsistent to trade reliably.
So what we then started to do, is that we focused on sub-game markets. We’ve seen that sub-game markets were a much more exploitable than full game markets. Which makes perfect sense because, I’m not sure if people that are hearing this podcast really understand the nature of global sports betting markets but really, everything is shaped by global market forces. Any number, any price or line that you see in this country on any non-Australian sport is shaped by the global market. There are maybe no more than 20 large syndicates that compose 80% of the turnover offshore. Those large syndicates are basically, those entities that have shaped this global price.
Given that’s the case, when you’re a small syndicate aspiring to grow in size from trading, you have to realise that you’re coming up against some of the best on the planet. Those entities, I’ve seen them. On their trading floor, they have seven figure R&D budgets. They are turning over monumental amounts of money, and if any small outfit thinks they can compete with the best in the business, you need to really be sharp and really be on your toes when it comes to your modelling and your overall process, because that’s who you’re coming up against.
With that realisation in mind, we then decided to pivot and focus on sub-game markets and prop markets like game props and player props. That choice to adjust has paid off spectacularly well so far, as we can see from this product offering that we gave out to our Champion clients three and a half months ago. From our own trading, we have seen that going through the undergrowth and trying not to draw attention to ourselves by trying to pretend that we’re the lion of the jungle when we’re not, when we’re actually the mouse trying to grow in this jungle environment.
That was our strategy. Start again, go back to the bottom of the food chain. Stop smashing our heads against the wall by trying to beat the heart of the market. That’s paying off really well.
That was the strategy and the mindset behind looking at first half markets. One year later, we’ve got a fairly successful first half solution that exploits the fact that these large syndicates are not focused on the first half, on the pre-game first half, that is. Simply because there’s not enough money for them to bother with pre-game first half markets. It also plays into our hands, because the first halves are, for most sports that we have covered, fairly algorithmic in nature. So all these concerns with form, fitness and class and team motivations are basically things that happen in the later stages of the game. We found if we combined first halves only, which were fairly algorithmic, with less focus by large syndicates on these obscure first half markets, we could come to a fairly decent advantage. That’s what we have to offer, along with our half a dozen player props per day.
Mark: Fantastic. We’ve actually never had an NBA package on Champion Bets until now. It’s something we get a lot of requests for, it’s obviously a very popular sport to be betting on, just with the time of the games and the number of games. The subscribers to this package, as you were saying, can expect bets every day, or most days?
Brett: There will be bets every day. We’ll have around half a dozen player prop bets per day, and there’ll be somewhere between 10 to 15 first half bets each week.
Mark: Fantastic. That’s about all we’ve got time for today. Thanks for joining us Brett. I think you’ve given everyone a great insight into how the new package will look, and what they’re dealing with in terms of your own background, so thanks for that. We look forward to a very long and hopefully fruitful relationship.
Brett: Cheers Mark. I’m really looking forward to our new public package doing as well as our private package has so far.